Tunnel three-dimensional point cloud acquisition system, structured light exposure adjusting system and method

Abstract

The invention belongs to the technical field of three-dimensional point cloud acquisition, and particularly relates to a tunnel three-dimensional point cloud acquisition system and a structured light exposure adjusting system and method. The adjusting system comprises an image processing module and a controller. The image processing module obtains the proportion information of the number of pixel points in each interval in the structured light grayscale image; the interval is at least divided into two according to the gray level; the controller adjusts the exposure according to the proportion information until the quality of the obtained structured light image is qualified: if the proportion of the number of the pixel points in the low gray level interval is larger than a set threshold value and the proportion of the number of the pixel points in the high gray level interval is smaller than the set threshold value, the exposure is increased, and shooting is carried out again; and if the proportion of the number of the pixel points in the high gray level interval is larger than a set threshold value and the proportion of the number of the pixel points in the low gray level interval is smaller than the set threshold value, the exposure is reduced, and shooting is carried out again. According to the invention, the problem that the image quality and the three-dimensional point cloud generation quality are influenced by insufficient exposure or overexposure of the projection structured light is solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of three-dimensional point cloud acquisition, and in particular relates to a tunnel three-dimensional point cloud acquisition system, a structured light exposure adjustment system and a method. Background Art

[0002] In the intelligent construction of the tunnel primary lining, it is necessary to detect the excavation surface and calculate the amount of grouting. In order to accurately and comprehensively obtain the tunnel wall data, some projects use laser 3D scanners based on the principle of laser ranging to perform 3D analysis of the tunnel contour, comprehensively analyze the 3D situation of the tunnel wall, and guide the information construction of the tunnel, which has very important engineering application value. However, its defects are high cost, sparse point cloud, and the water mist environment has an absorption effect on the laser, resulting in no echo signal.

[0003] Some engineering applications use triangular geometric relationships to measure wall data, such as structured light cameras. Structured light 3D measurement is relatively low cost, but insufficient or overexposure of the projected light during use will affect the image quality, thereby affecting the quality of the generated 3D point cloud data. How to solve the exposure problem is an important consideration for the application of structured light sensors in engineering. In addition, the point cloud accuracy of structured light cameras is not exactly the same in the point cloud collection area. How to ensure the measurement accuracy of each area is also an important difficulty in the application of structured light in tunnel measurement. Summary of the invention

[0004] The purpose of the present invention is to provide a tunnel three-dimensional point cloud acquisition system, a structured light exposure adjustment system and method, so as to solve the technical problem in the prior art that insufficient or overexposure of projected structured light will affect the image quality, thereby affecting the generation quality of three-dimensional point cloud data.

[0005] In order to solve the above technical problems, the present invention provides a technical solution of a structured light exposure adjustment system: a structured light exposure adjustment system, comprising an image processing module and a controller;

[0006] The image processing module is used to obtain the proportion information of the number of pixels in each interval in the structured light grayscale image; the interval is divided into at least two according to the grayscale level;

[0007] The controller is used to adjust the exposure of the structured light camera according to the proportion information of the number of pixels in each interval until the quality of the acquired structured light image is qualified: if the proportion of the number of pixels in the low gray level interval is greater than the set threshold, and the proportion of the number of pixels in the high gray level interval is less than the set threshold, the exposure is increased and the photo is retaken; if the proportion of the number of pixels in the high gray level interval is greater than the set threshold, and the proportion of the number of pixels in the low gray level interval is less than the set threshold, the exposure is reduced and the photo is retaken.

[0008] The beneficial effect of the above technical solution is that the technical solution of the structured light exposure adjustment system of the present invention belongs to an improved invention. The grayscale value of each pixel of the image represents the brightness of the pixel, so the exposure state of the image can be judged by the grayscale information. Therefore, the present invention judges whether it is overexposed or underexposed by the pixel ratio in each grayscale interval, and adjusts the exposure of the structured light camera according to whether it is overexposed or underexposed and then reshoots. The present invention solves the technical problem in the prior art that insufficient or overexposure of the projected structured light will affect the image quality, thereby affecting the generation quality of three-dimensional point cloud data.

[0009] Further, the intervals include a lower grayscale interval, an intermediate grayscale interval, and a higher grayscale interval; accordingly, the controller adjusts the exposure of the structured light camera according to the proportion information of the number of pixels in each interval in the following manner:

[0010] If the proportion of the number of pixels in the lower grayscale interval and the middle grayscale interval is greater than the set threshold, and the proportion of the number of pixels in the higher grayscale interval is less than the set threshold, increase the exposure and retake the photo;

[0011] If the proportion of the number of pixels in the higher grayscale interval and the middle grayscale interval is greater than the set threshold, and the proportion of the number of pixels in the lower grayscale interval is less than the set threshold, the exposure is reduced and the photo is taken again.

