阅读说明：本技术 一种相关双采样线阵传感器的自动报靶及弹道测量系统 (Automatic target-reporting and trajectory measuring system of related double-sampling linear array sensor ) 是由 张劲 于 2020-07-27 设计创作，主要内容包括：本发明提供一种相关双采样线阵传感器的自动报靶及弹道测量系统,对于低速弹丸使用至少2个高速线阵传感器检测弹着点位置,对于高速弹丸使用移相的至少3个高速线阵传感器来检测弹着点位置,还使用更多个传感器减小移相时间来检测弹着点位置及过靶时刻；本发明采用反射式光源,利用弹丸反射亮斑进行检测,不限制弹种,抗红外光干扰；本发明还采用标定灯条自调校,可消除不可逆形变误差。本方案采用相关双采样解决线阵传感器的零点及漂移,不需要个体调校,也无需数周期的初始化数据来消除零点及漂移；本发明求取亮斑坐标的自动阈值法还易于利用FPGA等硬件实现,减小了数据传输及处理压力,利于使用多个线阵传感器提高过靶时刻的测量精度。(The invention provides an automatic target-reporting and trajectory measuring system of a related double-sampling linear array sensor, which uses at least 2 high-speed linear array sensors to detect the position of a landing point for a low-speed projectile, uses at least 3 phase-shifted high-speed linear array sensors to detect the position of the landing point for a high-speed projectile, and uses more sensors to reduce the phase-shifting time to detect the position of the landing point and the target-passing time; the invention adopts a reflective light source, utilizes the shot to reflect bright spots for detection, does not limit the seeds, and resists infrared light interference; the calibration lamp strip is adopted for self-calibration, so that the irreversible deformation error can be eliminated. According to the scheme, zero point and drift of the linear array sensor are solved by adopting related double sampling, individual adjustment and calibration are not needed, and the zero point and drift are eliminated without initializing data in a plurality of periods; the automatic threshold value method for solving the bright spot coordinates is easy to realize by utilizing hardware such as an FPGA (field programmable gate array), reduces data transmission and processing pressure, and is favorable for improving the measurement precision at the moment of target passing by using a plurality of linear array sensors.)

1. An automatic target-reporting and trajectory measuring system of a related double-sampling linear array sensor is characterized by comprising: the system comprises a target surface, an infrared light source, three dark background side frames, a main control board, an infrared calibration light bar and N high-speed linear array sensors, wherein N is more than or equal to 2;

the infrared light source is arranged on one side edge of the target surface, the three dark background side frames are sequentially arranged on the other three side edges of the target surface, and the N high-speed linear array sensors are arranged on the side edge where the infrared light source is arranged; the main control board is respectively connected with the N high-speed linear array sensors in sequence; the main control board is connected with the infrared calibration light bar control board, the infrared calibration light bar control board is sequentially connected with the infrared calibration light bar, and N is larger than or equal to 2.

2. The automatic target reporting and ballistic measuring system of a correlated double sampling linear array sensor as claimed in claim 1, wherein: target paper is flatly laid on the target material of the target surface; the infrared light source emits infrared light parallel to the target surface, covers the whole target surface and irradiates to three opposite dark background side frames, and the infrared light source is not lightened at the initial electrifying moment, namely in the process of constructing the optical calibration table, and is continuously lightened after the optical calibration table is constructed.

3. The automatic target reporting and ballistic measuring system of a correlated double sampling linear array sensor as claimed in claim 1, wherein: the three dark background side frames have a certain width and are arranged around the other three side edges outside the edge where the target surface infrared light source is located, and the side frames are covered with light absorption materials, so that infrared light emitted by the infrared light source to the side frames can be effectively absorbed.

4. The automatic target reporting and ballistic measuring system of a correlated double sampling linear array sensor as claimed in claim 1, wherein: the main control board comprises a voice playing module, a display module and a network transmission module;

the main control board is used for sending out driving logics of the high-speed linear array sensors, the infrared light source and the infrared calibration light bar, receiving data collected by the high-speed linear array sensors, constructing each optical calibration table at the initial moment of power-on, inquiring the optical calibration tables in the subsequent target reporting working time, interpolating and calculating the geometric positions on the infrared calibration light bar corresponding to the positions of the reflection bright spot pixels detected by the high-speed linear array sensors, selecting two geometric positions from the geometric positions to form straight lines with the optical centers of the two high-speed linear array sensors respectively, obtaining target passing coordinates of the shot by intersection, and transmitting the target passing coordinates of the shot to an upper computer.

5. The automatic target reporting and ballistic measuring system of a correlated double sampling linear array sensor as claimed in claim 1, wherein: the infrared calibration light bar control board is controlled through infrared calibration light bar driving logic of the main control board, and each light chain of the infrared calibration light bar is sequentially lightened when an optical calibration meter is built at the initial power-on moment;

the infrared calibration light bar is composed of infrared emission LEDs which are uniformly distributed and grouped into a plurality of light chains on a PCB, and is sequentially arranged at the bottom side or the top side of the three dark background side frames;

the visual range of the N high-speed linear array sensors completely covers three dark background side frames, and the front ends of the high-speed linear array sensors are provided with window type infrared filters and are provided with common lenses with light slits.

6. An automatic target reporting and ballistic measuring method for a correlated double-sampling linear array sensor by using the automatic target reporting and ballistic measuring system for the correlated double-sampling linear array sensor of claim 1, characterized in that:

if the number of the high-speed linear array sensors is 2, the automatic target reporting and trajectory measuring system of the related double-sampling linear array sensor is used for realizing a method for detecting the target passing position of the shot of the low-speed shot;

if the number of the high-speed linear array sensors is 3, the automatic target reporting and trajectory measuring system of the related double-sampling linear array sensor is used for realizing a method for detecting the target passing position of the projectile of each speed projectile;

and if the number of the high-speed linear array sensors is more than 3, the automatic target reporting and trajectory measuring system of the related double-sampling linear array sensor is used for a method for detecting the target passing position and the target passing time of each speed projectile.

7. The automatic target reporting and ballistic measuring method of the correlated double-sampling linear array sensor as claimed in claim 6, wherein:

the method for detecting the target passing position of the low-speed projectile comprises the following steps:

step 1: the method comprises the steps that a main control board controls an infrared light source to be turned off, infrared calibration light bars are controlled to light lamp chains one by one through an infrared calibration light bar control board, namely, each lamp chain is turned off after being turned on for a plurality of scanning cycles, meanwhile, an image of each lamp chain is collected by outputting in-phase driving logic to 2 high-speed linear array sensors to sequentially light the infrared calibration light bars, a bright spot area is judged by an automatic threshold method, a pixel coordinate center of each bright spot of the lamp chain on an induction area (namely an imaging area) of the linear array sensor is gradually calculated in the bright spot area by a centroid method, the pixel coordinate center and the geometric position of an LED correspond to the geometric positions of the LED in an anticlockwise sequence according to the sequence of the pixel coordinate of each sensor, and an optical calibration table is constructed;

step 2: the method comprises the steps that an infrared calibration light bar is closed by a main control board through an infrared calibration light bar control board, an infrared light source is controlled to be lightened, the main control board outputs in-phase driving logic to N2 high-speed linear array sensors to acquire images of dark background side frames shielded by brighter shots, a bright spot area is judged by an automatic threshold value method, pixel coordinate centers of imaging bright spots reflected by the shots when the shots pass through on a linear array sensor sensing area (namely an imaging area) are calculated in the bright spot area by a centroid method, an optical calibration table is inquired, linear interpolation is carried out to calculate geometric positions on the infrared calibration light bar corresponding to the bright spots, the geometric positions corresponding to the two bright spots respectively form straight lines with optical centers of the two high-speed linear array sensors, and shot passing coordinates are obtained by intersection.

