阅读说明：本技术 测量物体空间运动的高速摄影数据处理方法 (High-speed photographic data processing method for measuring object space movement ) 是由 颜文选 赵继伟 杨昆劼 张颖 潘武贤 杨建宏 李超 汪玲 于 2019-09-20 设计创作，主要内容包括：本发明提出一种测量物体空间运动的高速摄影数据处理方法,按照图像中的像素点尺寸,对高速摄影得到的物体标记点原始位移-时间数据信息进行划分；提取每个像素点中的第一个数据点作为有效数据,其余数据为无效数据；之后基于相邻两个像素点之间距离非常小,可以将连续两个标记点的变速运动作为匀变速运动的思路对无效数据进行还原,得到真实运动状态,从而解决了一二阶导数的跳变问题,提高系统精度。(The invention provides a high-speed photographic data processing method for measuring object space motion, which divides original displacement-time data information of object mark points obtained by high-speed photography according to the size of pixel points in an image; extracting a first data point in each pixel point as effective data, and taking the rest data as invalid data; and then, based on the fact that the distance between two adjacent pixel points is very small, the variable speed motion of two continuous mark points can be used as the thought of uniform variable speed motion to restore invalid data to obtain a real motion state, so that the problem of jumping of a second-order derivative is solved, and the system precision is improved.)

1. A high-speed photographic data processing method for measuring the space motion of an object is characterized in that: the method comprises the following steps:

step 1: acquiring original displacement-time data information of object mark points obtained by high-speed photography;

step 2: dividing the original displacement-time data information obtained in the step 1 according to the size of pixel points in the image; extracting a first data point in each pixel point as effective data, and taking the rest data as invalid data;

and step 3: eliminating invalid data, and restoring the invalid data through the following processes:

step 3.1: marking the object mark point at successive times t₀，t₁，t₂，t₃…..t_m+n+kAt a position point of P₀,P₁,P₂,…,P_m,P_m+1,…,P_m+n-2,P_m+n-1,P_m+n,P_m+n+1,…,P_m+n+k-1,P_m+n+k,

The distance between the position of each mark point and the set coordinate origin is S₀，S₁，S₂…S_m+n+k(ii) a According to the division in step 2, wherein P₀,P₁,P₂,…,P_m-1Falls on the first pixel point, P_m,P_m+1,…,P_m+n-1Falls on the second pixel, P_m+n,P_m+n+1,…,P_m+n+k-1Falls on the third pixel point, P_m+n+kFalling on the fourth pixel point; wherein P is₀,P_m,P_m+n,P_m+n+kAs valid data points, corresponding S₀,S_m,S_m+n,S_m+n+kIs the valid distance, the rest are invalid data points;

step 3.2: restoring the data points falling in the first pixel point:

geti is 0,1, 2. m-1, the distance of the mark point falling in the first pixel point after reduction from the set coordinate origin is obtained, and then

Where T is the adjacent time interval and T is the adjacent time interval,

S¹ ₀＝S₀，S¹ _m＝S_m，S¹ _m+n＝S_m+n；

step 3.3: and restoring the data points falling in the second pixel point:

geti is m, m +1, m +2, …, m + n-1, the distance between the mark point in the second pixel point after reduction and the coordinate origin is set, then

Wherein

S¹ _m＝S_m，S¹ _m+n＝S_m+n，S¹ _m+n+k＝S_m+n+k；

Step 3.4: repeating the step 3.1 to the step 3.3 to finish the reduction of the data points in each pixel point to obtain the reduced displacement-time data [ s ]¹ ₀，s¹ ₁，…..]；

And 4, step 4: and performing data fitting on the reduced displacement-time data to obtain a displacement-time fitting function S ═ f (t) of the mark point, and further obtaining the speed v ═ df (t)/dt of the mark point, wherein the acceleration a ═ d²f(t)/dt²。

2. A high-speed photographic data processing method for measuring spatial movement of an object according to claim 1, characterized in that: and 4, performing data fitting on the reduced displacement-time data by adopting a polynomial fitting method.