[0012] Furthermore, the controller is also used to: if the proportion of the number of pixels in the middle grayscale interval is greater than a set threshold, and the proportion of the number of pixels in the lower grayscale interval and the higher grayscale interval is less than the set threshold, then adjust the exposure and shoot again, repeat at least twice, and the final image is obtained by averaging the repeatedly shot images.

[0013] Furthermore, if the proportion of the number of pixels in each interval is greater than a set threshold, the image quality is considered to be qualified.

[0014] The present invention also provides a technical solution for a tunnel three-dimensional point cloud acquisition system: a tunnel three-dimensional point cloud acquisition system, including a point cloud acquisition module and a structured light camera for acquiring a structured light image in the tunnel, and also including an image processing module and a controller:

[0015] The image processing module obtains the proportion information of the number of pixels in each interval in the structured light grayscale image; the interval is divided into at least two according to the grayscale level;

[0016] The controller adjusts the exposure of the structured light camera according to the proportion information until the quality of the structured light image is qualified: if the proportion of the number of pixels in the low gray level interval is greater than the set threshold, and the proportion of the number of pixels in the high gray level interval is less than the set threshold, the exposure is increased and the photo is retaken; if the proportion of the number of pixels in the high gray level interval is greater than the set threshold, and the proportion of the number of pixels in the low gray level interval is less than the set threshold, the exposure is reduced and the photo is retaken;

[0017] The point cloud acquisition module acquires the three-dimensional point cloud of the tunnel according to the structured light image with qualified quality.

[0018] The beneficial effect of the above technical solution is that the technical solution of a tunnel three-dimensional point cloud acquisition system of the present invention belongs to an improved invention. The grayscale value of each pixel of the grayscale image represents the brightness of the pixel, so the exposure state of the image can be judged by the grayscale information. Therefore, the present invention judges whether it is overexposed or underexposed by the pixel proportion in each grayscale interval, and adjusts the exposure of the structured light camera according to whether it is overexposed or underexposed and then re-shoots. The present invention solves the technical problem in the prior art that insufficient or overexposure of the projected structured light will affect the image quality, thereby affecting the generation quality of the three-dimensional point cloud data.

[0019] Further, the intervals include a lower grayscale interval, an intermediate grayscale interval, and a higher grayscale interval; accordingly, the controller adjusts the exposure of the structured light camera according to the proportion information of the number of pixels in each interval in the following manner:

[0020] If the proportion of the number of pixels in the lower grayscale interval and the middle grayscale interval is greater than the set threshold, and the proportion of the number of pixels in the higher grayscale interval is less than the set threshold, increase the exposure and retake the photo;

[0021] If the proportion of the number of pixels in the higher grayscale interval and the middle grayscale interval is greater than the set threshold, and the proportion of the number of pixels in the lower grayscale interval is less than the set threshold, the exposure is reduced and the photo is taken again.

[0022] Furthermore, the controller is also used to: if the proportion of the number of pixels in the middle grayscale interval is greater than a set threshold, and the proportion of the number of pixels in the lower grayscale interval and the higher grayscale interval is less than the set threshold, then adjust the exposure and shoot again, repeat at least twice, and the final image is obtained by averaging the repeatedly shot images.

[0023] Furthermore, if the proportion of the number of pixels in each interval is greater than a set threshold, the image quality is considered to be qualified.

[0024] Furthermore, the number of the structured light cameras is at least two, and the two structured light cameras constitute a binocular camera.

[0025] Furthermore, the structured light cameras are arranged in a straight line in sequence, a structured light projector for projecting structured light is arranged between every two adjacent structured light cameras, and the structured light cameras adjacent to the structured light projector on both sides of the structured light projector constitute a binocular camera.

[0026] Furthermore, the feature point matching method of the left and right camera images of the binocular camera is as follows: one of the left and right camera images is an original image, and the other is a target image; in the original image, a sample sub-area is taken as the center point of the feature point to be matched, and in the target image, a search sub-area slightly larger than the sample sub-area is used to traverse the target image, and the search sub-area with the highest similarity to the pixel points in the sample sub-area is taken as the target sub-area, and the center point of the target sub-area is taken as the feature point corresponding to the feature point to be matched in the target image.

[0027] Furthermore, the similarity is evaluated by a correlation coefficient, which is any one of a minimum difference square correlation coefficient, a normalized mutual correlation coefficient, and a covariance mutual correlation coefficient.

[0028] Furthermore, there is more than one structured light projector, and projection areas of at least two structured light projectors have overlapping parts.

[0029] The present invention also provides a technical solution of a structured light exposure adjustment method: a structured light exposure adjustment method, the method comprising:

[0030] S1. Obtaining the proportion information of the number of pixels in each interval of the structured light grayscale image respectively;

[0031] The interval is divided into at least two according to the gray level;

[0032] S2, if the proportion of the number of pixels in the low gray level interval is greater than the set threshold, and the proportion of the number of pixels in the high gray level interval is less than the set threshold, increase the exposure and retake the photo;

[0033] If the proportion of the number of pixels in the high gray level interval is greater than the set threshold, and the proportion of the number of pixels in the low gray level interval is less than the set threshold, then reduce the exposure and retake the photo;

[0034] Until the quality of the captured structured light image is qualified.