8. The automatic target reporting and ballistic measuring method of the correlated double-sampling linear array sensor as claimed in claim 6, wherein:

the method for detecting the passing target position of each type of speed projectile comprises the following steps:

step 1: the method comprises the steps that a main control board controls an infrared light source to be turned off, infrared calibration light bars are controlled to be turned on one by one through an infrared calibration light bar control board, namely, each light bar is turned off after being turned on for a plurality of scanning cycles, meanwhile, a phase shift driving logic is output to N-3 high-speed linear array sensors to collect images of the infrared calibration light bars, each light chain is sequentially turned on, a bright spot area is judged through an automatic threshold value method, the pixel coordinate center of each bright spot of each light chain on an induction area (namely an imaging area) of the linear array sensor is gradually calculated in the bright spot area through a centroid method, the pixel coordinate center and the geometric position of an LED correspond to the geometric positions of the LED in an anticlockwise sequence according to the sequence of the pixel coordinate of each sensor, and an optical calibration table is constructed;

step 2: the method comprises the steps that an infrared calibration light bar is closed by a main control board through an infrared calibration light bar control board, an infrared light source is controlled to be lightened, the main control board outputs phase-shifting driving logic to N-3 high-speed linear array sensors to collect images of side frames of a dark background, the side frames are shielded by bright shots, a bright spot area is judged by an automatic threshold method, pixel coordinate centers of imaging bright spots, reflected when the shots penetrate through, of linear array sensor sensing areas (imaging areas) are calculated in the bright spot area by a centroid method, an optical calibration table is inquired, geometric positions of the infrared calibration light bar corresponding to the bright spots are calculated through linear interpolation, any two groups of sensors and the geometric positions of the bright spots are taken, the geometric positions corresponding to the two bright spots form straight lines with the optical centers of the two high-speed linear array sensors respectively, and intersection is carried out to obtain target-passing.

9. The automatic target reporting and ballistic measuring method of the correlated double-sampling linear array sensor as claimed in claim 6, wherein:

the method for detecting the target passing position and the target passing time of each type of speed projectile comprises the following steps:

step 1: the method comprises the steps that a main control board controls an infrared light source to be turned off, infrared calibration light bars are controlled to be turned on one by one through an infrared calibration light bar control board, namely each light bar is turned off after being turned on for multiple scanning cycles, the main control board outputs phase-shifting driving logic to N high-speed linear array sensors, 3 high-speed linear array sensors are selected from the N high-speed linear array sensors to serve as original pixel acquisition sensors, images of the sensors returning to original pixel output and turning on each light bar in sequence are acquired, the phase-shifting phase of the original pixel acquisition sensors needs to be guaranteed, the non-exposure time of any one high-speed linear array sensor in the 3 original pixel acquisition sensors is covered by the exposure time of other two sensors, a bright spot area is judged by an automatic threshold value method, the pixel coordinate center of each bright spot on an induction area is calculated by a centroid method in the bright spot area, and the pixel coordinate center of each bright spot and the geometric position of an LED are arranged in a mode that the Should, in order to construct the optical calibration table of said 3 raw pixel acquisition sensors (i.e. the sensors returning the raw pixel output), N is greater than 3;

step 2: the method comprises the steps that a main control board closes an infrared calibration light bar through an infrared calibration light bar control board, an infrared light source is controlled to be lightened, the main control board outputs phase shift to drive a logic high-speed linear array sensor to collect an image of a dark background side frame shielded by a brighter shot, wherein 3 original pixel collection sensors in step 1 have original pixel output information, the rest sensors only have bright spot position information to save IO, a bright spot area is judged by an automatic threshold method, a centroid method is used for calculating a pixel coordinate center of an imaging bright spot, reflected when the shot passes through, on a linear array sensor sensing area (namely an imaging area) of the 3 high-speed linear array sensors, an optical calibration table is inquired for the 3 high-speed linear array sensors, a linear interpolation is used for calculating a geometric position on the infrared calibration light bar corresponding to the bright spot, and due to the phase shift, the fact that the non-exposure time of any one linear array sensor in the 3 original pixel collection sensors in step 1 can be guaranteed by other two sensing areas When the exposure time of the device is covered, when the 3 original pixel acquisition sensors have shot reflection bright spots when the same serial number is scanned and imaged periodically, 2 or 3 linear array sensors exist in the 3 original pixel acquisition sensors, any two groups of sensors and the geometric positions of the bright spots are taken, the geometric positions corresponding to the two bright spots respectively form straight lines with the optical centers of the two high-speed linear array sensors, and the shot passing-target coordinates are obtained by intersection;

and step 3: if the target passing coordinate can be calculated for the scanning period with a certain serial number, searching each sensor in the scanning period, if one or more sensors are searched and no bright spots appear in the scanning period, taking the middle point of the sensor in the middle phase in the non-exposure time of the scanning period as the shot target passing time, if the condition is not searched, continuing to search each scanning period before and after the scanning period, and if bright spots appear in two periods before and after the non-exposure time of one or more sensors, taking the middle point of the sensor in the middle phase in the non-exposure time as the target passing time.

Technical Field

The invention belongs to the technical field of digital shooting target scoring, and particularly relates to an automatic target scoring and trajectory measuring system of a related double-sampling linear array sensor.