Technical Field

The invention relates to the technical field of high-speed photography, in particular to a high-speed photography data processing method for measuring object space motion in a non-contact manner.

Background

In a test for researching the space displacement and the attitude angle of an object, the high-precision dynamic measurement of the motion displacement, the speed, the acceleration and the like of the object in a certain time period is often required, and data support is provided for subsequent analysis and optimal design.

The non-contact measurement of the object motion can be realized by adopting high-speed photography, and compared with other measurement methods, the method has the advantages of high measurement precision, small interference, low requirement on test environment, reusability and the like. At present, high-speed photography is widely applied to various test fields, such as measurement of a drop separation deflection angle, bullet flying acceleration, cannonball muzzle-out speed, rocket motion attitude and the like.

The high-speed photography technology can automatically capture displacement-time data in the motion process of an object, and later hopes to obtain speed-time and acceleration-time data of object mark points through the displacement-time data and further analyze the physical motion process of the object.

However, in practice, it is found that data jump occurs in different degrees in speed-time and acceleration-time data obtained by dividing the displacement difference between two adjacent points by the time difference, and the phenomenon does not conform to the actual motion state of an object, and the data jump phenomenon needs to be explained and solved.

For example, in a high-altitude falling experiment of an engine, conventionally, after high-speed photographing data is obtained, displacement-time data of a certain mark point is extracted, and a speed value of the mark point at the time is obtained by dividing a displacement difference between two adjacent points by a time difference, as shown in fig. 1. As can be seen from fig. 1, the speed of the mark point changes with time, and the data continuously jumps, which is not in accordance with the physical process of the engine falling in the air, and indicates that the collected displacement-time data needs to be specially processed. And the phenomenon of data jump in different degrees of speed-time and acceleration-time is found by analyzing the previous multiple test data.

By adopting common data processing methods such as common multipoint smoothing, filtering and the like, data jump cannot be eliminated, the result obtained by analyzing test data is not consistent with the actual motion of the object, and the change relation of the speed and the acceleration of the object along with time cannot be obtained. There is no prior art to review the relevant literature to explain this data hopping problem.

Disclosure of Invention

Technical problem to be solved

As described in the background art, speed-time and acceleration-time obtained by using displacement-time data in high-speed photographic data have data jump phenomena of different degrees, and data jump cannot be eliminated by adopting common data processing methods such as multipoint smoothing and filtering. Experiments show that when the sampling rate is increased, the data jump phenomenon is not solved and even has an aggravation trend, and further, the jump phenomenon is not analyzed to be caused by the fact that the sampling rate is too low.

As can be seen from fig. 1, the speed of the object has a zero value many times as the speed-time varies, i.e. the object has a motion stop in the air. The speed is calculated from two adjacent points, and the zero occurrence speed indicates that the displacement values of the mark points at the adjacent moments are the same. The raw displacement-time data of the marker points are thus analyzed, as shown in fig. 2. It can be seen from fig. 2 that there are continuous "steps" appearing on the displacement-time image of the object, and the occurrence of the step value is just the reason why the velocity-time curve of the object appears with zero value for many times and jumps occur.

Further, the principle of automatically capturing the motion track of the object through high-speed photography is analyzed, and the high-speed photography measurement is to replace the actual position of the mark point with the coordinate of the pixel point where the mark point of the object is located. Because the resolution of the high-speed camera is limited, the mark point may fall on the same pixel point at continuous time, and the high-speed photographic system can process the position falling on one pixel point by the same data, namely the coordinate value of the pixel point replaces the actual position of the mark point. This is the root cause of the step-like appearance of the displacement-time data directly obtained by high-speed photography. This results in discontinuity of the first order (velocity) and second order (acceleration) displacement-time data in the later data processing and multi-point data jump. The common multi-point smoothing and filtering data processing method cannot eliminate the jump, so that the test data is inconsistent with the actual motion of the object.

