阅读说明：本技术 一种长曝光倾斜摄影隧道全息测量方法 (Long exposure oblique photography tunnel holographic measurement method ) 是由 张逆进 陈志� 李世文 曾贤平 冯浩雄 于 2019-10-29 设计创作，主要内容包括：本发明公开了一种长曝光倾斜摄影隧道全息测量方法,本发明实施步骤包括在掌子面钻孔放样时标记像控点并测量坐标；通过长曝光摄影采集低照度环境下不同角度、高分辨率、携带岩层纹理的洞身图像；针对洞身图像通过识别多张洞身图像上特征点,以多像空间前方交会法为基础计算特征点空间坐标并展点形成网格表面模型,从而得到携带地层信息、洞身轮廓表面起伏的数字表面模型。本发明具有设备简单、作业便捷、受环境制约小、可全信息还原洞身情况、不影响洞内施工、适用性广泛的优点。(The invention discloses a holographic measurement method for a long exposure oblique photography tunnel, which comprises the implementation steps of marking image control points and measuring coordinates during drilling and lofting on a tunnel face; acquiring hole body images with different angles, high resolution and carrying rock stratum textures in a low-illumination environment by long-exposure photography; the method comprises the steps of identifying characteristic points on a plurality of hole body images aiming at the hole body images, calculating space coordinates of the characteristic points on the basis of a multi-image space forward intersection method, and expanding the points to form a grid surface model, so that a digital surface model carrying stratum information and the surface relief of the hole body outline is obtained. The invention has the advantages of simple equipment, convenient operation, small restriction by environment, capability of restoring the hole condition by full information, no influence on the construction in the hole and wide applicability.)

1. A long exposure oblique photography tunnel holographic measurement method is characterized by comprising the following implementation steps:

1) marking image control points and measuring coordinates during drilling and lofting of a tunnel face;

2) acquiring hole body images carrying rock stratum textures at different angles in a low-illumination environment by long-exposure photography;

3) aiming at the hole body images with different angles and carrying rock stratum textures, a grid surface model is formed by identifying characteristic points on a plurality of hole body images, calculating characteristic point space coordinates and spreading points on the basis of a multi-image space forward intersection method, and accordingly a digital surface model carrying stratum information and hole body contour surface relief is obtained.

2. The tunnel holographic measurement method of claim 1, wherein the step 1) of marking the image control points and measuring the coordinates during the drilling and lofting of the tunnel face further comprises marking the standby points and the check points, wherein the image control points, the standby points and the check points are respectively distributed in the range of the tunnel face and are not collinear, the number of the image control points is not less than 3, the standby points are used as standby points when the image control point points are lost, and the check points are used for checking the accuracy of results and the connection of the survey stations.

3. The long exposure oblique photography tunnel holographic measurement method of claim 1, characterized in that the detailed steps of step 2) include:

2.1) calibrating the camera field of view to determine the minimum shooting distance D_minDetermining that the shooting distance D of the camera is greater than the minimum shooting distance D_min；

2.2) determining all machine positions of the camera, wherein all machine positions of the camera comprise at least three continuous machine positions of a central machine position, a left-side machine position and a right-side machine position;

2.3) selecting one machine position as the current machine position, and installing a camera at the current machine position;

2.4) rechecking the image acquisition area aiming at the current machine position, and skipping to execute the next step if the rechecking is passed; otherwise, skipping to execute the step 2.1) to determine the shooting distance D of the camera again;

2.5) focusing and focus locking are carried out on the camera, and a zooming ring and a focusing ring on a camera lens are not moved after the focus is locked;

2.6) covering the viewfinder window to shield stray light from entering the camera through the viewfinder to interfere the imaging of the camera;

2.7) triggering shutter exposure, and then carrying out soft and scattered light supplement after the camera starts shooting;

2.8) the camera shutter is kept in an open state all the time in a weak illumination environment, so that light reflected by the surface of an object can be continuously accumulated on an imaging element, and bright and clear imaging is obtained in a low illumination environment after the exposure time specified by the shutter opening time;

2.9) judging whether all the machine positions are photographed completely, and if not, skipping to execute the step 2.3); otherwise, skipping to execute the step 3).

4. The long exposure oblique photography tunnel holographic measurement method of claim 3, wherein step 2.3) further comprises using a tripod to ensure the camera is fixed in spatial position when the camera is installed at the front machine position.

5. The long exposure oblique photography tunnel holographic measurement method of claim 3, wherein the step 2.3) further comprises using a remote controller to trigger a shutter when the camera is installed at the current position to avoid imaging blurring caused by vibration caused by manual operation of the camera.

6. The long exposure oblique photography tunnel holographic measurement method of claim 3, characterized in that step 2.1) performs camera field calibration to determine the minimum shooting distance D_minThe detailed steps comprise:

2.1.1) fixing a measuring tape on a wall along the horizontal direction, erecting a camera on a tripod and enabling the camera to be horizontal, wherein the sight line of the camera is vertical to the wall surface; the camera lens is pulled to the shortest focal length and opened to the largest aperture, a picture is taken, the distance from the focal plane of the camera to the wall surface is measured and recorded, the distance from the camera to the wall surface is changed, and the shooting is repeatedly carried out for a plurality of times according to the program;

2.1.2) during data processing, amplifying the photos to find out the scale difference of the two ends of each photo frame, namely the actually measured view field width B, and obtaining at least three groups of data of shooting distance D and view field width B;

2.1.3) fitting an optimal straight line by adopting a least square method to obtain a straight line equation of the field width B and the shooting distance D;

2.1.4) determining a linear equation of the height H of the field of view and the shooting distance D according to the height-width ratio of the camera image;

2.1.5) substituting the maximum width of the tunnel face into a linear equation of the field width B and the shooting distance D to obtain a first shooting distance D_BSubstituting the maximum height of the tunnel face into a linear equation of the view field height H and the shooting distance D to obtain a second shooting distance D_H(ii) a From the first shooting distance D_BSecond shooting distance D_HThe larger of the two is selected as the minimum shooting distance D_min。

