Three-dimensional depth perception method for coal rock identification

Abstract

The invention discloses a three-dimensional depth perception method for coal rock recognition. The method comprises the steps that a blue light speckle projector projects a blue light speckle coding pattern to coal rock; the binocular camera collects projected blue light speckle coding patterns at the same time; carrying out monocular structure light depth decoding on an input blue light speckle pattern of the left-eye camera or the right-eye camera and the blue light speckle reference pattern, and outputting a rough disparity map of the irradiated coal rock; calculating a rough depth value of the coal rock according to a monocular structure light depth calculation formula and the rough disparity map; substituting the rough depth value into a binocular depth calculation formula to reversely solve a corresponding binocular parallax value, and performing binocular accurate matching calculation on the input blue light speckle images of the left and right eyes according to a reference parallax value to obtain an accurate parallax image; and according to a binocular depth calculation formula, calculating the accurate disparity map to obtain an accurate depth value of the coal rock. According to the method, through monocular and binocular structured light fusion depth decoding, accurate depth information is obtained, and the depth ranging precision and the detail resolution capability are greatly improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of mine geological exploration and intelligent coal mines, and relates to a three-dimensional depth perception method for coal and rock identification. Background Art

[0002] Within my country's coal industry, intelligent and digital transformation has become crucial for promoting coal mine safety, improving production efficiency, and reducing environmental impact. In particular, perception technology within the mine environment, particularly coal and rock identification technology, as a core component of intelligent coal mine development, is crucial for improving operational safety, automation, and emergency response capabilities.

[0003] The working environment underground in coal mines is characterized by low light, low texture, dust interference, and dynamic occlusion, making traditional depth sensing technologies difficult to meet practical application requirements. For example, traditional ToF (Time of Flight) cameras face severe multipath effects in mine environments, especially in the cramped working space underground. Multiple signal reflections lead to distorted depth data. Furthermore, absorption by black coal rock leads to significant depth errors. Compounding this with complex environments such as high dust and humidity, ranging performance is significantly degraded. Another common depth sensing technology, binocular stereo vision, offers advantages such as flexible structure and low cost, but its application in mine environments also presents significant challenges. Underground lighting is often insufficient, and coal rock cross-sections often lack distinct texture features. This makes it difficult for binocular vision systems to obtain stable matching points, resulting in holes and mismatches in depth maps. Traditional depth estimation methods based on image feature matching struggle to function effectively in low-light and low-texture environments, compromising the accuracy and reliability of the overall system. Summary of the Invention

[0004] In response to the shortcomings of the existing technology, the purpose of the present invention is to provide a three-dimensional depth perception method for coal and rock identification to solve the problems of insufficient depth perception accuracy and reliability in existing coal mines. The present invention is based on active and passive fusion visual perception technology, actively projects blue light speckle structured light coding patterns onto coal and rock, and then obtains accurate depth information through binocular reception and monocular and binocular structured light fusion depth decoding.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0006] A three-dimensional depth perception method for coal rock identification includes the following steps:

[0007] S100: The blue light speckle projector projects a blue light speckle coded pattern onto the coal rock;

[0008] S200: binocular cameras simultaneously collect the projected blue light speckle coding pattern;

[0009] S300: Monocular structured light depth decoding: Select the input blue light speckle pattern of the left or right camera, perform monocular structured light depth decoding with the blue light speckle reference pattern of a known reference distance, and output a rough disparity map of the illuminated coal rock;

[0010] S400: Monocular structured light depth calculation: Based on the monocular structured light depth calculation formula, the internal and external parameters of the monocular camera and blue light speckle projector, and the rough disparity map, the rough depth value of the coal rock is calculated;

[0011] S500: Binocular structured light depth decoding: Substitute the rough depth value into the binocular depth calculation formula to reversely obtain the corresponding binocular disparity value as the reference disparity value for binocular disparity calculation. The input blue light speckle patterns of the left and right binoculars are subjected to binocular precise matching calculation within the effective range based on the reference disparity value to obtain an accurate disparity map.