The step value is just to say that the step value phenomenon occurs because the sampling rate of the system is too high under the condition that the resolution is constant. When the sampling rate is too high, the marking points are all located on the same pixel point at continuous time intervals, and the system can automatically judge that the marking points do not move in the time. It is therefore necessary to eliminate the occurrence of step values by selecting a suitable sampling rate, the magnitude of which is related to the speed of movement and the resolution of the object. However, in practical situations, when a subject is photographed and recorded by high-speed photography, since the motion process of the subject is unknown, the appropriate sampling rate cannot be predicted, and in most cases, the motion state of the subject is constantly changed, so that the appropriate sampling rate of the whole motion process cannot be determined.

Aiming at the new problems, the invention provides a new high-speed photographic data processing method for measuring the space motion of an object on the basis of finding key factors influencing data.

The technical scheme of the invention is as follows:

the high-speed photographic data processing method for measuring the space motion of the object is characterized by comprising the following steps of: the method comprises the following steps:

step 1: acquiring original displacement-time data information of object mark points obtained by high-speed photography;

step 2: dividing the original displacement-time data information obtained in the step 1 according to the size of pixel points in the image; extracting a first data point in each pixel point as effective data, and taking the rest data as invalid data;

and step 3: eliminating invalid data, and restoring the invalid data through the following processes:

step 3.1: marking the object mark point at successive times t₀，t₁，t₂，t₃…..t_m+n+kAt a position point of

P₀,P₁,P₂,…,P_m,P_m+1,…,P_m+n-2,P_m+n-1,P_m+n,P_m+n+1,…,P_m+n+k-1,P_m+n+k,

The distance between the position of each mark point and the set coordinate origin is S₀，S₁，S₂…S_m+n+k(ii) a According to the division in step 2, wherein P₀,P₁,P₂,…,P_m-1Falls on the first pixel point, P_m,P_m+1,…,P_m+n-1Falls on the second pixel, P_m+n,P_m+n+1,…,P_m+n+k-1Falls on the third pixel point, P_m+n+kFalling on the fourth pixel point; wherein P is₀,P_m,P_m+n,P_m+n+kAs valid data points, corresponding S₀,S_m,S_m+n,S_m+n+kIs the valid distance, the rest are invalid data points;

step 3.2: restoring the data points falling in the first pixel point:

geti is 0,1, 2. m-1, the distance of the mark point falling in the first pixel point after reduction from the set coordinate origin is obtained, and then

Where T is the adjacent time interval and T is the adjacent time interval,

S¹ ₀＝S₀，S¹ _m＝S_m，S¹ _m+n＝S_m+n；

step 3.3: and restoring the data points falling in the second pixel point:

geti is m, m +1, m +2, …, m + n-1, the distance between the mark point in the second pixel point after reduction and the coordinate origin is set, then

Wherein

S¹ _m＝S_m，S¹ _m+n＝S_m+n，S¹ _m+n+k＝S_m+n+k；

Step 3.4: repeating the step 3.1 to the step 3.3 to finish the reduction of the data points in each pixel point to obtain the reduced displacement-time data [ s ]¹ ₀，s¹ ₁，…..]；

And 4, step 4: performing data fitting on the reduced displacement-time data to obtain a displacement-time fitting function S ═ f (t) of the mark point, and further obtaining the markThe velocity v ═ df (t)/dt, and the acceleration a ═ d of the point²f(t)/dt²。

Advantageous effects

According to the novel method provided by the invention, high-speed shooting is carried out at a high sampling rate to obtain test information as much as possible in the test process, then displacement-time data of an object automatically acquired by the system is identified and divided according to pixel points, key data points are found, invalid data points are truly restored, polynomial fitting is carried out on the restored displacement-time data, step data and multi-point jumping of first-order and second-order data are eliminated, errors caused by automatic capturing of mark point positions by high-speed shooting are eliminated, and the system precision is improved.

Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.

Drawings

The above and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:

FIG. 1 is a high-speed photography for finding a speed-time image of two adjacent points;

FIG. 2 is a high speed photographic raw displacement-time image;

FIG. 3 is a schematic view of a high-speed photography test site;

FIG. 4 illustrates the position of the mark point at different times of the horizontal movement of the object;

FIG. 5 is a flow chart of a data processing algorithm;

FIG. 6 marks a point displacement-time image;

FIG. 7 labels point velocity versus time images;

FIG. 8 plots point acceleration a1 versus time image;

FIG. 9 marks the point acceleration a 2-time image.