[0035] The beneficial effect of the above technical solution is that the technical solution of a structured light exposure adjustment method of the present invention belongs to an improved invention. The grayscale value of each pixel of the image represents the brightness of the pixel, so the exposure state of the image can be judged by the grayscale information. Therefore, the present invention determines whether it is overexposed or underexposed by the pixel proportion in each grayscale interval, and adjusts the exposure of the structured light camera based on whether it is overexposed or underexposed and then reshoots. The present invention solves the technical problem in the prior art that insufficient or overexposure of projected structured light will affect the image quality, thereby affecting the generation quality of three-dimensional point cloud data.

[0036] Further, the intervals in S1 include a lower gray level interval, an intermediate gray level interval and a higher gray level interval; accordingly, S2 includes:

[0037] If the proportion of the number of pixels in the lower grayscale interval and the middle grayscale interval is greater than the set threshold, and the proportion of the number of pixels in the higher grayscale interval is less than the set threshold, increase the exposure and retake the photo;

[0038] If the proportion of the number of pixels in the higher grayscale interval and the middle grayscale interval is greater than the set threshold, and the proportion of the number of pixels in the lower grayscale interval is less than the set threshold, the exposure is reduced and the photo is taken again.

[0039] Furthermore, S2 also includes: if the proportion of the number of pixels in the middle grayscale interval is greater than the set threshold, and the proportion of the number of pixels in the lower grayscale interval and the higher grayscale interval is less than the set threshold, then adjust the exposure and shoot again, repeat at least twice, and the final image is obtained by averaging the repeatedly shot images.

[0040] Furthermore, if the proportion of the number of pixels in each interval is greater than a set threshold, the image quality is considered to be qualified.

[0041] The present invention also provides a technical solution for a method for collecting three-dimensional point clouds in a tunnel: a method for collecting three-dimensional point clouds in a tunnel, the method comprising: adjusting the exposure of a structured light camera by a structured light exposure adjustment method to collect a structured light image in the tunnel, and obtaining a three-dimensional point cloud of the tunnel based on the structured light image; the structured light exposure adjustment method comprises:

[0042] S1. Obtaining the proportion information of the number of pixels in each interval of the structured light grayscale image respectively;

[0043] The interval is divided into at least two according to the gray level;

[0044] S2, if the proportion of the number of pixels in the low gray level interval is greater than the set threshold, and the proportion of the number of pixels in the high gray level interval is less than the set threshold, increase the exposure and retake the photo;

[0045] If the proportion of the number of pixels in the high gray level interval is greater than the set threshold, and the proportion of the number of pixels in the low gray level interval is less than the set threshold, then reduce the exposure and retake the photo;

[0046] Until the quality of the captured structured light image is qualified.

[0047] The beneficial effect of the above technical solution is that the technical solution of a tunnel three-dimensional point cloud acquisition method of the present invention belongs to an improved invention. The grayscale value of each pixel of the grayscale image represents the brightness of the pixel, so the exposure state of the image can be judged by the grayscale information. Therefore, the present invention determines whether it is overexposed or underexposed by the pixel proportion in each grayscale interval, and adjusts the exposure of the structured light camera based on whether it is overexposed or underexposed and then reshoots. The present invention solves the technical problem in the prior art that insufficient or overexposure of the projected structured light will affect the image quality, thereby affecting the generation quality of the three-dimensional point cloud data.

[0048] Further, the intervals in S1 include a lower gray level interval, an intermediate gray level interval and a higher gray level interval; accordingly, S2 includes:

[0049] If the proportion of the number of pixels in the lower grayscale interval and the middle grayscale interval is greater than the set threshold, and the proportion of the number of pixels in the higher grayscale interval is less than the set threshold, increase the exposure and retake the photo;

[0050] If the proportion of the number of pixels in the higher grayscale interval and the middle grayscale interval is greater than the set threshold, and the proportion of the number of pixels in the lower grayscale interval is less than the set threshold, the exposure is reduced and the photo is taken again.

[0051] Furthermore, S2 also includes: if the proportion of the number of pixels in the middle grayscale interval is greater than the set threshold, and the proportion of the number of pixels in the lower grayscale interval and the higher grayscale interval is less than the set threshold, then adjust the exposure and shoot again, repeat at least twice, and the final image is obtained by averaging the repeatedly shot images.

[0052] Furthermore, if the proportion of the number of pixels in each interval is greater than a set threshold, the image quality is considered to be qualified.

[0053] Furthermore, the number of the structured light cameras is at least two, and the two structured light cameras constitute a binocular camera.

[0054] Furthermore, the structured light cameras are arranged in a straight line in sequence, a structured light projector for projecting structured light is arranged between every two adjacent structured light cameras, and the structured light cameras adjacent to the structured light projector on both sides of the structured light projector constitute a binocular camera.