[0012] S600: Binocular structured light depth calculation: Based on the binocular depth calculation formula, the internal and external parameters of the binocular camera and the precise disparity map, the precise depth value of the coal rock is calculated.

[0013] The present invention also includes the following technical features:

[0014] Specifically, the blue light speckle projector includes a blue light LD laser source, a collimating lens and a DoE optical diffraction device. The projected blue light speckle coding pattern is a unique speckle block composed of a number of randomly distributed blue light spots.

[0015] Specifically, the binocular cameras and the blue light speckle projector are on the same baseline, and projection and reception are performed synchronously; the binocular cameras are symmetrically distributed on both sides of the blue light speckle projector and maintain a baseline distance S, and the baseline distance between the binocular cameras is 2S.

[0016] Specifically, the blue light speckle projector and the binocular camera work synchronously, and projection and reception are coordinated and collected;

[0017] Or the blue light speckle projector and the white light LED lighting are switched for projection, and the binocular camera synchronously adopts time-division multiplexing to alternately collect the blue light speckle coding pattern, that is, input the blue light speckle map and RGB map, where the input blue light speckle map is used for depth decoding and the RGB map is used for texture information collection.

[0018] Specifically, the S300 includes:

[0019] S301: Select the left camera or the right camera, and pre-process its input blue light speckle pattern by histogram enhancement or binarization operation to extract blue light speckles in the image, so that the acquired input blue light speckle pattern is highly similar to the speckle image blocks of the blue light speckle reference image of the camera at a known reference distance;

[0020] S302: performing block matching disparity calculation on the pre-processed input blue light speckle pattern and a blue light speckle reference pattern with a known reference distance, so as to obtain a rough disparity map corresponding to the pre-processed input blue light speckle pattern.

[0021] Specifically, the blue light speckle reference pattern is a blue light speckle coding pattern projected by a blue light speckle projector to a position perpendicular to the optical center axis Z axis of the projector and at a distance d from the projector. ref On the reference reference plane, a static image is captured by the camera and the standard speckle pattern is stored in the memory after image preprocessing.

[0022] Specifically, in S400, the monocular structured light depth calculation formula calculates d, with the horizontal offset Δx as an input parameter:

[0023]

[0024] Among them, d ref is the known distance of the monocular blue speckle reference image, Δx is the disparity value corresponding to the optimal matching block obtained by block matching disparity calculation between the preprocessed input blue speckle pattern and the preprocessed blue speckle reference image with a known distance, S is the baseline distance between the left or right camera and the blue speckle projector, f is the focal length of the left or right camera, and μ is the dot pitch of the image sensor of the left or right camera.

[0025] Specifically, in S500, the binocular depth calculation formula reversely calculates the binocular disparity value Δx' as a reference disparity value for binocular disparity calculation, with the rough depth value d as an input parameter:

[0026]

[0027] Where S is the baseline distance between the left or right camera and the blue light speckle projector, f is the focal length of the left or right camera, and μ is the dot pitch of the left or right camera image sensor.

[0028] Specifically, in S500, the input blue light speckle patterns of the left and right binoculars are subjected to binocular precise matching calculation within a valid range according to the reference disparity value to obtain a precise disparity map, including:

[0029] Preprocessing the input blue light speckle patterns of the left camera and the right camera by histogram enhancement or binarization operation;

[0030] The input speckle image block in the left-eye input blue light speckle map is used to search for the optimal speckle image matching block in the right-eye input blue light speckle map according to the reference disparity value; or the input speckle image block in the right-eye input blue light speckle map is used to search for the optimal speckle image matching block in the left-eye input blue light speckle map according to the reference disparity value; and the precise disparity map corresponding to the input blue light speckle map is obtained.

[0031] Specifically, in S600, d is calculated according to the binocular depth calculation formula. l,r , with a horizontal offset Δx l,r As input parameters:

[0032]

[0033] Where, the horizontal offset Δx l,r Input blue light speckle pattern I for the left eye l The input speckle image block B' and its corresponding right eye input blue light speckle pattern I r The optimal offset of the optimal speckle image matching block B in the X direction is denoted by , S is the baseline distance between the left or right camera and the blue light speckle projector, f is the focal length of the left or right camera, and μ is the dot pitch of the image sensor of the left or right camera.