Background

Shooting is an indispensable item in shooting competitions and military training. The manual target scoring method has the problems of poor safety and timeliness. The automatic intelligent target scoring system can improve the efficiency and the precision and effectively ensure the safety of personnel. The automatic target reporting device is connected with the middle parts of a plurality of target surfaces in series along the trajectory track of the projectile and is capable of accurately measuring the target passing time and position coordinates of the projectile, and the automatic target reporting device can also form a trajectory measuring system and provide powerful support for criminal investigation and gun correction. Existing automatic target scoring systems are generally divided into five categories according to principles: the method comprises an ultrasonic shock wave method, a double-layer electrode short circuit sampling method, a laser pair tube or light curtain method, an image processing method and a high-sensitivity double-line array camera intersection method. Wherein the ultrasonic shock wave method utilizes the shock wave generated when the projectile flies at supersonic speed or the ultrasonic wave generated when the projectile punctures the target material to detect, for example, the shock wave target-reporting system with the patent number of CN200610102134.X requires that the projectile must be a supersonic speed projectile, and can not detect partial pistol with lower projectile speed; the accuracy of the ultrasonic shock wave method is greatly influenced by the wind speed and the target passing time of the projectile cannot be measured. The double-layer electrode short circuit sampling method adopts 2 layers of conductive layers and an insulating layer sandwiched between the conductive layers to manufacture a target material, and corresponding points are conducted to detect when a projectile passes through, for example, the device for automatically reporting the target in firing practice of light arms with the patent number of CN201310077701.0 and the working method thereof have the defects that the target material is expensive, the use cost is high, only metal projectiles can be measured, and the target passing time of the projectiles cannot be measured. An image processing method, such as an automatic shooting target scoring system based on image recognition disclosed in patent No. CN201310249220.3, uses a camera to shoot a target surface for image processing, so as to obtain the position of a shot point, has a certain requirement on the finish degree of the target surface, and has poor detection performance for the overlapping situation of the shot points, and cannot measure the shot passing time. The laser pair tube or light curtain method uses a laser pair tube array, a fan-shaped linear light source and a receiving tube array or an optical fiber array to form an infrared light curtain, and shields the infrared light curtain to be detected when a projectile passes through the infrared light curtain, for example, the light curtain array automatic target-reporting device and method of patent No. CN200710019000.6 have the defects that the precision requirement of light curtain manufacture and installation is higher, so the cost is high, and in addition, the pointing direction of the laser tube can be gradually changed along with the impact of the projectile in use, and the laser tube needs to be adjusted again at intervals. Companies such as Kurzzeit, germany, also use high-precision hole array structures to limit the laser-to-tube arrays to ensure that the light curtains maintain strict alignment and avoid re-alignment, but the machining precision requirement is high. The precision of the laser pair tube light curtain method is high, and the shot passing time can be measured; when the carrier-modulated laser is used for tube alignment, the infrared light anti-interference capability to outdoor sunlight is good, and when other light curtain implementation modes are used, the infrared light anti-interference capability is poor, and auxiliary structures such as slits and optical filters are generally needed. The high-sensitivity double-linear-array camera intersection method is also used for detecting by replacing a receiving tube with an ultrahigh-sensitivity linear-array camera or a linear-array sensor, the former is a shooting auxiliary training room based on a high-speed industrial camera such as CN201910505380.7, two or more high-sensitivity high-speed industrial linear-scan cameras are generally used for shooting of dozens of KHz/s, for example, a linear-scan camera of DALSA company in Canada is matched with a professional acquisition card for realization, the data transmission and processing pressure is large and the price is high, the high positioning precision can be still kept under the condition of a large target surface, but the target passing time of a projectile cannot be measured; the latter has several forms, one is with imaging structure, infrared light source and linear array sensor with imaging structure are located at both sides of the bullet to be measured, such as CN201510487949.3 automatic target reporting and speed measuring device and its positioning and speed measuring method, can position the bullet except for the glowing bullet accurately, and can measure the time when the bullet passes the target roughly by adding the second optical sensor, the time error is equal to the exposure period of the second optical sensor, because the carrier modulation can not be used, and the reflected sunlight when the bullet passes the target is also received by the linear array sensor to cause the detection dark spot to be 'filled up', therefore the infrared light anti-interference ability to the outdoor sunlight is poor, the interference can only be reduced by increasing the brightness of the background infrared light source and increasing the structures such as extinction helix, the sensitivity of the linear array sensor required by this kind of method needs to far exceed the common linear array CCD, the corresponding to the oversized pixel sensing surface and the larger extinction helix result in the structure is bigger, for example, the TSL208 has only 640 pixels and is 64mm long, and this scheme requires the sensor to operate in a continuous exposure mode in which the sensor is exposed for the whole scanning period, thereby limiting the selection range of the sensor; the other is that the device is not provided with an imaging structure, the infrared light source and the linear array sensors on the side edge of the target surface are also positioned on two sides of the projectile to be detected, such as a laser target reporting device and a laser target reporting method of CN201810599455.8, but the sensing surface of a single linear array sensor is required to be overlong, only the CIS linear array sensor can be used actually, and only splicing can be carried out to form the side edge of the target surface. The high-sensitivity double-linear-array camera intersection method is characterized in that a shot positioned between a linear-array sensor and a background light source shields the background light source for detection, the shot is required to be detected through a considerable part of the target surface sensing time occupying the exposure time to obtain obvious detection dark spots, and the method works in a continuous exposure mode in which the exposure time continuously occupies the whole sensor scanning period to avoid scanning omission, so that the dark spots are not obvious to cause poor detection performance of the high-speed shot.

Disclosure of Invention

The invention can accurately detect the target passing position of the projectile, is suitable for various types of projectiles, has simple installation process, easy maintenance and less interference by infrared light of the environment such as the sun, and the like, and the automatic target-scoring device can respectively provide the following functions according to the difference of the number of sensors and the driving logic:

detecting the target passing position of the projectile under the condition of low-speed projectile, detecting the target passing position of the projectile under the condition of various speeds of the projectile, and detecting the target passing position and the target passing moment of various speeds of the projectile;

the technical scheme of the device is that the automatic target-reporting and trajectory measuring system of the related double-sampling linear array sensor is characterized by comprising the following components: the system comprises a target surface, an infrared light source, three dark background side frames, a main control board, an infrared calibration light bar and N (N is more than or equal to 2) high-speed linear array sensors;

the infrared light source is arranged on one side edge of the target surface, the three dark background side frames are sequentially arranged on the other three side edges of the target surface, and the N (N is more than or equal to 2) high-speed linear array sensors are arranged on the side edge where the infrared light source is arranged; the main control board is respectively connected with the N (N is more than or equal to 2) high-speed linear array sensors in sequence; the main control board is connected with the infrared calibration light bar control board, and the infrared calibration light bar control board is sequentially connected with the infrared calibration light bar;

target paper is flatly laid on the target material forming the target surface;

the infrared light source emits infrared light parallel to the target surface, covers the whole target surface and irradiates the three dark background side frames on the opposite side, and is not lightened at the initial power-on moment, namely in the process of constructing the optical calibration table, and is continuously lightened after the optical calibration table is constructed;

the three dark background side frames have a certain width and are arranged around the other three side edges outside the edge where the target surface infrared light source is located, and the side frames are covered with light absorption materials which can effectively absorb infrared light emitted by the infrared light source to the side frames;

the main control board comprises a voice playing module, a display module and a network transmission module;

the main control board is used for sending out driving logics of the high-speed linear array sensors, the infrared light source and the optional infrared calibration light bars, receiving data collected by the high-speed linear array sensors, constructing an optical calibration table at the initial power-on moment, inquiring the optical calibration table and calculating geometric positions on the infrared calibration light bars corresponding to the bright spots in an interpolation mode, enabling the two geometric positions to form straight lines with optical centers of the two high-speed linear array sensors respectively, obtaining projectile target-passing coordinates through intersection, and transmitting the projectile target-passing coordinates to an upper computer;

the infrared calibration light bar control board is controlled through infrared calibration light bar driving logic of the main control board, and each light chain of the infrared calibration light bar is sequentially lightened when an optical calibration meter is built at the initial power-on moment;

the infrared calibration light bar is composed of infrared emission LEDs which are uniformly distributed and grouped into a plurality of light chains on a PCB, and is sequentially arranged at the bottom side or the top side of the three dark background side frames;

the visual range of the N high-speed linear array sensors completely covers three dark background side frames, and the front ends of the high-speed linear array sensors are provided with window type infrared filters and are provided with common lenses with light slits.