[0055] Furthermore, the feature point matching method of the left and right camera images of the binocular camera is as follows: one of the left and right camera images is an original image, and the other is a target image; in the original image, a sample sub-area is taken as the center point of the feature point to be matched, and in the target image, a search sub-area slightly larger than the sample sub-area is used to traverse the target image, and the search sub-area with the highest similarity to the pixel points in the sample sub-area is taken as the target sub-area, and the center point of the target sub-area is taken as the feature point corresponding to the feature point to be matched in the target image.

[0056] Furthermore, the similarity is evaluated by a correlation coefficient, which is any one of a minimum difference square correlation coefficient, a normalized mutual correlation coefficient, and a covariance mutual correlation coefficient.

[0057] Furthermore, there is more than one structured light projector, and projection areas of at least two structured light projectors have overlapping parts. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 Schematic diagram of feature point matching of left and right camera images in an embodiment of a tunnel three-dimensional point cloud acquisition system of the present invention;

[0059] Figure 2 It is a schematic diagram of the three-dimensional coordinate calculation principle of a space point in a world coordinate system in an embodiment of a tunnel three-dimensional point cloud acquisition system of the present invention;

[0060] Figure 3 It is a normalized grayscale histogram in the embodiment of the tunnel three-dimensional point cloud acquisition system of the present invention;

[0061] Figure 4 A grayscale histogram of a tunnel three-dimensional point cloud acquisition system under normal exposure in an embodiment of the present invention;

[0062] Figure 5 A grayscale histogram of overexposure in an embodiment of the tunnel three-dimensional point cloud acquisition system of the present invention;

[0063] Figure 6 A grayscale histogram when underexposed in an embodiment of the tunnel three-dimensional point cloud acquisition system of the present invention;

[0064] Figure 7 It is a schematic diagram of exposure adjustment of an embodiment of a tunnel three-dimensional point cloud acquisition system of the present invention;

[0065] Figure 8 A schematic diagram of multi-eye matrix measurement of an embodiment of a tunnel three-dimensional point cloud acquisition system of the present invention;

[0066] Fig. 9 A schematic diagram of binocular measurement of an embodiment of a tunnel three-dimensional point cloud acquisition system of the present invention;

[0067] Fig.10This is a front view of a multi-eye matrix measurement of an embodiment of a tunnel three-dimensional point cloud acquisition system of the present invention;

[0068] Fig.11 A side view of a multi-eye matrix measurement of an embodiment of a tunnel three-dimensional point cloud acquisition system of the present invention;

[0069] Fig.12 A top view of a multi-eye matrix measurement of an embodiment of a tunnel three-dimensional point cloud acquisition system of the present invention;

[0070] Fig.13 This is a three-dimensional structural diagram of the multi-eye matrix measurement of the tunnel three-dimensional point cloud acquisition system embodiment of the present invention.

[0071] Among them, 1. Install the circuit board; 2. Structured light camera; 3. Structured light projector. DETAILED DESCRIPTION

[0072] The present invention determines the exposure state of an image through grayscale information. The present invention determines whether the image is overexposed or underexposed by the percentage of pixels in each grayscale interval (, and adjusts the exposure of the structured light camera based on whether the image is overexposed or underexposed, and then reshoots the image. The present invention solves the technical problem in the prior art that insufficient or overexposure of the projected structured light will affect the image quality, thereby affecting the quality of the generated three-dimensional point cloud data.

[0073] Example of tunnel 3D point cloud acquisition system:

[0074] This embodiment uses a binocular structured light camera to obtain a structured light image, and uses the structured light image to obtain a tunnel three-dimensional point cloud. Due to the complex tunnel environment, when collecting structured light images, it is easy to have problems of overexposure or underexposure of the image. Therefore, this embodiment uses a structured light exposure adjustment method to adjust the exposure of the structured light camera to obtain a high-quality three-dimensional point cloud.

[0075] Specifically, the structured light projected in this embodiment is speckle structured light.

[0076] The tunnel three-dimensional point cloud acquisition system of this embodiment includes a structured light camera, a point cloud acquisition module, an image processing module and a controller. The structured light camera is used to acquire the structural image in the tunnel; the image processing module acquires the proportion information of the number of pixels in each interval in the structured light grayscale image; the controller adjusts the exposure of the structured light camera according to the proportion information of the number of pixels in each interval until the structured light image acquired by the structured light camera is of qualified quality; the point cloud acquisition module acquires the three-dimensional point cloud of the tunnel according to the qualified structured light image.

[0077] Among them, the image processing module and the controller constitute a structured light exposure adjustment system, and the various parts of the structured light exposure adjustment system cooperate with each other to realize the structured light exposure adjustment method.