7. The long exposure oblique photography tunnel holographic measurement method of claim 3, wherein the step 2.4) of performing image acquisition region review for the current machine position comprises: (1) checking whether the left area and the right area of the palm surface enter the camera view field: irradiating the left boundary position of the palm surface by using a condensing flashlight to enable the palm surface to generate light spots, observing the positions of the light spots through a viewfinder, finely adjusting a holder if the light spots do not exist in the viewfinder, and horizontally rotating a camera until the light spots enter the viewfinder; irradiating the boundary position on the right side of the palm surface by using a condensing flashlight to enable the palm surface to generate light spots, observing the positions of the light spots through a viewfinder, finely adjusting a holder if the light spots do not exist in the viewfinder, and horizontally rotating a camera until the light spots enter the viewfinder; (2) checking whether the upper and lower regions of the palm surface enter the camera view field: illuminating the boundary position on the upper side of the palm surface by using a focusing flashlight, observing the position of a light spot through a viewfinder, and finely adjusting the vertical rotation of a pan-tilt camera until the light spot enters the viewfinder if the light spot does not appear in the viewfinder; and (3) irradiating the boundary position of the lower side of the palm surface by using a condensing flashlight, observing the position of a light spot through a viewfinder, and finely adjusting the vertical rotation of the pan-tilt camera until the light spot enters the viewfinder if the light spot does not appear in the viewfinder.

8. The holographic measurement method for the long-exposure oblique photography tunnel according to claim 3, wherein the step 2.7) of soft-scattering supplementary lighting is to adopt a wide-angle end of a variable-focus flashlight to perform soft-scattering supplementary lighting on surrounding rock, and adjust the flashlight beam to a large-radius aperture with low brightness, so that the illuminated surface is uniformly illuminated.

9. The long exposure oblique photography tunnel holographic measurement method of claim 1, characterized in that the detailed steps of step 3) include:

3.1) inputting the hole body images with different angles and carrying rock stratum textures into photogrammetry and live-action modeling software, and producing and outputting a universal point cloud model;

3.2) aiming at the universal point cloud model, combining two modes of fixed point location coordinate check and random point location vision check to finish model precision check;

3.3) carrying out curved surface generation, section measurement and square measurement on the universal point cloud model after model precision check is finished, and solving the problem of hole body geometric measurement;

and 3.4) carrying out geological model analysis on the universal point cloud model after model precision check is completed to solve the problem of geological distribution.

10. The holographic measurement method for the long exposure oblique photography tunnel according to claim 9, wherein the step 3.2) of completing the model accuracy check by combining the fixed point location coordinate check and the random point location vision check comprises the steps of:

3.2.1) checking the point location marked in the image and having the measured coordinate to investigate the deviation, only the puncture point does not input the coordinate when the puncture point is controlled by the image in the previous period, obtaining the calculated coordinate of the point after the data processing is finished, and evaluating the precision of the universal point cloud model by contrasting the difference value of the calculated coordinate and the actually measured coordinate;

3.2.2) randomly measuring coordinates of a plurality of point locations on the surface of the hole body after the image control point measurement is finished in advance, after the calculation of the universal point cloud model is finished, spreading the point locations into the universal point cloud model, and observing whether the surface separation condition exists in the surface attachment condition of the actually measured point locations and the universal point cloud model so as to investigate the uniformity of error distribution and whether a coarse error exists.

Technical Field

The invention relates to a tunnel section measuring technology, in particular to a long-exposure oblique photography tunnel holographic measuring method.

Background

The tunnel section is usually measured by a total station or a laser scanner, the total station or the laser scanner is used for emitting laser by equipment to measure by a polar coordinate method, the equipment is expensive, only the surface profile data of the tunnel body can be obtained, geological information and blasting parameters cannot be recorded, the laser scattering is serious in smoke and dust and water-rich environments, and the signal return rate is low. The tunnel face geological measurement still stays at the geological sketch or local plane image storage stage, the geological logging subjective influence is large, and the integral traceability of geological conditions is not strong. The 'measuring section for measuring section' essentially ignores the geological property of the tunnel, only pursues the measuring precision of the outline of the tunnel body and ignores the control and balance of geological factors and blasting parameters, and is only passive and expedient in smooth blasting construction.

Disclosure of Invention

The technical problems to be solved by the invention are as follows: aiming at the problems in the prior art, the long exposure oblique photography tunnel holographic measurement method is simple in equipment, convenient and fast to operate, small in environmental restriction, capable of restoring the tunnel body condition through full information, free of influence on in-tunnel construction and wide in applicability.

In order to solve the technical problems, the invention adopts the technical scheme that:

a long exposure oblique photography tunnel holographic measurement method comprises the following implementation steps:

1) marking image control points and measuring coordinates during drilling and lofting of a tunnel face;

2) acquiring hole body images with different angles, high resolution and carrying rock stratum textures in a low-illumination environment by long-exposure photography;

3) aiming at hole body images with different angles, high resolutions and carrying rock stratum textures, a grid surface model is formed by identifying characteristic points on a plurality of hole body images, calculating characteristic point space coordinates and spreading points on the basis of a multi-image space forward intersection method, and accordingly a digital surface model carrying stratum information and hole body contour surface relief is obtained.

Optionally, the step 1) further includes marking backup points and checking points when marking the image control points and measuring coordinates during face drilling and lofting, the image control points, the backup points and the checking points are respectively distributed in the face range and are not collinear, the number of the image control points is not less than 3, the backup points are used as a backup when the image control point points are lost, and the rechecking points are used for checking the accuracy of the result and the connection of the survey station.