[0034] Compared with the prior art, the present invention has the following technical effects:

[0035] This method uses the sparse depth map obtained by monocular structured light depth decoding under blue speckle projection as guidance information for high-resolution binocular structured light depth decoding. This significantly reduces the computational complexity of binocular structured light depth decoding (both the matching block and the search range are significantly reduced), resulting in a more accurate and detailed depth map. This method combines the advantages of a monocular structured light camera and binocular stereo ranging, overcoming the challenges of black absorption in coal and rock, low light levels, and weak textures. By fusing monocular and binocular structured light depth decoding, it significantly improves depth ranging accuracy and detail resolution. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a flow chart of a three-dimensional depth perception method for coal and rock identification according to an embodiment of the present invention.

[0037] Figure 2 Schematic diagram of monocular structured light depth calculation according to an embodiment of the present invention.

[0038] Figure 3 2 is a schematic diagram of inputting a speckle image block and searching for an optimal speckle image matching block according to an embodiment of the present invention.

[0039] Figure 4 2 is a schematic diagram of binocular block matching depth calculation according to an embodiment of the present invention. DETAILED DESCRIPTION

[0040] Specific embodiments of the present invention are given below. It should be noted that the present invention is not limited to the following specific embodiments, and all equivalent modifications made on the basis of the technical solution of this application fall within the protection scope of the present invention.

[0041] Example 1:

[0042] This embodiment provides a three-dimensional depth perception method for coal rock identification. Based on active and passive fusion visual perception technology, a blue light speckle structured light coding pattern is actively projected onto the coal rock. The speckle coding pattern is received by the left and right binoculars on the same baseline. After single and binocular structured light fusion depth decoding, accurate depth information is obtained.

[0043] Figure 1 The schematic diagram shows the overall process of the three-dimensional depth perception method for coal rock identification according to an embodiment of the present invention. Figure 2 、 Figure 3 、 Figure 4 To describe this method.

[0044] The three-dimensional depth perception method for coal and rock identification of the present invention comprises the following steps:

[0045] S100: Blue light speckle projector projects blue light speckle coded pattern onto coal rock:

[0046] The blue light speckle projector includes a blue light LD laser source, a collimator and a DoE (optical diffraction element). The projected blue light speckle coding pattern remains unchanged. The speckle block composed of several randomly distributed blue light speckle codes within a certain range is unique.

[0047] S200: Binocular cameras simultaneously capture the projected blue light speckle coding pattern:

[0048] The binocular cameras and the blue light speckle projector are on the same baseline, and projection and reception are performed synchronously. The binocular cameras have the same specifications and are symmetrically distributed on both sides of the blue light speckle projector, maintaining a certain baseline distance S, that is, the baseline distance between the two cameras is 2S.

[0049] One working mode is that the blue light speckle projector and the binocular camera work synchronously, that is, projection and reception are coordinated and collected;

[0050] Another working mode is that the blue light speckle projector and the white light LED lighting can be switched for projection. The binocular camera can synchronously use time-division multiplexing to alternately collect the blue light speckle coding pattern (i.e., the input blue light speckle pattern) and the RGB image, where the input blue light speckle pattern is used for depth decoding and the RGB image is used for texture information collection.

[0051] S300: Monocular structured light depth decoding:

[0052] The monocular structured light depth decoding module selects the input blue light speckle pattern of the left or right camera, performs monocular structured light depth decoding with the blue light speckle reference pattern of a known reference distance, and outputs a rough disparity map of the illuminated coal rock.

[0053] Specifically, monocular structured light depth decoding includes the following steps:

[0054] S301: Select the left camera or the right camera, and pre-process its input blue light speckle pattern through histogram enhancement or binarization operation to extract blue light speckles in the image, so that the acquired input blue light speckle pattern is highly similar to the speckle image blocks of the blue light speckle reference image of the camera at a known reference distance.