Detailed Description

As shown in fig. 3, before the test, the tripod is properly and stably supported, and then the high-speed camera main body with the lens mounted thereon is fixed on the tripod through the tripod head and is powered on. The camera lens is adjusted to be opposite to the object to be measured. The side surface of the object to be measured, which can be shot by a high-speed camera, is marked, a mark point is made at the position to be researched in the test, a target point is pasted at the position of the mark point, and the computer can automatically track the movement of the mark point and capture and record the coordinate information of the mark point. Two marking points are made at two geometric ends of the side surface of the object which can be shot by the camera, and the actual distance between the two marking points is recorded to be used as the actual distance of the ruler. The object to be measured starts moving, and the camera simultaneously carries out high-speed photography until the whole test is finished.

After the test, a scale is firstly set in a software control interface of the high-speed camera, the starting point and the end point of the scale (namely the two geometric ends of the side surface of the object to be measured which can be shot by the camera) are clicked on a picture to be measured, and the computer can automatically calculate the image distance between the starting point and the end point of the scale. The actual size of the reference scale is entered in the pop-up dialog box and the system will automatically calculate the actual size represented by the current video unit pixel. The computer software can analyze and process the collected images and data, and can automatically track and quickly measure the mark points of the shot object to obtain the position coordinates of the mark points at different moments.

The above processes are all conventional processes in the existing high-speed imaging method.

Because the picture shot by high-speed photography consists of a plurality of pixel points, the position range of each pixel point is determined, and the position coordinate data of the mark points automatically captured by high-speed photography can be divided according to the pixel point area. Because the computer is adopted for automatic capture, the system takes the pixel position of the mark point as the actual position of the mark point, so that when the mark point falls on the same pixel point at different moments, the system reads that the mark point is always at the same position. Therefore, in the method, the coordinate data of the first marking point appearing in each pixel point is removed, and the coordinate data of the other points are regarded as invalid interference data. Therefore, in the method provided by the invention, first data (including corresponding displacement and time) on each pixel point is extracted, other invalid interference data are deleted, then the deleted data points are restored, and the restored data are used as the actual displacement-time of the mark points.

Referring to FIG. 4, for example, the object moves horizontally, and the mark point is at time t₀,t₁,t₂,…,t_m+n+k+j-1Position point P of time₀,P₁,P₂,…,P_m,P_m+1,…,P_m+n-2,P_m+n-1,P_m+n,P_m+n+1,…,P_m+n+k-1,P_m+n+k,P_m+n+k+1,…,P_m+n+k+j-1In which P is₀,P₁,P₂,…,P_m-1Falls on the first pixel point, P_m,P_m+1,…,P_m+n-1Falls on the second pixel, P_m+n,P_m+n+1,…,P_m+n+k-1Falls on the third pixel point, P_m+n+k,P_m+n+k+1,…,P_m+n+k+j-1And the fourth pixel point is dropped.

At successive times t₀，t₁，t₂，t₃,…,t_m+n+k-1Where the adjacent time intervals are the same as T, the interval time used for taking two pictures by high-speed photography. The system automatically captures the horizontal distance S between the position of the mark point and the origin of coordinates₀，S₁，S₂,…,S_m+n+k-1Then, on the first pixel point, remove S₀And the rest is S₁，S₂,…,S_m-1The data is invalid. At the second pixel, remove S_mAnd the rest is S_m+1，S_m+2,…,S_m+n-1The data is invalid. At the third pixel point, remove S_m+nAnd the rest is S_m+n+1，S_m+n+2,…,S_m+n+k-1The data is invalid.

Because the resolution ratio of the existing high-speed photography is very high, the distance between two adjacent pixel points is very small, and the variable-speed motion of the mark point between two continuous pixels can be used as the uniform variable-speed motion for research.