Optionally, the detailed steps of step 2) include:

2.1) calibrating the camera field of view to determine the minimum shooting distance D_minDetermining that the shooting distance D of the camera is greater than the minimum shooting distance D_min；

2.2) determining all machine positions of the camera, wherein all machine positions of the camera comprise at least three continuous machine positions of a central machine position, a left-side machine position and a right-side machine position;

2.3) selecting one machine position as the current machine position, and installing a camera at the current machine position;

2.4) rechecking the image acquisition area aiming at the current machine position, and skipping to execute the next step if the rechecking is passed; otherwise, skipping to execute the step 2.1) to determine the shooting distance D of the camera again;

2.5) focusing and focus locking are carried out on the camera, and a zooming ring and a focusing ring on a camera lens are not moved after the focus is locked;

2.6) covering the viewfinder window to shield stray light from entering the camera through the viewfinder to interfere the imaging of the camera;

2.7) triggering shutter exposure, and then carrying out soft and scattered light supplement after the camera starts shooting;

2.8) the camera shutter is kept in an open state all the time in a weak illumination environment, so that light reflected by the surface of an object can be continuously accumulated on an imaging element, and bright and clear imaging is obtained in a low illumination environment after the exposure time specified by the shutter opening time;

2.9) judging whether all the machine positions are photographed completely, and if not, skipping to execute the step 2.3); otherwise, skipping to execute the step 3).

Optionally, the step 2.3) further includes using a tripod to ensure the camera is fixed in the spatial position when the camera is installed at the current position.

Optionally, the step 2.3) further includes triggering a shutter by using a remote controller when the camera is installed at the current position to avoid imaging blur caused by vibration caused by manually operating the camera.

Optionally, step 2.1) performs camera field calibration to determine the minimum shooting distance D_minThe detailed steps comprise:

2.1.1) fixing a measuring tape on a wall along the horizontal direction, erecting a camera on a tripod and enabling the camera to be horizontal, wherein the sight line of the camera is vertical to the wall surface; the camera lens is pulled to the shortest focal length and opened to the largest aperture, a picture is taken, the distance from the focal plane of the camera to the wall surface is measured and recorded, the distance from the camera to the wall surface is changed, and the shooting is repeatedly carried out for a plurality of times according to the program;

2.1.2) during data processing, amplifying the photos to find out the scale difference of the two ends of each photo frame, namely the actually measured view field width B, and obtaining at least three groups of data of shooting distance D and view field width B;

2.1.3) fitting an optimal straight line by adopting a least square method to obtain a straight line equation of the field width B and the shooting distance D;

2.1.4) determining a linear equation of the height H of the field of view and the shooting distance D according to the height-width ratio of the camera image;

2.1.5) substituting the maximum width of the tunnel face into a linear equation of the field width B and the shooting distance D to obtain a first shooting distance D_BSubstituting the maximum height of the tunnel face into a linear equation of the view field height H and the shooting distance D to obtain a second shooting distance D_H(ii) a From the first shooting distance D_BSecond shooting distance D_HThe larger of the two is selected as the minimum shooting distance D_min。

Optionally, the step 2.4) of performing image acquisition region review on the current machine position includes: (1) checking whether the left area and the right area of the palm surface enter the camera view field: irradiating the left boundary position of the palm surface by using a condensing flashlight to enable the palm surface to generate light spots, observing the positions of the light spots through a viewfinder, finely adjusting a holder if the light spots do not exist in the viewfinder, and horizontally rotating a camera until the light spots enter the viewfinder; irradiating the boundary position on the right side of the palm surface by using a condensing flashlight to enable the palm surface to generate light spots, observing the positions of the light spots through a viewfinder, finely adjusting a holder if the light spots do not exist in the viewfinder, and horizontally rotating a camera until the light spots enter the viewfinder; (2) checking whether the upper and lower regions of the palm surface enter the camera view field: illuminating the boundary position on the upper side of the palm surface by using a focusing flashlight, observing the position of a light spot through a viewfinder, and finely adjusting the vertical rotation of a pan-tilt camera until the light spot enters the viewfinder if the light spot does not appear in the viewfinder; and (3) irradiating the boundary position of the lower side of the palm surface by using a condensing flashlight, observing the position of a light spot through a viewfinder, and finely adjusting the vertical rotation of the pan-tilt camera until the light spot enters the viewfinder if the light spot does not appear in the viewfinder.

Optionally, the step 2.7) of performing soft-scattering supplementary lighting specifically includes performing soft-scattering supplementary lighting on the surrounding rock by using a wide-angle end of a variable-focus flashlight, and adjusting a flashlight beam to a large-radius aperture with low brightness, so that the illuminated surface is uniformly illuminated.

Optionally, the detailed steps of step 3) include:

3.1) inputting the hole body images with different angles, high resolution and rock texture into photogrammetry and live-action modeling software, and producing and outputting a universal point cloud model;

3.2) aiming at the universal point cloud model, combining two modes of fixed point location coordinate check and random point location vision check to finish model precision check;

3.3) carrying out curved surface generation, section measurement and square measurement on the universal point cloud model after model precision check is finished, and solving the problem of hole body geometric measurement;

and 3.4) carrying out geological model analysis on the universal point cloud model after model precision check is completed to solve the problem of geological distribution.

Optionally, the step 3.2) of completing model accuracy check by combining two modes of fixed point location coordinate check and random point location vision check includes:

3.2.1) checking the point location marked in the image and having the measured coordinate to investigate the deviation, only the puncture point does not input the coordinate when the puncture point is controlled by the image in the previous period, obtaining the calculated coordinate of the point after the data processing is finished, and evaluating the precision of the universal point cloud model by contrasting the difference value of the calculated coordinate and the actually measured coordinate;

3.2.2) randomly measuring coordinates of a plurality of point locations on the surface of the hole body after the image control point measurement is finished in advance, after the calculation of the universal point cloud model is finished, spreading the point locations into the universal point cloud model, and observing whether the surface separation condition exists in the surface attachment condition of the actually measured point locations and the universal point cloud model so as to investigate the uniformity of error distribution and whether a coarse error exists.

Compared with the prior art, the invention has the following advantages:

1. the equipment is simple. Only one common single lens reflex camera and a tripod are needed to complete field data acquisition.

2. The operation is convenient. The coordinates of the station of the equipment do not need to be measured, and the equipment can be used along with the frame. The single image acquisition time is only 2 minutes, and only 4 images are acquired for 20 m continuous sections of the tunnel.