The infrared calibration light bar control panel and the infrared calibration light bar are optional, and are only used once before leaving a factory to perform optical calibration under the condition except for high power projectiles such as aeronautical cannons and the like, and are removed after leaving the factory.

If the number of the high-speed linear array sensors is 2, the method is applied to the target passing position detection of the projectile under the condition of low-speed projectile, and the technical scheme of the method is a target passing position detection method of the low-speed projectile, and the method comprises the following steps:

step 1: the method comprises the steps that a main control board controls an infrared light source to be turned off, infrared calibration light bars are controlled to light lamp chains one by one through an infrared calibration light bar control board, namely, each lamp chain is turned off after being turned on for a plurality of scanning cycles, meanwhile, an image of each lamp chain is collected by outputting in-phase driving logic to 2 high-speed linear array sensors to sequentially light the infrared calibration light bars, a bright spot area is judged by an automatic threshold method, a pixel coordinate center of each bright spot of the lamp chain on an induction area (namely an imaging area) of the linear array sensor is gradually calculated in the bright spot area by a centroid method, the pixel coordinate center and the geometric position of an LED correspond to the geometric positions of the LED in an anticlockwise sequence according to the sequence of the pixel coordinate of each sensor, and an optical calibration table is constructed;

step 2: the method comprises the steps that an infrared calibration light bar is closed by a main control board through an infrared calibration light bar control board, an infrared light source is controlled to be lightened, the main control board outputs in-phase driving logic to N2 high-speed linear array sensors to acquire images of dark background side frames shielded by brighter shots, a bright spot area is judged by an automatic threshold value method, pixel coordinate centers of imaging bright spots reflected by the shots when the shots pass through on a linear array sensor sensing area (namely an imaging area) are calculated in the bright spot area by a centroid method, an optical calibration table is inquired, linear interpolation is carried out to calculate geometric positions on the infrared calibration light bar corresponding to the bright spots, the geometric positions corresponding to the two bright spots respectively form straight lines with optical centers of the two high-speed linear array sensors, and shot passing coordinates are obtained by intersection.

If the number of the high-speed linear array sensors is 3, the method is applied to the detection method of the over-target position under the condition of various speed projectiles, and the technical scheme of the method is the detection method of the over-target position of various speed projectiles, and the method comprises the following steps:

step 1: the method comprises the steps that a main control board controls an infrared light source to be turned off, infrared calibration light bars are controlled to be turned on one by one through an infrared calibration light bar control board, namely, each light bar is turned off after being turned on for a plurality of scanning cycles, meanwhile, a phase shift driving logic is output to N-3 high-speed linear array sensors to collect images of the infrared calibration light bars, each light chain is sequentially turned on, a bright spot area is judged through an automatic threshold value method, the pixel coordinate center of each bright spot of each light chain on an induction area (namely an imaging area) of the linear array sensor is gradually calculated in the bright spot area through a centroid method, the pixel coordinate center and the geometric position of an LED correspond to the geometric positions of the LED in an anticlockwise sequence according to the sequence of the pixel coordinate of each sensor, and an optical calibration table is constructed;

step 2: the main control board closes the infrared calibration light bar through the infrared calibration light bar control board, controls the infrared light source to light, outputs phase shift driving logic to N ═ 3 high-speed linear array sensors to collect images of dark background side frames shielded by bright shots, judges bright spot areas by an automatic threshold value method, calculates the pixel coordinate center of the imaging bright spot reflected by the shot passing through the linear array sensor sensing area (namely imaging area) in the bright spot areas by a centroid method, inquires an optical calibration table and calculates the geometric position on the infrared calibration light bar corresponding to the bright spot by linear interpolation, because the phase shift can ensure that the non-exposure time of any linear array sensor is covered by the exposure time of other two sensors, at least 2 or 3 linear array sensors have bright spots when imaging in the same sequence number period, takes any two groups of linear array sensors and the geometric position of the bright spot, and respectively forming straight lines by the geometric positions corresponding to the two bright spots and the optical centers of the two high-speed linear array sensors, and obtaining the shot target-passing coordinate by intersection.

If the method is applied to the detection of the target passing position and the target passing time of various speed projectiles, the technical scheme of the method is the detection method of the target passing position and the target passing time of various speed projectiles, and the method comprises the following steps:

step 1: the main control board controls the infrared light source to be closed, the infrared calibration light bars are controlled to be lightened one by one through the infrared calibration light bar control board, namely each light chain is lightened for a plurality of scanning cycles and then is closed, the main control board outputs phase shift driving logic to more than 3 high-speed linear array sensors, but only images of each light chain which are lightened in sequence of 3 sensors returning to the original pixel output are collected, the phase shift phase of the original pixel collection sensor is ensured, the non-exposure time of any one high-speed linear array sensor in the 3 sensors returning to the original pixel output is covered by the exposure time of other two sensors, the bright spot area is judged by an automatic threshold method, the pixel coordinate center of each bright spot on the induction area is calculated by a centroid method in the bright spot area, and the pixel coordinate center and the geometric position of the LED are in one-to-counter-clockwise correspondence with the geometric position of the LED according to the sequence of the pixel coordinates of each sensor from small to large, to build an optical calibration table of the 3 sensors returning the original pixel outputs;

step 2: the main control board closes the infrared calibration light bar through the infrared calibration light bar control board, controls the infrared light source to light, outputs phase shift driving logic to the high-speed linear array sensor to collect the image of the dark background side frame shielded by the bright pill, wherein, the 3 sensors have original pixel output information, the other sensors only have bright spot position information to save IO, judges the bright spot area by an automatic threshold value method, calculates the pixel coordinate center of the imaging bright spot reflected when the pill passes through the sensing area (namely the linear array imaging area) of the 3 sensors by a centroid method in the bright spot area, inquires the optical calibration table for the 3 sensors and calculates the geometric position of the infrared calibration corresponding to the bright spot by linear interpolation, because the phase shift can ensure that the non-exposure time of any linear array light bar sensor is covered by the exposure time of the other two sensors, therefore, when the images are formed in the scanning period with the same serial number, 2 or 3 linear array sensors have bright spots, any two groups of sensors and the geometric positions of the bright spots are taken, the geometric positions corresponding to the two bright spots respectively form straight lines with the optical centers of the two high-speed linear array sensors, and the shot target-passing coordinates are obtained through intersection.

And step 3: if the target passing coordinate can be calculated for the scanning period with a certain serial number, searching each sensor in the scanning period, if one or more sensors are searched and no bright spots appear in the scanning period, taking the middle point of the sensor in the middle phase in the non-exposure time of the scanning period as the shot target passing time, if the condition is not searched, continuing to search each scanning period before and after the scanning period, and if bright spots appear in two periods before and after the non-exposure time of one or more sensors, taking the middle point of the sensor in the middle phase in the non-exposure time as the target passing time.