[0078] like Figure 7 As shown, in this embodiment, the method for the controller to adjust the exposure of the structured light camera according to the proportion information of the number of pixels in each interval (i.e., the structured light exposure adjustment method) is as follows:

[0079] The image processing module converts the structured light image obtained by the structured light camera into a grayscale image, and obtains a grayscale histogram based on the grayscale image. The grayscale histogram is a function of the grayscale level. It indicates the number of pixels with a certain grayscale level in the image and can reflect the frequency of each grayscale in the image. Let the input image be I(x, y), with x×y pixels and a grayscale level of L. Let h(i) be the grayscale histogram of I(x, y):

[0080]

[0081] in

[0082]

[0083] Normalize the grayscale histogram and press 1 to normalize it. The normalized grayscale histogram is as follows: Figure 3 As shown in the figure, the vertical axis represents the ratio of the number of pixels of a certain gray level to the total number of pixels, that is, the number of pixels of a certain gray level after normalization.

[0084]

[0085] The grayscale value of each pixel in a grayscale image represents the brightness of the pixel, so different exposures will affect the distribution of the grayscale histogram, such as Figures 4 to 6 As shown in the figure, the images with normal exposure, overexposure and underexposure and the corresponding grayscale histograms. It can be seen that when the image is normally exposed, the grayscale values ​​of each pixel are mostly concentrated in the middle; if the image is overexposed, the grayscale values ​​of each pixel will be concentrated on the right; if the image is underexposed (i.e., underexposed), the grayscale values ​​of each pixel will be concentrated on the left.

[0086] Therefore, this embodiment divides the entire grayscale range into three intervals, namely 0-50 (i.e., a lower grayscale range), 51-200 (i.e., a middle grayscale range), and 201-255 (i.e., a higher grayscale range), and counts the proportion of the number of pixels in each grayscale range (i.e., a normalized value).

[0087] If the proportion of the number of pixels in the intervals of 0 to 50 and 51 to 200 is greater than 10% (i.e., the set threshold), and the proportion of the number of pixels in the interval of 201 to 255 is less than 10%, the image is considered underexposed, and the exposure is increased and the image is re-shot.

[0088] If the proportion of the number of pixels in the intervals of 201-255 and 51-200 is greater than 10% (i.e., the set threshold), and the proportion of the number of pixels in the interval of 0-50 is less than 10%, the image is considered to be overexposed, and the exposure is reduced and re-photographed.

[0089] If the proportion of the number of pixels in the interval of 51 to 200 is greater than 10% (i.e. the set threshold), and the proportion of the number of pixels in the interval of 0 to 50 and the interval of 201 to 255 is less than 10%, it is considered that the image exposure is uneven, so the exposure is adjusted and the photo is taken again. This is repeated at least twice, and the final image is obtained by averaging the repeatedly taken images.

[0090] If the proportion of the number of pixels in the three intervals is greater than 10%, the image quality is considered to be qualified.

[0091] Until the quality of the captured structured light image is qualified.

[0092] If the number of pixels in the range of 51 to 200 accounts for less than 10%, it is considered that the tunnel ambient light is too bright or too dark. At this time, adjusting the camera exposure cannot obtain a qualified image, and the ambient light needs to be adjusted (such as fill light, etc.).

[0093] In other implementations, it can also be divided into two intervals, a low gray level interval and a high gray level interval, such as 0 to 127 (low gray level interval) and 127 to 255 (high gray level interval), as follows:

[0094] If the proportion of the number of pixels in the range of 0 to 127 is greater than the set threshold, and the proportion of the number of pixels in the range of 127 to 255 is less than the set threshold, the image is considered underexposed, and the exposure is increased and re-shot.

[0095] If the proportion of the number of pixels in the range of 0 to 127 is less than the set threshold, and the proportion of the number of pixels in the range of 127 to 255 is greater than the set threshold, the image is considered to be overexposed, and the exposure is reduced and re-shot.

[0096] If the proportion of the number of pixels in the two intervals is greater than the set threshold, the image quality is considered qualified.

[0097] The threshold value may be set based on experience.

[0098] The method for the point cloud acquisition module to acquire the three-dimensional point cloud of the tunnel based on the qualified structured light image is as follows:

[0099] The binocular structured light vision measurement system uses two cameras to obtain two images of an object from different angles, then finds the corresponding feature points of the two images, and completes the measurement of the three-dimensional shape of the object through epipolar geometry. The structured light projection is to create features for the measured surface, and then the digital image correlation method is used to match the feature points.

[0100] like Figure 1 As shown, point P is any point in the left camera image (i.e., the original image). In order to calculate the size of the spatial position, a sample sub-area with a size of (2M+1)×(2M+1) pixels centered on point P is selected to correspond to the position of the sample sub-area in the right camera image, which is called the target sub-area. In the right camera image (i.e., the target image), a search sub-area with a size of (2N+1)*(2N+1)(N>M) is selected. The correlation function is used to calculate and compare the similarity between the sample sub-area and each sub-area in the search sub-area. Among them, the algorithm determines that the sub-area that makes the correlation function reach an extreme value has the highest similarity with the sample sub-area and is determined to be the target sub-area to be searched. The center point P of the sample sub-area and the center point P' of the target sub-area are corresponding points, and the coordinate difference between P and P' is the disparity. The position of the point in space can be obtained by performing spatial geometric operations based on the disparity.