[0055] S302: performing block matching disparity calculation on the pre-processed input blue light speckle pattern and a blue light speckle reference pattern with a known reference distance to obtain a rough disparity map corresponding to the pre-processed input blue light speckle pattern;

[0056] In this step, the pre-processed input blue light speckle pattern is compared with the blue light speckle reference image at a known reference distance by block matching disparity calculation (the specific calculation is as follows) based on a certain size of speckle image block (the size of the speckle image block is k×l, k and l are positive integers). Figure 2 As shown in Figure 1, the specific strategy is to search for the best matching block with the highest similarity within a K×L (K and L are integers greater than k and l) search window of the monocular blue light speckle reference image for a k×l input speckle image block (the search process can refer to the traditional image block matching method) to obtain the disparity value Δx corresponding to the best matching block, and then calculate the coarse disparity map corresponding to the input blue light speckle image block by block.

[0057] Before starting work, it is necessary to collect and solidify the blue light speckle reference image of the left or right camera as a reference for matching and comparison. The blue light speckle reference image is a blue light speckle coding pattern projected by the blue light speckle projector to a point perpendicular to the optical center axis (Z axis) of the projector and at a distance d from the projector. ref On a plane (the plane can be composed of a projection cloth, a flat plate, etc., for presenting a clear and stable image, and the plane can be called a reference reference plane), a static image is captured by a camera, and after image preprocessing, it is stored and solidified in a memory for matching the reference and the standard pattern for the monocular structured light block matching parallax calculation. Preferably, the blue light speckle reference image of the blue light speckle projector is composed of a plurality of scattered spots with a known distance d ref Standard speckle pattern.

[0058] S400: Monocular structured light depth calculation:

[0059] According to the monocular structured light depth calculation formula, the internal and external parameters of the monocular camera and blue light speckle projector, and the rough disparity map, the rough depth value d of the coal rock is calculated.

[0060] In this embodiment, d is calculated according to the following monocular depth calculation formula (1), taking the horizontal offset Δx as an input parameter as an example:

[0061]

[0062] like Figure 2 In which, d ref is the known distance of the monocular blue speckle reference image, Δx is the disparity value corresponding to the optimal matching block obtained by block matching disparity calculation between the preprocessed input blue speckle pattern and the preprocessed blue speckle reference image with a known distance, S is the baseline distance between the left or right camera and the blue speckle projector, f is the focal length of the left or right camera, and μ is the dot pitch of the image sensor of the left or right camera.

[0063] S500: Binocular structured light depth decoding:

[0064] The binocular structured light depth decoding module substitutes the rough depth value obtained in step S400 into the binocular depth calculation formula to reversely calculate the corresponding binocular disparity value (which serves as the reference disparity value for binocular disparity calculation). The input blue light speckle patterns of the left and right binoculars are subjected to binocular precise matching calculation within the effective range based on the reference disparity value to obtain a more detailed and accurate disparity map.

[0065] In this embodiment, the binocular disparity value Δx' (serving as a reference disparity value for binocular disparity calculation) is reversely calculated according to the following binocular depth calculation formula (2), taking the rough depth value d as an input parameter as an example:

[0066]

[0067] Where S is the baseline distance between the left or right camera and the blue light speckle projector, f is the focal length of the left or right camera, and μ is the dot pitch of the left or right camera image sensor.

[0068] The input blue light speckle patterns of the left and right eyes are calculated for binocular precise matching within the effective range based on the reference disparity value Δx'. One mode is to input the input speckle image block in the blue light speckle pattern of the left eye, and input the blue light speckle image block in the right eye based on the reference disparity value Δx'. Figure 1 The optimal speckle image matching block is searched within a certain range; the other mode is that the right eye inputs the input speckle image block in the blue light speckle image, and the left eye inputs the blue light speckle image block according to the reference disparity value Δx' Figure 1 Search for the best speckle image matching block within a certain range. The details are as follows:

[0069] S501: The input blue light speckle patterns of the left camera and the right camera are preprocessed by the same preprocessing method as step S301, and the input blue light speckle patterns are preprocessed by histogram enhancement or binarization operation.