3. Is less restricted by the environment. The long exposure photography technology can ensure that the clear acquisition of images can still be realized under the condition of insufficient tunnel illumination and visibility.

4. The hole body condition can be restored through full information. The achievement of the method comprises not only a tunnel body outline digital surface model, but also all visible information such as the occurrence, color and luster, water richness and blasting hole arrangement of the surrounding rock, and the parameters of the model can be accurately measured, so that the identification and prejudgment of the surrounding rock in front and the rechecking and adjustment of blasting parameters are realized, which cannot be realized by the existing measuring technology.

5. The construction in the hole is not affected. Due to the long exposure imaging characteristic, even if the camera is shielded for a short time in the imaging process, the image acquisition is not influenced, so that the traffic is not required to be interrupted during measurement, and the compatibility with construction operation is good.

6. The applicability is wide. The invention can be used for the following operation contents in the excavation construction of tunnel engineering: measuring continuous sections of tunnel cave sections and accounting concrete pouring square amount before lining; the method comprises the following steps of (1) carrying out layout recording and parameter optimization on blasting hole sites on a tunnel face before tunnel blasting; the tunnel boring section reveals surrounding rock information recording and forward geological inference.

Drawings

FIG. 1 is a schematic diagram of a basic flow of a method according to an embodiment of the present invention.

FIG. 2 is a diagram illustrating the labeling of the image control point, the spare point and the review point according to an embodiment of the present invention.

Fig. 3 is a schematic diagram of coordinate acquisition information of image control points in the embodiment of the present invention.

Fig. 4 shows the contrast of the images with different exposure times under the same illumination.

Fig. 5 is a schematic view of the field coverage of the camera.

Fig. 6 is a schematic diagram of a camera and lens field calibration process according to an embodiment of the invention.

Fig. 7 is a schematic diagram of an arrangement of image capturing positions of a camera according to an embodiment of the present invention.

Fig. 8 is a schematic diagram of the left (right) margin check of the tunnel face and the camera viewfinder window in the embodiment of the invention, wherein the left side is the schematic diagram of the left (right) margin check of the tunnel face, and the right side is the schematic diagram of the camera viewfinder window.

Fig. 9 is a schematic diagram of upper (lower) margin checking of a tunnel face and a camera viewfinder window in the embodiment of the invention, wherein the left side is the schematic diagram of upper (lower) margin checking of the tunnel face, and the right side is the schematic diagram of the camera viewfinder window.

FIG. 10 is a schematic diagram of focusing on a spot in an auto-focus mode according to an embodiment of the present invention, wherein a view finder is shown on the left side.

FIG. 11 is a schematic diagram of the light view (left) and stray light interference (right) of the lens according to the embodiment of the present invention.

Fig. 12 is a schematic diagram of the focusing (left) and light scattering (right) effects of the variable-focus flashlight in the embodiment of the invention.

Fig. 13 is a schematic diagram showing the comparison of the imaging effects of direct exposure (left) and soft and diffuse fill light (right).

FIG. 14 is a plot of point cloud data from an Agisosoft PhotoSacan example of the present invention.

Fig. 15 is a schematic diagram of image-controlled coordinate entry and point location specification checking according to an embodiment of the present invention.

Fig. 16 is a schematic diagram of coordinates of check points on a model according to an embodiment of the present invention.

FIG. 17 is a diagram illustrating coordinates of check points on a model according to an embodiment of the present invention.

FIG. 18 is a hole body contour under run distribution analyzed by Geomagic Studio in accordance with an embodiment of the present invention.

FIG. 19 is an example of an editable cross-sectional representation of AutoCAD derived by Geomagic Studio in accordance with an embodiment of the present invention.

FIG. 20 is a secondary lining thickness distribution analyzed by Geomalic Studio in the example of the present invention.

FIG. 21 shows the actual measurement of the arrangement of the blastholes based on the orthographic image of the palm surface of Agisoft PhotoSacan in the embodiment of the present invention.

FIG. 22 is a cross-sectional view of a tunnel face joint distribution based on an Agisoft PhotoSacan ortho-image according to an embodiment of the present invention.

FIG. 23 is a joint spatial distribution estimated based on the Civli 3D geological model in an embodiment of the present invention.

FIG. 24 is a sagittal polar projection of the palm surface and the joint.

Fig. 25 is a schematic flow chart of the embodiment of the present invention applied to tunneling construction.

Detailed Description

The tunnel holographic measurement method of the long exposure oblique photography of the invention will be further described in detail below by taking the tunnel river rock horizontal tunnel of Xinhuashan tunnel of Zhangji Huai railway as an example.

As shown in fig. 1, the implementation steps of the tunnel holographic measurement method for long exposure oblique photography in this embodiment include:

1) marking image control points and measuring coordinates during drilling and lofting of a tunnel face;

2) acquiring hole body images with different angles, high resolution and carrying rock stratum textures in a low-illumination environment by long-exposure photography;

3) aiming at hole body images with different angles, high resolutions and carrying rock stratum textures, a grid surface model is formed by identifying characteristic points on a plurality of hole body images, calculating characteristic point space coordinates and spreading points on the basis of a multi-image space forward intersection method, and accordingly a digital surface model carrying stratum information and hole body contour surface relief is obtained.

This embodiment still includes mark spare point, check point when mark image control point and measurement coordinate in 1) face drilling lofting, just image control point, spare point, check point distribute respectively in face scope and not collinear, and the quantity of image control point is no less than 3, and the spare point is used for as reserve when image control point location is lost, the point of checking again is used for checking the achievement precision and the survey station is plugged into. The image collected by the camera can be reconstructed without external reference coordinates, the model can only ensure relative precision and cannot guide actual construction, therefore, a plurality of control points need to be arranged on the body of the hole in the early stage, and the point positions on the image are pierced in the later stage, so that the results are unified into a coordinate system used for construction. The number of the image control points is not less than 3, and equipment points are added according to the conditions in the tunnel, so that point position loss is prevented. In addition, a plurality of check points are randomly measured on the side wall of the hole body for checking the result precision and the test station connection. The image control points are shown in fig. 2. The marking and measurement of the image control point, the standby point and the check point can be completed together when the peripheral eyes of the palm surface are lofted. The point locations are flexibly arranged without strict requirements, and only need to be distributed in the range of the tunnel face and not be collinear, and the image control coordinates recorded by the electronic handbook are shown in figure 3.