[0101] In the operation of digital correlation algorithm, the correlation coefficient is a function used to evaluate the similarity between the grayscale value f(x,y) of the left camera image sub-area and the grayscale value g(x',y') of the right camera image sub-area. The algorithm is based on the correlation coefficient, searching in one of the image sub-areas, and finding the most similar sub-area in the other image, even the image sub-area with the extreme correlation coefficient. The correlation coefficient will affect the accuracy of the entire calculation and the convergence speed of the algorithm, so the selection of the correlation coefficient is crucial to the entire algorithm. The correlation coefficient has multiple definitions, which are mainly divided into two categories, namely the mutual correlation coefficient and the minimum square distance correlation coefficient. After rigorous mathematical derivation, experts and scholars have proved that the two correlation coefficients are equivalent to each other. The three representative mutual correlation coefficients are: the minimum difference square sum correlation coefficient C1, the normalized mutual correlation coefficient C2 and the covariance mutual correlation coefficient C3; the specific expressions are as follows:

[0102]

[0103] In the formula, and It represents the grayscale average value of all pixels in the sample sub-area and the target sub-area. Its specific expression is:

[0104]

[0105] Camera structure position parameter calibration: When the relative position between the two cameras is fixed, the camera parameters of the left and right cameras can be calibrated to obtain their relative position relationship, that is, the structure position parameters of the binocular vision measurement system. Two cameras with fixed positions simultaneously shoot the same calibration plate to obtain multiple calibration plate images with different postures, and solve all the internal and external parameters of the left and right cameras, that is, A l , A r , R l , R r , T l , T r .

[0106] According to the camera imaging model, the world coordinates (X w , Y w , Z w ) and the camera coordinates (x l ,y l 、z l )The relationship is:

[0107]

[0108] The world coordinates (X w , Y w , Z w ) and the camera coordinates (x r ,y r 、z r )The relationship is:

[0109]

[0110] Because the same calibration plate is photographed, different coordinates of the left and right camera coordinate systems in the same world coordinate system are obtained, and the above formula is unified:

[0111]

[0112] In order to express the relative position relationship of the left and right camera coordinate systems, the rotation matrix R = R l R r -1 and the translation vector T = T l -R l R r -1 T r Assuming that the world coordinate system coincides with the left camera coordinate system, the camera matrices of the left and right cameras are P and l =A l [I|0], P r =A r[R|T]. R and T are the structural position parameters of the system, which can be obtained by calibrating the external parameters of the two cameras.

[0113] After completing the pairing of the matching points of the left and right images, the spatial position relationship of the point is calculated. After calibrating the internal and external parameters of the left and right cameras, the coordinates of the spatial point M are inversely calculated from the coordinates of the pixel points on the imaging plane of the left and right cameras. The principle diagram of the inverse calculation of the three-dimensional world coordinates of the spatial point is as follows: Figure 2 shown.

[0114] According to the calibrated internal parameters of the camera, the coordinates (x ul ,y ul ) and (x ur ,y ur ), and then according to the projection imaging model of the camera, we can get the camera coordinates (x l ,y l ,z l ) and (x r ,y r ,z r ) is:

[0115]

[0116] The coordinate relationship between the left and right cameras is:

[0117]

[0118] in,

[0119]

[0120] Assume that the world coordinate system coincides with the left camera coordinate system, that is, [X w ,Y w ,Z w ] T =[x l, y l ,z l ] T .

[0121]

[0122] The spatial three-dimensional coordinates of the spatial point M can be obtained according to the imaging point M l and M r The coordinates in the image plane (x ul ,y ul ) and (x ur ,y ur )The system structure position parameters are solved:

[0123] X w =Z w x ul / f l

[0124] Y w =Z w y ul / f l

[0125]

[0126] This embodiment uses multi-eye matrix structured light to realize the arrangement between speckle projectors and image cameras in a matrix form. The patterns projected by two adjacent structured light projectors overlap, and finally a rectangular light source strip is formed. The overlapping part increases the point cloud density.

[0127] Conventional structured light has monocular and binocular structures, such as Figure 8 and Fig. 9 As shown in the figure, the effective acquisition area is the overlapping area of ​​the camera field of view and the structured light field of view. The characteristics of the collected point cloud data are that the point cloud is dense and accurate in the area close to the axis of the camera optical center, and the point cloud is sparse and accurate in the area far from the axis of the camera optical center. As the working distance of the camera increases, the point cloud at different depths becomes more sparse and the accuracy decreases.

[0128] In order to solve the problem of uneven point cloud density and inconsistent accuracy, this embodiment proposes a multi-eye matrix structured light form, such as Figures 10 to 13 As shown, it includes a mounting circuit board 1 and five structured light cameras 2 and four structured light projectors 9 mounted on the mounting circuit board 1. The five structured light cameras 2 are C1, C2, C3, C4, and C5 from left to right, and the four structured light projectors 9 are S1, S2, S3, and S4 from left to right, which can form sub-structured light modules of C1-S1-C2, C2-S2-C3, C3-S3-C4, and C4-S4-C5. The point cloud acquisition areas of the four sub-modules are spliced ​​to obtain high-density and high-precision point cloud acquisition.