[0070] S502: One mode is binocular input of blue light speckle pattern I l , I r Mutual binocular block matching calculation, see Figure 3 , we can get the left eye input blue light speckle pattern I l The corresponding precise disparity map. Specifically:

[0071] Input blue light speckle pattern I in the left eye l Extract a certain size of input speckle image block B'block m×n , the center point is o'; according to the reference parallax value Δx', input the blue light speckle pattern I in the right eye r Extract the matching search window Match of a certain size with the position o corresponding to the center point o' of the input speckle image block as the center (the horizontal disparity value between o' and o is Δx') M×N , the size is M×N, M and N are integers, which can be equal or unequal, generally M≥N, M>m, N≥n; then in the matching search window Match M×N Extract all speckle matching blocks that are the same size as the input speckle image blocks match k , size is m×n, matching block center point o' k , k is an integer, indicating the number of matching blocks.

[0072] Preferably, the input speckle image block size m×n is smaller than the speckle image block size k×l selected by the monocular structured light depth decoding in step 302; Match search window Match M×N The size M×N is also smaller than the direct binocular matching calculation.

[0073] Next, we calculate the input speckle image blocks respectively Match with k speckle matching blocks k The similarity value match_value between k , which is used as a similarity measurement indicator for image block matching.

[0074] Finally, among all similarity values ​​match_value k Find the minimum value in the speckle matching block match k That is the input speckle image block B'block m×n The optimal speckle image matching block B to be searched, the position information corresponding to the minimum value is the input speckle image block block m×n The offset of the center point o (Δxl,r , Δy l,r ), that is, the motion vector of the input speckle image block B'. Figure 3 As shown, the input speckle image block is the left eye input blue light speckle image I l The middle gray represents the area, and the best matching block is the right eye input blue light speckle pattern I r The slash in the matching search window represents the area, and its center point is o k Matching search window block M×N The optimal offset of the center point o (the center point o corresponds to the center point o of the input speckle image block) is (Δx l,r , Δy l,r ), respectively representing the displacement in the X and Y axis directions. The offset value is calculated by taking the absolute value of the coordinate value (x, y) of the center point o of the matching search window minus the coordinate value (x', y') of the center point of the optimal matching block according to the X and Y axes, and is expressed in the number of pixels.

[0075] S600: Binocular structured light depth calculation:

[0076] The depth calculation module calculates the precise depth value of the coal rock based on the binocular depth calculation formula, the internal and external parameters of the binocular camera and the precise disparity map.

[0077] In this embodiment, d is calculated according to the following binocular depth calculation formula (3): l,r , with a horizontal offset Δx l,r As an example of input parameters, see Figure 4 :

[0078]

[0079] Where, the horizontal offset Δx l,r Input blue light speckle pattern I for the left eye l The input speckle image block B' and its corresponding right eye input blue light speckle pattern I r The optimal offset of the optimal speckle image matching block B in the X direction, that is, the right eye input blue light speckle pattern I r The x-coordinate value of the center point of the matching search window of the input image block B minus the x-coordinate value of the center point of the optimal matching block B found in the matching search window is taken as the absolute value and expressed in the number of pixels. S is the baseline distance between the left or right camera and the blue light speckle projector, f is the focal length of the left or right camera, and μ is the dot pitch of the image sensor of the left or right camera.

[0080] Preferably, the size of the input speckle image block is selected based on the relative uniqueness of the input speckle image block within a certain range in the horizontal or vertical direction, that is, the characteristics of the input speckle image block are different from those of other input speckle image blocks of the same size and can be distinguished from other input speckle image blocks of the same size.

[0081] Preferably, the similarity value calculation method of the present invention adopts a method of calculating the sum of absolute differences (SAD) between corresponding pixels of the input image block and the matching block, but is not limited to this method.