Note S¹ ₀，S¹ ₁,…,S¹ _m+nIs t₀，t₁,…,t_m+nDistance source for mark point of timeDisplacement after dot reduction.

Get S¹ ₀＝S₀，S¹ _m＝S_m，S¹ _m+n＝S_m+nLet us remember t₀Velocity v of the time mark point₁From t₀To t_m+nThe time mark point is used for making horizontal acceleration a₁The uniform variable-speed movement of (1) is that

Can be obtained from the above two formulas

Thus, it is possible to provide

Wherein i is 0,1, 2. cndot. m-1, and the position coordinates of all the mark points on the first pixel point are obtained.

Similarly remember S¹ _m，S¹ _m+1,…,S¹ _m+n+kIs t in the process of uniform variable speed movement_m，t_m+1,…,t_m+n+kAnd (4) horizontally shifting the mark point of the moment from the original point after reduction.

Get S¹ _m＝S_m，S¹ _m+n＝S_m+n，S¹ _m+n+k＝S_m+n+kLet us remember t_mVelocity v of the time mark point₂From t_mTo t_m+n+kThe time mark point is used for making horizontal acceleration a₂Is uniformly variable in speedIn exercise, there are

Can be obtained from the above two formulas

Thus, it is possible to provide

And obtaining the position coordinates of the mark point falling on the second pixel point, wherein i is m, m +1, m +2, … and m + n-1.

According to the method, the position coordinate A [ s ] in the movement process of the mark point can be obtained again¹ ₀，s¹ ₁，…..]And the data is used as the displacement data after the reduction of the marking points, so that errors introduced by using the positions of the pixel points as the position coordinates of the marking points can be eliminated, and the coordinate data of the actual motion process of the object can be reduced. A [ s ]¹ ₀，s¹ ₁，…..]Performing polynomial data fitting to obtain a displacement-time fitting function S ═ f (t) of the mark point, and further obtaining the speed v ═ df (t)/dt of the mark point, and the acceleration a ═ d²f(t)/dt²。

Similarly, the displacement-time, speed-time and acceleration-time data of the mark point after restoration in the vertical direction can also be obtained.

The invention is described below with reference to specific examples:

according to the method provided by the invention, the falling safety test process of a certain type of engine is researched by utilizing high-speed photography, the position of the center of mass on the side of the engine, which is opposite to a high-speed camera, is taken as a mark point, and the time interval T of the high-speed photography is 0.000666667 s. FIG. 6 is an image of displacement data-time restored by the marker points obtained by the original displacement of the marker points automatically captured by the computer and data processing by the method of the present invention.

As can be seen from fig. 6, the result obtained by directly fitting the original data has a larger error than the actual motion of the mark point, and the fitting result of the restored data can better match the motion trajectory of the object. Further, the original data is divided by the time interval by the difference value of the displacement data of two adjacent points to obtain the movement speed V of the mark point₁Meanwhile, the method of the invention is utilized to carry out time first-order derivation on the fitting result of the reduction data to obtain the speed V of the mark point₂FIG. 7 is V₁，V₂Images with time.

As can be seen in FIG. 7, the velocity V found in the raw data₁The object will have multiple data transitions and the velocity of the object will have multiple nulls. And V2 is a smooth curve.

Similarly, V adjacent to the mark point₁The acceleration a of the mark point can be obtained by dividing the phase difference by the time interval through numerical value calculation₁And performing time second-order derivation on the reduced displacement data to obtain the acceleration a of the mark point₂. FIG. 8 and FIG. 9 are a₁、a₂Images with time.

It can further be seen from the acceleration-time diagram that the acceleration values processed using the raw data have already jumped to 10⁴Magnitude, data is not available. The acceleration value of the obtained mark point is very close to the local gravity acceleration by the method provided by the invention, and the absolute value of the acceleration is gradually reduced, which is caused by the air resistance of the object in the air and accords with the actual motion state of the object.

Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made in the above embodiments by those of ordinary skill in the art without departing from the principle and spirit of the present invention.