[0129] Structured light exposure adjustment system embodiment:

[0130] A structured light exposure adjustment system includes an image processing module and a controller; the image processing module is used to obtain the proportion information of the number of pixels in each interval in the structured light grayscale image; the interval is divided into at least two according to the grayscale level; the controller is used to adjust the exposure of the structured light camera according to the proportion information of the number of pixels in each interval until the quality of the acquired structured light image is qualified: if the proportion of the number of pixels in the low grayscale interval is greater than the set threshold, and the proportion of the number of pixels in the high grayscale interval is less than the set threshold, then increase the exposure and re-shoot; if the proportion of the number of pixels in the high grayscale interval is greater than the set threshold, and the proportion of the number of pixels in the low grayscale interval is less than the set threshold, then reduce the exposure and re-shoot. The specific structured light exposure adjustment system has been introduced in sufficient detail in the above-mentioned tunnel three-dimensional point cloud acquisition system embodiment, and will not be repeated here.

[0131] Embodiment of the method for adjusting structured light exposure:

[0132] A method for adjusting structured light exposure, the method comprising: S1, respectively obtaining information on the proportion of the number of pixels in each interval of a structured light grayscale image; the interval is divided into at least two according to the grayscale level; S2, if the proportion of the number of pixels in the low grayscale interval is greater than a set threshold, and the proportion of the number of pixels in the high grayscale interval is less than the set threshold, then increase the exposure and reshoot; if the proportion of the number of pixels in the high grayscale interval is greater than the set threshold, and the proportion of the number of pixels in the low grayscale interval is less than the set threshold, then reduce the exposure and reshoot; until the quality of the structured light image captured is qualified. The present invention solves the technical problem in the prior art that insufficient or overexposure of projected structured light will affect the image quality, thereby affecting the quality of the generated three-dimensional point cloud data. The specific structured light exposure adjustment method has been introduced in sufficient detail in the above-mentioned tunnel three-dimensional point cloud acquisition system embodiment, and will not be repeated here.

[0133] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention is described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions recorded in the aforementioned embodiments without creative work, or replace some of the technical features therein with equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. A structured light exposure adjustment system, characterized in that: including an image processing module and a controller; The image processing module is used to obtain the proportion information of the number of pixels in each interval in the structured light grayscale image; the interval is divided into at least two according to the grayscale level; The controller is used to adjust the exposure of the structured light camera according to the proportion information of the number of pixels in each interval until the quality of the acquired structured light image is qualified: if the proportion of the number of pixels in the low gray level interval is greater than the set threshold, and the proportion of the number of pixels in the high gray level interval is less than the set threshold, the exposure is increased and the photo is retaken; if the proportion of the number of pixels in the high gray level interval is greater than the set threshold, and the proportion of the number of pixels in the low gray level interval is less than the set threshold, the exposure is reduced and the photo is retaken.

2. The structured light exposure adjustment system according to claim 1, characterized in that: The intervals include a lower grayscale interval, an intermediate grayscale interval, and a higher grayscale interval; accordingly, the controller adjusts the exposure of the structured light camera according to the proportion information of the number of pixels in each interval as follows: If the proportion of the number of pixels in the lower grayscale interval and the middle grayscale interval is greater than the set threshold, and the proportion of the number of pixels in the higher grayscale interval is less than the set threshold, increase the exposure and retake the photo; If the proportion of the number of pixels in the higher grayscale interval and the middle grayscale interval is greater than the set threshold, and the proportion of the number of pixels in the lower grayscale interval is less than the set threshold, the exposure is reduced and the photo is taken again.

3. The structured light exposure adjustment system according to claim 2, characterized in that: The controller is also used to: if the proportion of the number of pixels in the middle grayscale interval is greater than a set threshold, and the proportion of the number of pixels in the lower grayscale interval and the higher grayscale interval is less than the set threshold, then adjust the exposure and shoot again, repeat at least twice, and the final image is obtained by averaging the repeatedly shot images.

4. The structured light exposure adjustment system according to any one of claims 1 to 3, characterized in that: If the proportion of the number of pixels in each interval is greater than the set threshold, the image quality is considered to be qualified.

5. A tunnel three-dimensional point cloud acquisition system, comprising a point cloud acquisition module and a structured light camera for acquiring a structured light image in a tunnel, characterized in that: Also includes image processing module and controller: The image processing module obtains the proportion information of the number of pixels in each interval in the structured light grayscale image; the interval is divided into at least two according to the grayscale level; The controller adjusts the exposure of the structured light camera according to the proportion information until the quality of the structured light image is qualified: if the proportion of the number of pixels in the low gray level interval is greater than the set threshold, and the proportion of the number of pixels in the high gray level interval is less than the set threshold, the exposure is increased and the photo is retaken; if the proportion of the number of pixels in the high gray level interval is greater than the set threshold, and the proportion of the number of pixels in the low gray level interval is less than the set threshold, the exposure is reduced and the photo is retaken; The point cloud acquisition module acquires the three-dimensional point cloud of the tunnel according to the structured light image with qualified quality.