The method for detecting the position of the target-passing position is easy to solve by hardware, simplifies the structure of software and hardware, is easy to finish bright spot edge detection in hardware by an automatic dynamic threshold bright spot detection algorithm with hysteresis comparison, avoids uploading or processing the original brightness value of the linear array sensor, does not need to store the original brightness values of a plurality of periods, reduces the data transmission or processing pressure, and is simpler and more convenient compared with the traditional high-sensitivity double-linear array camera intersection method and can expand and use more linear array sensors.

The method for detecting the position of the target-passing position adopts the window infrared filter and the reflective light path, thereby effectively reducing the interference of ambient infrared light such as sunlight and the like and avoiding the adoption of a relatively complex extinction structure such as an extinction spiral with a relatively large volume and the like.

In order to solve the problem of missing scanning, the method for detecting the position of the target passing position of each type of speed projectile adopts at least 3 high-speed linear array sensors, and solves the problem of missing scanning by using a mode that the non-exposure time and the exposure time of the related double sampling stages among the sensors are mutually covered, thereby avoiding the defect of poor detection performance of the traditional high-sensitivity double linear array camera intersection method for the high-speed projectile

Furthermore, the method for detecting the target passing position and the target passing time of the projectile with various speeds utilizes more high-speed linear array sensors to carry out phase shifting, reduces the detection error of the projectile at the target passing time to T/n (T is exposure time, and n is the number of sensors), and therefore, phase shifting time slices can be reduced by using a plurality of sensors to shift the phases so as to obtain higher detection precision of the target passing time.

Furthermore, the invention adopts a plurality of devices for detecting the target passing position and the target passing time of various types of pills to be sequentially arranged on the possible track range of the pills, each device adopts the same main clock frequency, line synchronization pulses are sequentially transmitted to each device, the transmission delay among the devices is measured, and the time system function among the devices is completed by utilizing the line synchronization pulses, so that the time and the position coordinates of the pills passing through each device are obtained, and complete trajectory information is formed.

Drawings

FIG. 1: the invention is an integral system block diagram.

FIG. 2: exemplary embodiment 1 of the invention is schematically illustrated.

FIG. 3: wavelength-transmittance curve of the window infrared filter.

FIG. 4: exemplary implementation 1 in-phase drive timing and exposure/non-exposure time of the sensors 103 and 104.

FIG. 5: and arranging each lamp chain of the infrared calibration lamp strip PCB.

FIG. 6: the installation mode in the side frame.

FIG. 7: exemplary embodiment 2 of the present invention.

FIG. 8: exemplary implementation 2 phase shift drive timing and exposure/non-exposure times for sensors 103, 112, and 104.

FIG. 9: exemplary embodiment 3 of the present invention.

FIG. 10: exemplary implementation 3 phase shift drive timing and exposure/non-exposure times for sensors 103, 113, 112, 114, and 104.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The following describes an embodiment of the present invention with reference to fig. 1 to 10:

the overall system block diagram of the present invention is shown in fig. 1. An automatic target-reporting and trajectory measuring system of a related double-sampling linear array sensor is characterized by comprising: the system comprises a target surface, an infrared light source, three dark background side frames, a main control board, an infrared calibration light bar and N (N is more than or equal to 2) high-speed linear array sensors;

the infrared light source is arranged on one side edge of the target surface, the three dark background side frames are sequentially arranged on the other three side edges of the target surface, and the N (N is more than or equal to 2) high-speed linear array sensors are arranged on the side edge where the infrared light source is arranged; the main control board is respectively connected with the N (N is more than or equal to 2) high-speed linear array sensors in sequence; the main control board is connected with the infrared calibration light bar control board, and the infrared calibration light bar control board is sequentially connected with the infrared calibration light bar;

target paper is flatly laid on the target material forming the target surface;

the infrared light source emits infrared light parallel to the target surface, covers the whole target surface and irradiates the three dark background side frames on the opposite side, and is not lightened at the initial power-on moment, namely in the process of constructing the optical calibration table, and is continuously lightened after the optical calibration table is constructed;

the three dark background side frames have a certain width and are arranged around the other three side edges outside the edge where the target surface infrared light source is located, and the side frames are covered with light absorption materials which can effectively absorb infrared light emitted by the infrared light source to the side frames;

the main control board comprises a voice playing module, a display module and a network transmission module;

the main control board is used for sending out driving logics of the high-speed linear array sensors, the infrared light source and the optional infrared calibration light bars, receiving data collected by the high-speed linear array sensors, constructing an optical calibration table at the initial power-on moment, inquiring the optical calibration table, performing interpolation calculation on geometric positions on the infrared calibration light bars corresponding to the bright spots, enabling the two geometric positions to respectively form straight lines with optical centers of the two high-speed linear array sensors, obtaining projectile target-passing coordinates through intersection, and transmitting the projectile target-passing coordinates to an upper computer;

the infrared calibration light bar control board is controlled through infrared calibration light bar driving logic of the main control board, and each light chain of the infrared calibration light bar is sequentially lightened when an optical calibration meter is built at the initial power-on moment;

the infrared calibration light bar is composed of infrared emission LEDs which are uniformly distributed and grouped into a plurality of light chains on a PCB, and is sequentially arranged at the bottom side or the top side of the three dark background side frames;

the visual range of the N high-speed linear array sensors completely covers three dark background side frames, and the front ends of the high-speed linear array sensors are provided with window type infrared filters and are provided with common lenses with light slits.

The infrared calibration light bar control panel and the infrared calibration light bar are optional, and are only used once before leaving a factory to perform optical calibration under the condition except for high power projectiles such as aeronautical cannons and the like, and are removed after leaving the factory.

The following examples 1-3 can be regarded as the implementation of the system block diagram with certain tailoring (in the case of example 1, only 2 sensors do not shift the phase).