The basic principle of the method of the embodiment is as follows: firstly, hole body images with different angles, high resolution and rich rock texture are collected under a low-illumination environment through a long-exposure photography technology, then a digital model is reconstructed from the collected photos by utilizing an oblique photogrammetry technology, and a digital surface model carrying formation information and hole body contour surface relief is further formed.

The biggest difficulty in acquiring high-resolution images of a tunnel body in the tunnel is that the illumination in the tunnel is serious and insufficient (for example, when an image of a tunnel face is acquired before blasting, the tunnel face is painted black), and the quality of the acquired image directly influences the accuracy of a digital surface model. If the flash lamp is used for light supplement, the air in the hole is rich in suspended particles, so that the flash light is seriously scattered, and the flash lamp in the gas tunnel is forbidden to be used, so that the conventional photography technology cannot successfully acquire images. In the long exposure photography technique, in a weak light environment, a camera shutter is kept in an open state (such as a B-gate mode in a single lens reflex), light reflected by the surface of an object is accumulated on an imaging element continuously, bright and clear imaging can be obtained in a low-illumination environment as long as the relative position between the camera and a photographed object is kept fixed (a tripod is used for stabilizing the camera), and the shutter opening time is long enough (exposure time), fig. 4 shows the difference of images obtained by adopting different exposure times under the same illumination condition, wherein the exposure time in fig. 4(a) is 5s, and the exposure time in fig. 4(B) is 60 s.

The basic principle of establishing the digital surface model of oblique photography is as follows: shooting the reconstructed object from different angles to obtain images of the object from different angles, completing the identification of the characteristic points on a plurality of images by a computer graphic system, and calculating the space coordinates of the characteristic points and expanding the points on the basis of a multi-image space forward intersection method to form a grid surface model. Therefore, the higher the resolution of the acquired image is, the richer the details are, the more the computer identifies the feature points, and the higher the achievement precision is. The method works in the tunnel, the coordinates of the camera station are unknown, satellite positioning is not available, and the attitude parameters of the camera cannot be measured, so that the condition for solving the coordinates of the feature points is much harsher than that for oblique photography measurement in an open-air environment.

In this embodiment, the detailed steps of step 2) include:

2.1) calibrating the camera field of view to determine the minimum shooting distance D_minDetermining that the shooting distance D of the camera is greater than the minimum shooting distance D_min；

2.2) determining all machine positions of the camera, wherein all machine positions of the camera comprise at least three continuous machine positions of a central machine position, a left-side machine position and a right-side machine position;

2.3) selecting one machine position as the current machine position, and installing a camera at the current machine position;

2.4) rechecking the image acquisition area aiming at the current machine position, and skipping to execute the next step if the rechecking is passed; otherwise, skipping to execute the step 2.1) to determine the shooting distance D of the camera again;

2.5) focusing and focus locking are carried out on the camera, and a zooming ring and a focusing ring on a camera lens are not moved after the focus is locked;

2.6) covering the viewfinder window to shield stray light from entering the camera through the viewfinder to interfere the imaging of the camera;

2.7) triggering shutter exposure, and then carrying out soft and scattered light supplement after the camera starts shooting;

2.8) the camera shutter is kept in an open state all the time in a weak illumination environment, so that light reflected by the surface of an object can be continuously accumulated on an imaging element, and bright and clear imaging is obtained in a low illumination environment after the exposure time specified by the shutter opening time;

2.9) judging whether all the machine positions are photographed completely, and if not, skipping to execute the step 2.3); otherwise, skipping to execute the step 3).

When shooting is carried out in the tunnel, the illumination is extremely low, a shot picture range cannot be observed through a viewfinder, the image acquisition is easy to be incomplete when a camera is too close to the tunnel face, and the image acquisition is easy to be unclear when the camera is too far away, as shown in fig. 5 (a); the distance between the camera view field and the face is preferably just enough to cover the face, as shown in fig. 5 (b).

In order to shorten the preparation time, the shooting distance meeting the full coverage of the field of view is predetermined, the shooting range of the camera lens is measured outside the hole, the relationship between the field of view width and the shooting distance is determined, and the process is the camera field of view calibration. Fig. 6 shows the field calibration process and the results of the camera and the corresponding lens. In this embodiment, step 2.1) performs camera field calibration to determine the minimum shooting distance D_minThe detailed steps comprise:

2.1.1) fixing the measuring tape on a wall along the horizontal direction, erecting a camera on a tripod and enabling the camera to be horizontal, wherein the sight line of the camera is vertical to the wall surface, as shown in fig. 6 (a); the camera lens is pulled to the shortest focal length (18mm), opened to the largest aperture (F3.5), a picture is taken, the distance from the camera focal plane to the wall surface is measured and recorded, as shown in fig. 6(b) and 6(c), the distance from the camera to the wall surface is changed, and the shooting is repeated for a plurality of times (at least 3 times) according to the program;

2.1.2) during data processing, amplifying the photos to find out the scale difference of the two ends of each photo frame, namely the actually measured view field width B, and obtaining at least three groups of data of shooting distance D and view field width B;

2.1.3) fitting an optimal straight line by adopting a least square method to obtain a straight line equation of the field width B and the shooting distance D;

2.1.4) determining a linear equation of the height H of the field of view and the shooting distance D according to the height-width ratio of the camera image;

2.1.5) substituting the maximum width of the tunnel face into a linear equation of the field width B and the shooting distance D to obtain a first shooting distance D_BSubstituting the maximum height of the tunnel face into a linear equation of the view field height H and the shooting distance D to obtain a second shooting distance D_H(ii) a From the first shooting distance D_BSecond shooting distance D_HThe larger of the two is selected as the minimum shooting distance D_min。