[0082] Move the center point o of the projected image block to the next pixel point in the same row and repeat step 502 to calculate the depth value corresponding to the next pixel point. In this way, the left-eye input blue light speckle pattern I is calculated point by point from left to right and from top to bottom. l The corresponding accurate depth value of the entire image.

[0083] Similarly, the right eye inputs blue light speckle pattern I r Alternatively, the blue light speckle pattern I may be input to the left eye according to steps S502 and S600. l Search for the optimal speckle image matching block within a certain range and obtain the corresponding accurate depth value of the entire image.

[0084] As an example, the binocular camera of the present invention uses two independent cameras with the same performance indicators (same optical lens and image sensor), which are arranged symmetrically and equidistantly on both sides of the blue light speckle projector. Their optical axes are parallel to the optical axis of the coded pattern projector and are on the same baseline. However, the baselines of the two cameras can also be adjusted according to different needs, or two cameras with different focal lengths or models can be used.

[0085] As an example, the search strategy for matching blocks of the present invention adopts traditional full search block matching, but various other improved search strategies can also be adopted; the similarity value calculation method adopts the sum of absolute differences (SAD) method, but is not limited to this method; all methods that adopt the content flow similar to the present invention should be included in the scope of the claims of the present invention.

[0086] While the above embodiments are implemented in a specific system, they are not intended to limit the present invention. The present invention can be similarly applied to similar coded pattern projection and image sensor systems. The present invention not only supports structured light patterns generated by various laser sources, such as infrared, visible light, ultraviolet light, and invisible light, but also accommodates projection schemes with various patterns, such as dots, blocks, crosses, and stripes. Therefore, modifications and improvements that do not depart from the spirit and scope of the present invention are intended to be encompassed by the claims above.

[0087] The preferred embodiments of the present invention are described in detail above in conjunction with the accompanying drawings. However, the present invention is not limited to the specific details in the above embodiments. Within the technical concept of the present invention, various simple modifications can be made to the technical solution of the present invention, and these simple modifications all fall within the scope of protection of the present invention.

[0088] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any appropriate manner without contradiction. To avoid unnecessary repetition, the present invention will not further describe various possible combinations.

[0089] In addition, the various embodiments of the present invention may be arbitrarily combined, and as long as they do not violate the concept of the present invention, they should also be regarded as the contents disclosed by the present invention.

Claims

1. A three-dimensional depth perception method for coal rock identification, characterized in that: The following steps are involved: S100: The blue light speckle projector projects a blue light speckle coded pattern onto the coal rock; S200: binocular cameras simultaneously collect the projected blue light speckle coding pattern; S300: Monocular structured light depth decoding: Select the input blue light speckle pattern of the left or right camera, perform monocular structured light depth decoding with the blue light speckle reference pattern of a known reference distance, and output a rough disparity map of the illuminated coal rock; S400: Monocular structured light depth calculation: Based on the monocular structured light depth calculation formula, the internal and external parameters of the monocular camera and blue light speckle projector, and the rough disparity map, the rough depth value of the coal rock is calculated; S500: Binocular structured light depth decoding: Substitute the rough depth value into the binocular depth calculation formula to reversely obtain the corresponding binocular disparity value as the reference disparity value for binocular disparity calculation. The input blue light speckle patterns of the left and right binoculars are subjected to binocular precise matching calculation within the effective range based on the reference disparity value to obtain an accurate disparity map. S600: Binocular structured light depth calculation: Based on the binocular depth calculation formula, the internal and external parameters of the binocular camera and the precise disparity map, the precise depth value of the coal rock is calculated.

2. The three-dimensional depth perception method for coal rock identification according to claim 1, characterized in that: The blue light speckle projector includes a blue light LD laser source, a collimating lens and a DoE optical diffraction device. The projected blue light speckle coding pattern is a unique speckle block composed of a number of randomly distributed blue light spots.

3. The three-dimensional depth perception method for coal rock identification according to claim 1, characterized in that: The binocular cameras and the blue light speckle projector are on the same baseline, and projection and reception are performed synchronously; the binocular cameras are symmetrically distributed on both sides of the blue light speckle projector and maintain a baseline distance S, and the baseline distance between the binocular cameras is 2S.