The line CCD/CMOS sensor in the high speed line sensors in embodiments 1 to 4 does not refer to any type of line sensor, and it is not required to be able to continuously expose the line sensor in the whole period, nor to have ultra high sensitivity, so some line sensors with medium sensitivity can also meet the requirements, such as RL0512P, RL1024P, RL2048P of Perkinelmer, IL-P3 and IL-P4 of DALSA, Dragster 2K7 and 4K7 of AMS, as long as these sensors can sense near infrared light within 780 to 1000nm, and still have a short scanning period (generally less than 100 us) and good consistency of photoelectric gain (i.e. sensitivity) when a CDS (correlated double sampling) operation mode with at least 10bit ad sampling accuracy is used. Because the ultrahigh sensitivity is not required, a sensor with smaller pixels can be selected, so that enough sensor pixels are provided in the acceptable sensor volume, the higher positioning precision can be still ensured under the condition of a large target surface, and the high-speed linear array sensor with the pixels of 2K or more can be selected as much as possible. The linear array sensor with ultrahigh sensitivity such as AMS TAOS TSL208R is to be avoided, firstly, in order to obtain ultrahigh sensitivity (ultrahigh sensitivity is easily obtained if a photosensitive surface of a pixel is large and a photosensitive area is large), a pixel needs to be made large, so that the number of pixels is small, for example, TSL208R only has 512 pixels under the length of a 64mm long sensing area, so that in the embodiment, because the length of the overlong sensing area can only use a cylindrical mirror to ensure that the distance between the sensing surface and the target surface of the sensor is not too large, so that the positioning error is prevented from increasing when the projectile is obliquely incident, which also causes the problem that the structure volume is too large because an extinction spiral needs to be used in cooperation with the cylindrical mirror; second, ultrahigh sensitivityThe photoelectric gain uniformity of the sensor is not good, for example, the individual photoelectric gain of the TSL208R is 16-28V/(uJ/cm)²) The output voltage during saturation sensing fluctuates in such a large range, which causes the device difference to be 1.6-2.4V, so that the sensor board of the TSL208R needs to measure the photoelectric gain of the sensor board aiming at the individual TSL208R and correspondingly set the zero point and the gain of the ADC front-stage operational amplifier, which causes a tedious debugging process aiming at the individual TSL208R sensor board; and thirdly, the ultra-high sensitivity sensor is not high in the low-speed working clock due to the fact that the single-pixel charge packet electric quantity is large, the scanning period is long after the related double sampling step with 10-bit AD sampling precision is added, for example, the shortest scanning period is 102.6us when the TSL208R does not use the related double sampling under the highest working clock of 5MHz, the shortest scanning period is 512.2us after the related double sampling step with 10-bit AD sampling precision is used, wherein the non-exposure time during the related double sampling period is at least 204us, and a lot of pills passing through the induction surface can be missed in the non-exposure time. Because the embodiment does not require a continuous exposure mode and ultrahigh sensitivity when the exposure time occupies the whole scanning period, zero point influence and reset noise can be removed by using related double sampling and other modes, a complex adjustment process is avoided, and meanwhile, a sensor with small volume and more pixels can be selected as much as possible to ensure the positioning accuracy of a large target surface.

Example 1 is described below, with corresponding parts exaggerated in size for clarity.

The method provided by the embodiment of the invention comprises the following steps of:

step 1: the method comprises the steps that a main control board controls an infrared light source to be turned off, infrared calibration light bars are controlled to light lamp chains one by one through an infrared calibration light bar control board, namely, each lamp chain is turned off after being turned on for a plurality of scanning cycles, meanwhile, an image of each lamp chain is collected by outputting in-phase driving logic to 2 high-speed linear array sensors to sequentially light the infrared calibration light bars, a bright spot area is judged by an automatic threshold method, a pixel coordinate center of each bright spot of the lamp chain on an induction area (namely an imaging area) of the linear array sensor is gradually calculated in the bright spot area by a centroid method, the pixel coordinate center and the geometric position of an LED correspond to the geometric positions of the LED in an anticlockwise sequence according to the sequence of the pixel coordinate of each sensor, and an optical calibration table is constructed;

step 2: the method comprises the steps that an infrared calibration light bar is closed by a main control board through an infrared calibration light bar control board, an infrared light source is controlled to be lightened, the main control board outputs in-phase driving logic to N2 high-speed linear array sensors to acquire images of dark background side frames shielded by brighter shots, a bright spot area is judged by an automatic threshold value method, pixel coordinate centers of imaging bright spots reflected by the shots when the shots pass through on a linear array sensor sensing area (namely an imaging area) are calculated in the bright spot area by a centroid method, an optical calibration table is inquired, linear interpolation is carried out to calculate geometric positions on the infrared calibration light bar corresponding to the bright spots, the geometric positions corresponding to the two bright spots respectively form straight lines with optical centers of the two high-speed linear array sensors, and shot passing coordinates are obtained by intersection.

Embodiment 1 of the invention is as shown in fig. 2, and an automatic target scoring device comprises a target surface, an infrared light source, three side frames, a main control board, 2 high-speed linear array sensors, an optional infrared calibration light bar control board and an infrared calibration light bar. Where components other than the target surface should be mounted on the rear side of the ballistic structure to avoid damage from projectiles. The concrete constitution is as follows:

leveling a target 111, and flatly paving target paper 110 attached to the target, wherein the target and the target paper form a target surface;

the high-power infrared light source 105 is composed of a 850nm infrared LED lamp strip with high power and a long cylindrical lens arranged side by side with the infrared LED lamp strip, 60 5050 packaged 850nm infrared LEDs are uniformly arranged on the lamp strip per meter of length, the power is about 14.4W, the long cylindrical lens collects light of the infrared LED lamp strip to enable most of the light to be parallel to a target surface, and therefore when a projectile penetrates through a detection surface formed by the infrared LED lamp strip and the high-speed linear array sensor, the projectile can be illuminated by the infrared light source to reflect infrared light, and therefore bright spots are imaged at corresponding positions of an induction area of the high-speed linear array sensor;

a dark background 109 composed of three side frames, as shown in fig. 6, the side frames are composed of black aluminum alloy sections, extinction velvet is attached to one side of the side frames facing the center of the target surface, and except for the part reflected by the shot, the rest infrared light emitted by the infrared light source 105 reaches the three side frames and is absorbed by the extinction velvet;

the calibration method comprises the following steps that a main control board 101 is provided with a ZYNQ7020 FPGA, the FPGA outputs a driving time sequence to close an infrared light source 105 in a calibration period, the infrared calibration light bar control board 102 is controlled to sequentially light LEDs of 1 st to 5 th light chains of infrared calibration light bars 106, 107 and 108 in the initialization stage (the sequential lighting of the sub light chains is to avoid that imaging bright spots of adjacent LEDs are fused due to too close distance and are difficult to detect, the precision of a calibration table can be improved by using denser LEDs for the sub light chains), the lighting time of each light chain lasts for 2 scanning periods, each lighted LED images as a bright spot at a corresponding position of an induction area of a high-speed linear array sensor, the FPGA detects and calculates a pixel coordinate center of each bright spot on the induction area by a centroid method according to pixel brightness, and corresponds to the geometric position of the LED, and a calibration table is formed; entering a normal working stage after the initialization stage is finished, namely target reporting working time, outputting a driving time sequence by the FPGA to turn on the infrared light source 105 and turn off all infrared calibration light bars, detecting by the FPGA and calculating a pixel coordinate center of an imaging bright spot reflected when a projectile passes through an induction area by a centroid method according to pixel brightness, checking a calibration table and calculating geometric positions on the infrared calibration light bars corresponding to the bright spot in an interpolation manner, wherein the two geometric positions respectively form straight lines with the optical centers of the two high-speed linear array sensors, and intersection is carried out to obtain target passing coordinates of the projectile;