In this embodiment, an optimal straight line is fitted by using a least square method, and a straight line equation for obtaining the field width B and the shooting distance D is shown in fig. 6 (D). From the calibration equation, the relationship between the distance D (mm) of the camera from the tunnel face and the width B (mm) of the camera view field is: b-1.2773. D_B90.121, if the aspect ratio of the camera imaging is 2:3, the relationship between the field height H and D is: H2B/3 0.8515. D_H-60.081. If the maximum width B of the palm surface is 8m, H is 6m, then D can be obtained according to the above-mentioned calibration equation_B＝6.3m，D_H7.1m, minimum shot distance D_minGet D_BAnd D_HThe large value (7.1m) of the camera is about 7.1m away from the palm surface, so the camera frame is arranged 7.5 m behind the palm surface when shooting.

As shown in fig. 7, in step 2.2), all positions of the camera are determined, where all positions of the camera include at least three consecutive positions, i.e., a central position, a left-side position, and a right-side position, and the position supplementary shooting can be automatically added when necessary. The orientation of the camera should be adjusted every time the arrangement is carried out, so that the images shot by all the positions can completely cover the tunnel face.

Absolute stability needs to be ensured in the long exposure shooting process, and blurring can be caused by vibration in the shooting process. In this embodiment, in step 2.3), when the camera is installed at the current position, a tripod is further adopted to ensure that the camera is fixed at the spatial position. In addition, in the step 2.3), when the camera is installed at the current machine position, a remote controller is adopted to trigger a shutter, so that imaging blurring caused by vibration caused by manual operation of the camera is avoided.

In this embodiment, a canon EOS-500D camera is specifically adopted as the camera, and by taking canon EOS-500D camera parameter settings as an example, exposure parameters after installation in place are set as follows (other camera parameter settings are similar):

adjusting a camera shutter to a BULB mode; the focal length is adjusted to a minimum (18mm in this example); the aperture is set to maximum (F3.5 in this example); the sensitivity ISO is preferably set to 3200 (3200 is preferable for an APS-C frame single lens reflex camera, 6400 for a middle frame camera, 12800 for a full frame single lens reflex camera), the white balance is set to an automatic mode (AWB), and the focusing mode is preset to an artificial intelligent SERVO focusing mode (AI SERVO).

The illumination in the tunnel is extremely low, and the shooting area cannot be directly observed through a camera viewfinder. In order to avoid incomplete image acquisition of the tunnel face caused by the deviation of the shooting area, image acquisition area rechecking in the step 2.4) is required before shooting.

In this embodiment, the step 2.4) of performing image acquisition area review on the current machine position includes:

(1) checking whether the left area and the right area of the palm surface enter the camera view field: referring to fig. 8, a spot flashlight is adopted to irradiate the boundary position of the left side of the palm surface, so that the palm surface can generate a spot, the spot position can be observed through the viewfinder, if the spot does not exist in the viewfinder, the pan-tilt is finely adjusted, and the camera horizontally rotates until the spot enters the viewfinder; irradiating the boundary position on the right side of the palm surface by using a condensing flashlight to enable the palm surface to generate light spots, observing the positions of the light spots through a viewfinder, finely adjusting a holder if the light spots do not exist in the viewfinder, and horizontally rotating a camera until the light spots enter the viewfinder;

(2) checking whether the upper and lower regions of the palm surface enter the camera view field: referring to fig. 9, a spotlight flashlight is adopted to irradiate the boundary position on the upper side of the palm surface, the position of a light spot is observed through a viewfinder, and if the light spot does not appear in the viewfinder, the pan-tilt camera is finely adjusted to vertically rotate until the light spot enters the viewfinder; and (3) irradiating the boundary position of the lower side of the palm surface by using a condensing flashlight, observing the position of a light spot through a viewfinder, and finely adjusting the vertical rotation of the pan-tilt camera until the light spot enters the viewfinder if the light spot does not appear in the viewfinder.

The single lens reflex focusing process has the following characteristics that ① in an automatic focusing mode (AF), the camera executes a focusing program firstly, the shutter cannot be triggered to shoot when the focusing is not finished, ② in a manual focusing Mode (MF), the shutter can be directly triggered no matter whether the focusing is successful or not under the current camera focus setting, the automatic focusing cannot work normally due to insufficient light in a tunnel, the manual focusing cannot judge the focusing condition by naked eyes, the focusing failure can cause the shooting picture to be fuzzy, and the collected image cannot be used for data post-processing, so in the embodiment, step 2.5) focuses and locks the focus on the camera, and particularly, ① in the preset artificial intelligent SERVO focusing mode (AI SERVO), confirms that a focusing mode button on the camera lens is in an AF (automatic focusing) mode state, as shown in the left image of FIG. 10, ② irradiates a focusing flashlight on a palm surface to enable light spots to cover one of 9 focuses (a white box in the right image of FIG. 10) visible in a viewfinder, a semi-pressing a shutter indicator lamp, changing into a green indicator lamp, and changing to prevent the focusing to change when the focusing mode is changed to the focusing and the focusing is changed into a focusing ring, as shown in the left focusing ring ③, the focusing mode, the focusing is changed, and the focusing is completed when the focusing is changed.

The viewfinder of the camera is reflected to human eyes from the reflector through the built-in eye flat pentaprism to observe the lens framing condition (as shown on the left of figure 11). When the single lens reflex camera is used for long-exposure shooting, the reflector is lifted (as shown on the right of figure 11). During shooting, although a reflecting plate is arranged between the horizontal pentaprism and the sensor for blocking, the horizontal pentaprism cannot be completely sealed, external stray light (although a shooting environment is dark, long exposure is extremely sensitive to light, a screen is always opened and cannot be closed during exposure, and the screen becomes a main stray light source) can still be reflected to the sensor through a gap between the horizontal pentaprism and the sensor (as shown in the right side of fig. 11), interference is caused to a shot image, and the situation is particularly obvious in long-exposure shooting. Therefore, before the camera shutter is triggered, the viewfinder window needs to be covered to shield stray light from entering the camera through the viewfinder to interfere with the imaging of the camera. Step 2.6) in this embodiment carries out the closing cap to the view finder window and gets into when camera internal disturbance camera formation of image through the view finder with shielding stray light, and the view finder closing cap can be got by oneself and view finder window size is got, adopts rubber cutting preparation, simple and practical.