4. The three-dimensional depth perception method for coal rock identification according to claim 1, characterized in that: The blue light speckle projector works synchronously with the binocular camera, and projection and reception are coordinated and collected; Or the blue light speckle projector and the white light LED lighting are switched for projection, and the binocular camera synchronously adopts time-division multiplexing to alternately collect the blue light speckle coding pattern, that is, input the blue light speckle map and RGB map, where the input blue light speckle map is used for depth decoding and the RGB map is used for texture information collection.

5. The three-dimensional depth perception method for coal rock identification according to claim 1, characterized in that: The S300 includes: S301: Select the left camera or the right camera, and pre-process its input blue light speckle pattern by histogram enhancement or binarization operation to extract blue light speckles in the image, so that the acquired input blue light speckle pattern is highly similar to the speckle image blocks of the blue light speckle reference image of the camera at a known reference distance; S302: performing block matching disparity calculation on the pre-processed input blue light speckle pattern and a blue light speckle reference pattern with a known reference distance, so as to obtain a rough disparity map corresponding to the pre-processed input blue light speckle pattern.

6. The three-dimensional depth perception method for coal rock identification according to claim 5, characterized in that: The blue light speckle reference image is a blue light speckle coding pattern projected by a blue light speckle projector to a position perpendicular to the optical center axis Z axis of the projector and at a distance d from the projector. ref On the reference reference plane, a static image is captured by the camera and the standard speckle pattern is stored in the memory after image preprocessing.

7. The three-dimensional depth perception method for coal rock identification according to claim 1, characterized in that: In S400, the monocular structured light depth calculation formula calculates d, with the horizontal offset Δx as an input parameter: Among them, d ref is the known distance of the monocular blue speckle reference image, Δx is the disparity value corresponding to the optimal matching block obtained by block matching disparity calculation between the preprocessed input blue speckle pattern and the preprocessed blue speckle reference image with a known distance, S is the baseline distance between the left or right camera and the blue speckle projector, f is the focal length of the left or right camera, and μ is the dot pitch of the image sensor of the left or right camera.

8. The three-dimensional depth perception method for coal rock identification according to claim 1, characterized in that: In S500 , the binocular depth calculation formula reversely calculates the binocular disparity value Δx′ as a reference disparity value for binocular disparity calculation, with the rough depth value d as an input parameter: Where S is the baseline distance between the left or right camera and the blue light speckle projector, f is the focal length of the left or right camera, and μ is the dot pitch of the left or right camera image sensor.

9. The three-dimensional depth perception method for coal rock identification according to claim 1, characterized in that: In the above S500, the input blue light speckle patterns of the left and right binoculars are subjected to binocular precise matching calculation within a valid range according to the reference disparity value to obtain a precise disparity map, including: Preprocessing the input blue light speckle patterns of the left camera and the right camera by histogram enhancement or binarization operation; The input speckle image block in the left-eye input blue light speckle map is used to search for the optimal speckle image matching block in the right-eye input blue light speckle map according to the reference disparity value; or the input speckle image block in the right-eye input blue light speckle map is used to search for the optimal speckle image matching block in the left-eye input blue light speckle map according to the reference disparity value; and the precise disparity map corresponding to the input blue light speckle map is obtained.

10. The three-dimensional depth perception method for coal rock identification according to claim 1, characterized in that: In the step S600, d is calculated according to the binocular depth calculation formula. l,r , with a horizontal offset Δx l,r As input parameters: Where, the horizontal offset Δx l,r Input blue light speckle pattern I for the left eye l The input speckle image block B' and its corresponding right eye input blue light speckle pattern I r The optimal offset of the optimal speckle image matching block B in the X direction is denoted by , S is the baseline distance between the left or right camera and the blue light speckle projector, f is the focal length of the left or right camera, and μ is the dot pitch of the image sensor of the left or right camera.

Images