the 2 high-speed linear array sensors 103 and 104 arranged on the sides outside the three side frames are composed of imaging structures such as a high-speed linear array CCD/CMOS sensor, an interface circuit, a window optical filter, an imaging lens and the like; the high-speed linear array sensor has a large visual angle, so that the detection fan-shaped range can cover the whole target surface. The high-speed linear array CCD/CMOS sensor receives driving logic from the main control board FPGA and works in a mode of being beneficial to eliminating zero points and drift and suppressing noise in a correlated double sampling mode or a correlated multiple sampling mode. It should be noted that, during the related double sampling AD conversion, the high-speed linear CCD/CMOS sensor needs to be in the non-exposure state for a certain period of time, and the working timing and the exposure/non-exposure time are as shown in fig. 4, and the high-speed projectile may be missed during the non-exposure time, so the detection performance for the high-speed projectile is not good. A similar problem exists with some high speed industrial line scan cameras, and how to solve the detection of high speed projectiles is shown in example 2. Specifically, for the high-speed linear array CMOS sensor of the present embodiment, the sensitivity is 65.5V/Lux · s, 2080 pixels are in total, the pixel pitch is 4um, and the length of the sensing area is only 8.32mm, so that a small-sized M12 lens holder can be used to reduce the volume of the sensor and reduce the distance between the sensing surface and the target surface of the sensor, a sector angle range of not less than 89 degrees can be obtained when a 4.23mm lens is used, a sector angle range of not less than 92 degrees can be obtained when a 4mm lens is used, a sector angle range of not less than 98 degrees can be obtained when a 3.6mm lens is used, a sector angle range of not less than 112 degrees can be obtained when a 2.8mm lens is used, and a sector angle range of not less than 122 degrees can be obtained when a 2.3mm lens is used. In this embodiment 1, 4.23mm lenses are used for 2 high-speed line sensors, and in this embodiment 2, 2.3mm lenses are used for 1 increased high-speed line sensor 112. A window filter is required to be attached to a sensing area of the high-speed linear array sensor, a wavelength range allowed to be transmitted by the window filter (also called a narrowband filter) should be matched with an infrared light source, for example, a wavelength range transmitted by the window filter matched with an infrared light source of 850nm is as shown in fig. 3, so as to reduce interference of ambient light such as the sun. The window filter can also be omitted or a common high-pass filter (passing wavelengths higher than 780 nm) can be used indoors or in cases where the requirements on ambient light interference are not high.

The infrared calibration light bar control board 102 is composed of an automatic regulation booster circuit with each light chain enabling end, and lights the corresponding light chain in a constant current mode according to the set current when the corresponding light chain is in an enabling opening state. The infrared calibration light bar control board 102 receives driving logic of the infrared calibration light bar from the main control board to drive each light chain enable terminal.

The PCB of the infrared calibration light bars 106, 107, 108 mounted in the three side frames and the mounting in the side frames are shown in fig. 5 and 6, respectively. The infrared calibration light bar adopts 850nm small patch infrared side emission LEDs with a larger beam angle, the infrared side emission LEDs are arranged at fixed intervals of 6.35mm, the height of each infrared side emission LED is about 1mm, most of the emitted infrared light is parallel to the surface of the PCB, and the laying range of the extinction velvet in the side frame covers the irradiation range of the infrared light source 105 on the opposite side. For a common light projectile, the infrared calibration light bar control board 102 and the infrared calibration light bars 106, 107 and 108 can be used only once when an optical calibration table is determined by factory calibration, and are not reserved after factory calibration; for heavy projectiles, the infrared calibration light bar control board 102 and the infrared calibration light bars 106, 107 and 108 are kept all the time, the optical calibration is automatically performed once in the calibration time and the optical calibration table is updated when the electric projectile is powered on in each use, and the projectile target-passing coordinate error caused by unrecoverable deformation caused by impact is avoided through the automatic calibration process.

The following description is provided to how to complete the establishment of the optical calibration table and the measurement of the shot target-passing coordinate in example 1, to explain the reason why the detection performance of the optical calibration table for a high-speed shot is poor, and to explain the reason why the infrared calibration light bar control board and the infrared calibration light bar are not retained/retained respectively for a light shot and a heavy shot after the optical calibration table is established.

When the power-on starts, the main control board 101 respectively outputs 5 enabling pulse signals to the infrared calibration light bar control board 102 through 5 control channels, the width of each pulse signal corresponds to 2 scanning periods in fig. 4, the channels 1-5 of the voltage boosting constant current source circuit controlled by the sub-channels on the infrared calibration light bar control board 102 are sequentially turned on and turned off at the time of 2 scanning periods, respectively, the lamp chain 1, the lamp chain 2, the lamp chain 3, the lamp chain 4 and the lamp chain 5 are turned on and turned off, the LEDs turned on in the lamp chain 1-the lamp chain 5 in the 10 scanning periods are respectively imaged as a series of bright spots on the induction surfaces of the high-speed linear array sensors 103 and 104, correspondingly, a series of small peaks are formed on a data curve output by the pixels, the center position coordinates of the small peaks are obtained by a centroid method, and a corresponding table between the geometric positions of the infrared LEDs and the peak coordinates is obtained, namely an optical calibration table.

Impact tests were conducted on this example using a lightweight impact pendulum (simulating light projectiles of small arms) and a heavy impact pendulum (simulating heavy projectiles such as cannons) from directions perpendicular to the target surface and parallel to the target surface, respectively, and a 6mm diameter metal cylinder was affixed to a target surface of about 0.6mx0.9m size to simulate the effect of projectile incidence from all directions on the stability and accuracy of the example. Experiments show that the stability and the precision of the embodiment are not obviously influenced under the condition that the impact pendulum is incident from the position vertical to the target surface; for the condition that the impact pendulum is incident from being parallel to the target surface and hits the side face of the target surface, the light impact pendulum can cause the shaking of the positioning result of about 0.4mm at most, but the shaking is eliminated after the experiment is finished, namely the condition that the light impact pendulum is incident from being parallel to the target surface does not cause obvious influence on the stability and the precision of the embodiment, the heavy impact pendulum can cause the deviation of the positioning result exceeding 2mm after about 6-9 times of impact incident from being parallel to the target surface, and the deviation can not be eliminated after the experiment is finished and kept fixed, namely the condition that the heavy impact pendulum is incident from being parallel to the target surface can cause the obvious fixed deviation of the over-target coordinate of the embodiment. Correspondingly, the optical calibration meters are respectively obtained by utilizing the infrared calibration light bars before and after the impact experiment, the optical calibration meters are hardly changed by using the light impact pendulum, the optical calibration meters are slightly deviated by using the heavy impact pendulum after about 6-9 times of impact incident parallel to the target surface, and the deviation value is not changed after the experiment is stopped, which indicates that the directions of the high-speed linear array sensors 103 and 104 are slightly and permanently changed relative to the target surface. Because the projectile is incident close to the parallel of the target surface, the projectile is generally a fluid projectile, and therefore the accumulated fixed error can be eliminated by self-calibration at each power-on. For light shot, after an optical calibration table is automatically obtained once when the shot leaves a factory, the infrared calibration light bar control board 102 and the infrared calibration light bars 106, 107 and 108 can be removed; for heavy projectiles, the self-calibration process of automatically solving and updating the optical calibration table is carried out at each power-on start so as to ensure the measurement accuracy of the projectile target-passing coordinates.