The step 2.7) of triggering shutter exposure in this embodiment includes ① switching the shooting mode button on the remote controller emitter to BULB state ②. exposure time estimation the light metering system built in the camera cannot calculate the correct exposure time for extreme light conditions in the tunnel and determining the exposure time based on the visual environment illumination and the exposure table, not only depending on experience and time, so this embodiment uses the photography light meter to calculate the exposure time, the light metering table is set up according to the aperture and light sensing of the camera, which can calculate the correct exposure time rapidly according to the current environment light ③. away from the camera, pressing the shutter trigger button of the remote controller, the camera enters the long exposure shooting state, the lower right corner of the screen will display the exposed time, when the time on the camera screen reaches the exposure time calculated by the light metering table, pressing the shutter key of the remote controller to stop the exposure.

After the tunnel is excavated, the surface of the tunnel body is uneven, long exposure can only clearly record details of the light receiving surface, and obvious shadow cast can appear on the backlight surface. Details of shadow areas are lost, broken holes can appear in the digital surface model, and information recording is incomplete. By adopting the wide-angle end of the variable-focus flashlight to carry out soft and scattered light supplement on surrounding rocks, the problems of uneven light receiving surface and shadow area can be solved. In this embodiment, the step 2.7) of performing soft-scattering supplementary lighting specifically means performing soft-scattering supplementary lighting on surrounding rock by using a wide-angle end of a variable-focus flashlight, and adjusting a flashlight beam to a large-radius aperture with low brightness, so that a light receiving surface irradiates uniformly. The variable-focus flashlight is characterized in that the focusing and scattering of the projection light spots can be adjusted: when focusing operation is carried out, the flashlight beam is adjusted to be a small-radius light spot, the brightness is high, and the irradiation point is concentrated (as shown on the left side of FIG. 12); when the soft and scattering light supplement operation is performed, the flashlight beam is adjusted to be a large-radius aperture, the brightness is low, and the light receiving surface is uniformly irradiated (as shown on the right side of fig. 12). The soft and diffuse light supplement is performed after the camera starts shooting. Firstly, a flashlight lamp head is arranged at a wide-angle end, the brightness of the lamp head is adjusted to be the lowest, a flashlight is turned on, the flashlight is upward and faces the direction of a hole, the surface of the hole body is quickly and uniformly swept within the exposure time according to the exposure time estimated by a photometer, so that the backlight surface of surrounding rock of the hole body uniformly receives light (in the light supplementing process, the flashlight cannot be directed towards a camera lens at any time), and the light supplementing effect is shown in fig. 13.

According to the photo collecting method, the image is collected by advancing at regular intervals, and the number of the collected images is about 5-6. Considering the imaging quality, an effective model of 10 meters can be generated on the premise of ensuring the model precision of the image acquired each time, the cyclic footage of the full section is controlled according to 2.5-3 meters, the section is acquired once by 3-4 cyclic footages at most, and the encryption is carried out if necessary.

In this embodiment, the detailed steps of step 3) include:

3.1) inputting the hole body images with different angles, high resolution and rock texture into photogrammetry and live-action modeling software, and producing and outputting a universal point cloud model; the photogrammetry and live-action modeling software of the embodiment specifically adopts the agisoft photoscan software to process the acquired original image, and produces and outputs a general point cloud model (LAS data exchange format) for point cloud processing and data analysis, as shown in fig. 14.

3.2) aiming at the universal point cloud model, combining two modes of fixed point location coordinate check and random point location vision check to finish model precision check;

3.3) carrying out curved surface generation, section measurement and square measurement on the universal point cloud model after model precision check is finished, and solving the problem of hole body geometric measurement;

and 3.4) carrying out geological model analysis on the universal point cloud model after model precision check is completed to solve the problem of geological distribution.

In this embodiment, the step 3.2) of completing model accuracy check by combining two modes of fixed point location coordinate check and random point location vision check includes:

3.2.1) checking the marked point location with measured coordinate in the image to examine the deviation, and only the puncture point does not input the coordinate when the image controls the puncture point in the previous period, after the data processing is finished, the calculated coordinate of the point is obtained, and the precision of the universal point cloud model is evaluated by contrasting the difference value of the calculated coordinate and the measured coordinate, the checking method aims at carrying out error quantitative evaluation on the key position, shown by figure 15, ① - ⑦ points are distributed on site, and the coordinates are collected, wherein ①, ④ and ⑤ points record the measured coordinate for correcting the coordinate system when the puncture point, ②, ③, ⑥ and ⑦ points only puncture the point on the picture, and do not record the coordinate, after the model is generated, the coordinate value on the accounting model is calculated, and compared with the measured coordinate, the error between the designated point on the model and the actual position of the point can be obtained, and the calculation result is shown in figure 16 and table 1.

Table 1 model and actual measurement fixed point coordinate error comparison table.

As can be seen from Table 1, the maximum error value of the model at the image control points (point Nos. ①, ④ and ⑤) of the set coordinates is 9mm, which is derived from the identification deviation at the time of the puncture point.

3.2.2) randomly measuring coordinates of a plurality of point locations on the surface of the hole body after the image control point measurement is finished in advance, after the calculation of the universal point cloud model is finished, spreading the point locations into the universal point cloud model, and observing whether the surface separation condition exists in the surface attachment condition of the actually measured point locations and the universal point cloud model so as to investigate the uniformity of error distribution and whether a coarse error exists. The checking method aims at visually evaluating the overall accuracy of the model. In FIG. 17, the field random measuring points are the sphere in the graph and are plotted according to the field actual measurement coordinates; the gray color is the surface model generated by the method of the embodiment, the spread points are all located on the model curved surface (in the semi-embedded curved surface), and no obvious measuring point is separated from the surface of the model, which indicates that the overall deviation distribution of the model is relatively uniform.