6. The tunnel 3D point cloud acquisition system according to claim 5, characterized in that: The intervals include a lower grayscale interval, an intermediate grayscale interval, and a higher grayscale interval; accordingly, the controller adjusts the exposure of the structured light camera according to the proportion information of the number of pixels in each interval as follows: If the proportion of the number of pixels in the lower grayscale interval and the middle grayscale interval is greater than the set threshold, and the proportion of the number of pixels in the higher grayscale interval is less than the set threshold, increase the exposure and retake the photo; If the proportion of the number of pixels in the higher grayscale interval and the middle grayscale interval is greater than the set threshold, and the proportion of the number of pixels in the lower grayscale interval is less than the set threshold, the exposure is reduced and the photo is taken again.

7. The tunnel three-dimensional point cloud acquisition system according to claim 6, characterized in that: The controller is also used to: if the proportion of the number of pixels in the middle grayscale interval is greater than a set threshold, and the proportion of the number of pixels in the lower grayscale interval and the higher grayscale interval is less than the set threshold, then adjust the exposure and shoot again, repeat at least twice, and the final image is obtained by averaging the repeatedly shot images.

8. The tunnel 3D point cloud acquisition system according to claim 5, characterized in that: If the proportion of the number of pixels in each interval is greater than the set threshold, the image quality is considered to be qualified.

9. The tunnel three-dimensional point cloud acquisition system according to claim 5, characterized in that: The number of the structured light cameras is at least two, and the two structured light cameras constitute a binocular camera.

10. The tunnel 3D point cloud acquisition system according to claim 8, characterized in that: The structured light cameras are arranged in a straight line in sequence, a structured light projector for projecting structured light is arranged between every two adjacent structured light cameras, and the structured light cameras adjacent to the structured light projector on both sides of the structured light projector constitute a binocular camera.

11. The tunnel three-dimensional point cloud acquisition system according to claim 9 or 10, characterized in that: The feature point matching method of the left and right camera images of the binocular camera is as follows: one of the left and right camera images is an original image, and the other is a target image; in the original image, a sample sub-area is taken as the center point of the feature point to be matched, and in the target image, a search sub-area slightly larger than the sample sub-area is used to traverse the target image, and a search sub-area with the highest degree of similarity to the pixel points in the sample sub-area is taken as the target sub-area, and the center point of the target sub-area is taken as the feature point corresponding to the feature point to be matched in the target image.

12. The tunnel three-dimensional point cloud acquisition system according to claim 11, characterized in that: The similarity is evaluated by a correlation coefficient, which is any one of a minimum difference square correlation coefficient, a normalized cross-correlation coefficient, and a covariance cross-correlation coefficient.

13. The tunnel 3D point cloud acquisition system according to claim 10, characterized in that: There is more than one structured light projector, and projection areas of at least two structured light projectors have overlapping parts.

14. A method for adjusting structured light exposure, characterized in that: The method includes: S1. Obtaining the proportion information of the number of pixels in each interval of the structured light grayscale image respectively; The interval is divided into at least two according to the gray level; S2, if the proportion of the number of pixels in the low gray level interval is greater than the set threshold, and the proportion of the number of pixels in the high gray level interval is less than the set threshold, increase the exposure and retake the photo; If the proportion of the number of pixels in the high gray level interval is greater than the set threshold, and the proportion of the number of pixels in the low gray level interval is less than the set threshold, then reduce the exposure and retake the photo; Until the quality of the captured structured light image is qualified.

15. The method for adjusting structured light exposure according to claim 14, characterized in that: The intervals in S1 include a lower gray level interval, an intermediate gray level interval, and a higher gray level interval; accordingly, S2 includes: If the proportion of the number of pixels in the lower grayscale interval and the middle grayscale interval is greater than the set threshold, and the proportion of the number of pixels in the higher grayscale interval is less than the set threshold, increase the exposure and retake the photo; If the proportion of the number of pixels in the higher grayscale interval and the middle grayscale interval is greater than the set threshold, and the proportion of the number of pixels in the lower grayscale interval is less than the set threshold, the exposure is reduced and the photo is taken again.

16. The method for adjusting structured light exposure according to claim 15, characterized in that: S2 also includes: if the proportion of the number of pixels in the middle grayscale interval is greater than the set threshold, and the proportion of the number of pixels in the lower grayscale interval and the higher grayscale interval is less than the set threshold, then adjust the exposure and shoot again, repeat at least twice, and the final image is obtained by averaging the repeatedly shot images.

17. The method for adjusting structured light exposure according to any one of claims 14 to 16, characterized in that: If the proportion of the number of pixels in each interval is greater than the set threshold, the image quality is considered to be qualified.

Images