For the sensor timing and exposure/non-exposure time example shown in FIG. 4, the master clock MCLK operates at 120MHz, each scan period is 6300 MCLK clock cycles (52.5 us total), with the non-exposure time 1388 MCLK clock cycles (11.57 us total) and the exposure time 4912 MCLK clock cycles (40.93 us total). Due to the adoption of 10-bit sampling precision, namely the requirement that the pixel level sampling slope and the reset level sampling slope are both at least 1024 MCLK clock cycles long, the length of the non-exposure time in the correlated double sampling CDS working mode cannot be smaller than the length of the pixel level sampling slope, and therefore if a high-speed projectile passes through a sensing area, the non-exposure time is subjected to leakage detection. Comparing the table below, it can be seen that if the pellet length is equivalent to that of a 92-pistol and the speed is equivalent to that of a 95-pistol, the time for the pellet to pass through the sensing area is less than 11us, which is shorter than the non-exposure time of the present embodiment, and detection leakage may occur.

TABLE 1 common projectile lengths and velocities for light and heavy weapons

Kinds of weapons	Length of the projectile	Initial speed of the projectile	Can sense time
				9mm92 type pistol	10mm	350m/s	34.2us
5.8mm95 type automatic rifle	25mm	950m/s	28.4us
				12.7mm89 type heavy machine gun	55mm	825m/s	69.1us
130mm59 type cannon	450mm	930m/s	486.0us

The following briefly describes how the high-speed linear array sensor eliminates the zero, drift and reset noise problems through the correlated double sampling CDS mode. In the embodiments 1-4, the high-speed linear array sensors all work in the correlated double sampling CDS mode. The high-speed linear array sensor is provided with sensing node arrays and latch array circuits which correspond to pixel arrays of the linear array CCD/CMOS sensor one by one, and is also provided with an operational amplifier, an accumulator, a counter for AD conversion and a comparator circuit. In a scanning period, in the first step, under the simultaneous action of a row RESET pulse and an electronic SHUTTER pulse, a pixel array and an induction node array are RESET; secondly, at the falling edge of the SHUTTER pulse, after the pixel signal is reset, the pixel array enters the exposure time to start integration; thirdly, after the pixel output DOUT of the previous period is finished, the latch array is set to be 0; fourthly, under the action of a row RESET RESET pulse, the sensing node arrays are RESET so as to collect background signal levels, then a RESET level sampling slope is generated under the action of a counter, the RESET levels of the sensing nodes corresponding to the pixels are compared with the RESET level sampling slope through a comparator, and when the levels are equal, the value of the counter is written into the position, corresponding to the pixels, of the latch array in a negative value mode; step 5, under the action of an electronic SHUTTER SHUTTER pulse, the level of a pixel array is amplified by an operational amplifier and enters a sensing node array, namely, the exposure time is ended at the falling edge of the SHUTTER pulse and the acquisition of the level of a pixel signal is completed, then, a pixel level sampling slope is generated under the action of a counter, the pixel level of the sensing node corresponding to each pixel is compared with the pixel level sampling slope by a comparator, when the levels are equal, the value of the counter is accumulated with a negative value (representing a reset level) at the corresponding position of a latch array by an accumulator, and the result is placed at the corresponding position of the latch array, so that zero point, drift and reset noise are eliminated from the output of the pixel level; step 6 starts the output of the pixel level DOUT of the period, and the values in the latch array are sequentially output by DOUT under the control of MCLK.

How the main control board only processes or uploads the sensing data of the pixels in the bright spot area of each sensor by adopting an automatic threshold method is briefly described below, and the method is suitable for embodiments 1 to 4. The sensor outputs a pixel DOUT to the main control board, the bright spot indication signal BOUT is obtained by processing DOUT by an automatic dynamic threshold method with hysteresis comparison, the algorithm can be completed by hardware, specifically, a hysteresis value hys is generally selected to be 4 in the following formula, and DOUT [ n ] represents a pixel level output value of an nth pixel:

if (DOUT [ n-1]/2+ DOUT [ n-2]/4+ DOUT [ n-3]/8+ hys <7 + DOUT [ n ]/8) output BOUT [ n ] is 1, which indicates that the pixel n is in the bright spot region

if (DOUT [ n-1]/2+ DOUT [ n-2]/4+ DOUT [ n-3]/8-hys >7 + DOUT [ n ]/8) output BOUT [ n ] is 0, indicating that the pixel n is not in the bright spot region

The BOUT [ n ] main control board can determine whether the sensor has bright spots in the period, and the bright spot area and the pixel level output value when the bright spots appear. The main control board can only process the pixel data of the bright spot area, the transmission and processing burden is greatly reduced through the method, the 667MHz ARM cortex xA9 processor of the PS part in the ZYNQ7020 FPGA can complete output processing and coordinate calculation, and the original data does not need to be uploaded to an upper computer for processing. After the bright spot area is determined by using an automatic threshold value method, the center of the bright spot, namely the position of a peak, can be obtained by using a centroid method.

To solve the problem of leak detection for high speed projectiles of example 1, example 2 is given below.

The method of embodiment 2 of the invention is a method for detecting the passing target position of pills with various speeds, which comprises the following steps:

step 1: the method comprises the steps that a main control board controls an infrared light source to be turned off, infrared calibration light bars are controlled to be turned on one by one through an infrared calibration light bar control board, namely, each light bar is turned off after being turned on for a plurality of scanning cycles, meanwhile, a phase shift driving logic is output to N-3 high-speed linear array sensors to collect images of the infrared calibration light bars, each light chain is sequentially turned on, a bright spot area is judged through an automatic threshold value method, the pixel coordinate center of each bright spot of each light chain on an induction area (namely an imaging area) of the linear array sensor is gradually calculated in the bright spot area through a centroid method, the pixel coordinate center and the geometric position of an LED correspond to the geometric positions of the LED in an anticlockwise sequence according to the sequence of the pixel coordinate of each sensor, and an optical calibration table is constructed;

step 2: the main control board closes the infrared calibration light bar through the infrared calibration light bar control board, controls the infrared light source to light, outputs phase shift driving logic to N ═ 3 high-speed linear array sensors to collect images of dark background side frames shielded by bright shots, judges bright spot areas by an automatic threshold value method, calculates the pixel coordinate center of the imaging bright spot reflected by the shot passing through the linear array sensor sensing area (namely imaging area) in the bright spot areas by a centroid method, inquires an optical calibration table and calculates the geometric position on the infrared calibration light bar corresponding to the bright spot by linear interpolation, because the phase shift can ensure that the non-exposure time of any linear array sensor is covered by the exposure time of other two sensors, at least 2 or 3 linear array sensors have bright spots when imaging in the same sequence number period, takes any two groups of linear array sensors and the geometric position of the bright spot, and respectively forming straight lines by the geometric positions corresponding to the two bright spots and the optical centers of the two high-speed linear array sensors, and obtaining the shot target-passing coordinate by intersection.