In this embodiment, the step 3.3) of performing surface generation, section measurement and square measurement on the universal point cloud model after completing model accuracy check to solve the problem of hole body geometric measurement mainly includes:

3.3.1) section measurement:

in this embodiment, a curved surface model generated by using geogenic Studio software is introduced into a reference standard excavation section model, and an advanced excavation distribution cloud map of an actual excavation profile can be obtained, as shown in fig. 18 (for convenience of display, any section of a hole body measurement result is intercepted, the section length is 632mm, and the illustration is adopted in this embodiment), different colors on the hole body excavation profile represent different underexcavation values, and a statistical table in the lower right corner of the figure shows that the maximum overbreak amount is 227mm (appearing in a vault); the maximum underdigging amount is 580mm (the maximum underdigging amount is found in an inverted arch because the inverted arch is not excavated), and visual and accurate reference is provided for the overall overbreak condition of the hole body. By utilizing the surface model generated by the Geomagic Studio software, the position of the cross section can be arbitrarily appointed to intercept the cross section of the hole body and export an editable CAD graph, as shown in FIG. 19.

3.3.2) square measurement:

in this embodiment, a cloud map of actual lining thickness distribution can be obtained by using a tunnel body curved surface model generated by geogenic Studio software and importing a standard secondary lining section model in contrast, as shown in fig. 20. The thickness of the originally designed secondary lining is 300mm, and the design pouring square amount of the measuring section is 3.56m³/m×0.632m＝2.25m³. Actually measuring the maximum lining thickness of 500mm (appearing on the vault) and the minimum lining thickness of-289 mm (appearing on the inverted arch because the inverted arch is not excavated), statistically displaying the linear average thickness of 260mm of the secondary lining in the measuring section, and calculating the pouring square amount of 3.05m of the measuring section according to the designed profile of the secondary lining, wherein the outward expansion of 260mm³/m×0.632m＝1.93m³。

3.3.3) measurement of geological properties of the hole body:

in the embodiment, the geological model processing analysis is performed by adopting AutoCAD Civil 3D software, so that the geological distribution problem is solved, and the method mainly comprises the following aspects:

① blasting parameter verification

And (3) performing perspective and inclination correction on the picture by using a point cloud model based on Agisoft PhotoSacn software to obtain measurable tunnel surface orthographic projection. The arrangement positions of the blast holes on the tunnel face are clearly recorded by the high-resolution camera, so that the arrangement number and the arrangement positions of the blast holes at the time can be rapidly and accurately measured in the later period, the arrangement of the blast holes on the tunnel face orthoscopic image is shown in a figure 21, and the parameter analysis of the peripheral blast holes based on the image is shown in a table 2.

Table 2: and (5) actually measuring the peripheral eye parameter analysis table.

As shown in fig. 21 and table 2, the average pitch of the holes is 569mm, the coefficient of variation is 0.466, the coefficient of variation is too large (preferably, the coefficient of variation of the peripheral hole pitch is not more than 0.3), and the uniformity of the hole pitch distribution is not good. The real and quantitative blasting parameters are mastered, and the next circular drilling and blasting construction parameters can be accurately adjusted by combining the surrounding rock conditions.

② geological check and attitude estimation

The face orthoscopic image produced by the Agisoft PhotoSacn software can be used for performing the exhibition and the shape estimation on the face joint distribution. The tunnel face joint information is recorded in the point cloud model, and each point in the point cloud model carries coordinate information and object color information, so that the tunnel face orthoimage obtained by the model not only contains joint conventional plane information (such as color and hole position relation), but also contains joint spatial position information (such as joint occurrence), and the tunnel face front geological information can be conveniently estimated. The tunnel face joint information recorded by the ortho image is shown in fig. 22, and the spatial geological model and the estimated joint occurrence are shown in fig. 23. Different from the conventional joint surface, the method of the embodiment scans the sections one by one to obtain the distribution of the joint surface, so that the method is a curved surface, has more detailed characteristics and can intuitively and accurately estimate the stratum relief condition. The distribution of the sagittal polar projection of the facet joints can also be obtained from the spatial distribution of the joints, as shown in fig. 24.

As shown in fig. 25, when the method of the present embodiment is applied to tunneling construction, basic steps of the tunneling construction include:

s1) marking image control points and measuring coordinates during drilling and lofting of the tunnel face by adopting the step 1) of the method;

s2) withdrawing the excavation rack before blasting;

s3) acquiring hole body images with different angles, high resolution and carrying rock stratum textures in a low-illumination environment by long-exposure photography in the step 2) of the method;

s4), on one hand, blasting, mucking and supporting operation in the tunnel are carried out; on the other hand, step 3) of the method is adopted to calculate the space coordinates of the characteristic points and spread the points to form a grid surface model by identifying the characteristic points on a plurality of hole body images and on the basis of a multi-image space forward intersection method, so that a digital surface model carrying stratum information and surface relief of hole body contours is obtained, the rendering of surrounding rock textures is completed, then the blasted hole body contours are analyzed on the basis of the digital surface model, and the information points of the rock stratum layer are circularly revealed in the current round on the basis of the rechecking address condition and blasting parameters of the texture model; blasting parameter adjustment, undermining treatment and lining square amount statistics can be carried out according to requirements;

s5) comparing with the previous cycle stratum information point, deducing the next cycle stratum characteristic, and executing the step S3); and finally, establishing the stratum distribution information of the whole tunnel until tunneling is completed.

The above description is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments, and all technical solutions belonging to the idea of the present invention belong to the protection scope of the present invention. It should be noted that modifications and embellishments within the scope of the invention may occur to those skilled in the art without departing from the principle of the invention, and are considered to be within the scope of the invention.