CN113344945B - Automatic rock mass blasting block size analysis device and method based on binocular vision - Google Patents
Automatic rock mass blasting block size analysis device and method based on binocular vision Download PDFInfo
- Publication number
- CN113344945B CN113344945B CN202110600753.6A CN202110600753A CN113344945B CN 113344945 B CN113344945 B CN 113344945B CN 202110600753 A CN202110600753 A CN 202110600753A CN 113344945 B CN113344945 B CN 113344945B
- Authority
- CN
- China
- Prior art keywords
- rock
- module
- image
- unit
- block
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000011435 rock Substances 0.000 title claims abstract description 406
- 238000004458 analytical method Methods 0.000 title claims abstract description 28
- 238000000034 method Methods 0.000 title claims abstract description 14
- 238000005422 blasting Methods 0.000 title claims description 14
- 238000009826 distribution Methods 0.000 claims abstract description 50
- 230000011218 segmentation Effects 0.000 claims description 131
- 238000000605 extraction Methods 0.000 claims description 71
- 238000006243 chemical reaction Methods 0.000 claims description 55
- 238000012937 correction Methods 0.000 claims description 46
- 238000004364 calculation method Methods 0.000 claims description 42
- 238000003860 storage Methods 0.000 claims description 42
- 238000013500 data storage Methods 0.000 claims description 30
- 230000010354 integration Effects 0.000 claims description 27
- 238000001914 filtration Methods 0.000 claims description 24
- 238000001514 detection method Methods 0.000 claims description 15
- 230000004927 fusion Effects 0.000 claims description 15
- 230000009466 transformation Effects 0.000 claims description 15
- 238000004806 packaging method and process Methods 0.000 claims description 9
- 230000001960 triggered effect Effects 0.000 claims description 9
- 230000008859 change Effects 0.000 claims description 6
- 230000008676 import Effects 0.000 claims description 6
- 230000005540 biological transmission Effects 0.000 claims description 4
- 238000012935 Averaging Methods 0.000 claims description 3
- 239000003086 colorant Substances 0.000 claims description 3
- 230000003247 decreasing effect Effects 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000001629 suppression Effects 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 8
- 238000012544 monitoring process Methods 0.000 abstract description 3
- 230000002349 favourable effect Effects 0.000 abstract description 2
- 238000012545 processing Methods 0.000 abstract description 2
- 238000011897 real-time detection Methods 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 9
- 238000005259 measurement Methods 0.000 description 3
- 238000000691 measurement method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration using histogram techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/80—Geometric correction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
- G06T2207/10012—Stereo images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30132—Masonry; Concrete
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30204—Marker
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention discloses a binocular vision-based rock burst block size automatic analysis device and method. The basic module controls the hardware part and combines other modules to form an overall system framework, and equipment index configuration units in the basic module are used for configuring equipment information and parameters. After the system recognizes the rock mass image, a user can view and display the distribution information of rock mass blocks through the controller. The invention provides the technology based on automatic monitoring, digital technology, binocular vision image processing and the like for real-time image data acquisition and rock mass distribution analysis, has good real-time performance and reliability, and is favorable for real-time detection of on-site rock mass distribution.
Description
Technical Field
The invention relates to rock mass blasting block size detection in the field of geotechnical engineering, in particular to a binocular vision-based rock mass blasting block size automatic analysis device and method.
Background
In rock blasting engineering, the blocking degree is one of important parameters for evaluating the effect of the rock blasting engineering, and the subsequent operations such as shipment, mechanical crushing and the like are directly influenced. The detection of the rock mass block degree of the blasting has the advantages of complex working procedure, high manual measurement cost and large error, and very complex work is required to be carried out when checking the rock mass degree of a certain rock mass degree on site.
At present, a direct measurement method or a traditional screening method is often adopted for obtaining rock mass size information of rock mass block detection, for example, the direct measurement method is to set a plurality of straight lines on a tested rock mass pile, observe the length of each rock mass occupying the straight line, and calculate the rock mass distribution of each level. On one hand, the method takes a straight line belt as a main tool to carry out visual estimation on the rock mass, and has individual measurement errors; on the other hand, the technician may misread and misrecognize the data, so that a large deviation is generated between the rock record length and the actual length of the rock record, or the data record format is incorrect, and the subsequent analysis work becomes difficult.
Disclosure of Invention
The invention aims to solve the technical problem of providing a binocular vision-based rock mass blasting block size automatic analysis device and a binocular vision-based rock mass blasting block size automatic analysis method, wherein the device can acquire the distribution information of the blasting block sizes in real time, provides a technical means for developing the blasting block size analysis in a complex environment, and provides references and bases for blasting parameter design and optimization.
In order to solve the problems in the prior art, the invention adopts the following technical scheme:
the rock burst block size automatic analysis device based on binocular vision comprises a camera, a processor, a memory, a display and a controller, wherein the processor comprises a base module, a camera depth conversion module, a rock depth map extraction module, a rock segmentation module, a single rock actual size estimation module, a rock block size statistics module and a public data storage area; the basic module comprises a binocular image acquisition module, an image storage module, an image copying module, an information display module, an equipment index configuration module and an equipment control module; the binocular image acquisition module receives an image obtained by the camera and stores the image into the memory, the image storage module is connected with the memory, the information display module is connected with the display, the equipment control module is connected with the controller, the image copying module is connected with the basic module, the camera depth conversion module, the rock depth map extraction module, the rock segmentation module, the single rock actual size estimation module and the rock block size statistics module, the image copying module is connected with the memory in the basic module, and the equipment index configuration module is respectively connected with the camera, the memory, the display and the controller and is used for controlling the data mode of equipment input or output and related parameters, units and indexes of the system; the image storage module is respectively connected with the camera depth conversion module, the rock depth map extraction module and the rock segmentation module, and the image storage module and the equipment index configuration module are respectively connected with the single rock actual size estimation module; the information display module is respectively connected with the camera depth conversion module, the rock depth map extraction module, the rock segmentation module, the single rock actual size estimation module and the rock block size statistics module and is used for providing display events for the modules;
The distortion correction unit of the camera depth conversion module is connected with the image storage module and is used for correcting distortion of the acquired original image; the camera depth conversion module is connected with the rock block segmentation module; one output end of the camera depth conversion module is connected with the rock mass depth map extraction module, and one input end of the camera depth conversion module is connected with the single rock mass actual size estimation module; one output end of the camera depth conversion module is connected with the image copying module, and one output end of the rock depth map extraction module is connected with the information display module and is used for providing a display event and displaying a parallax map obtained by calculating a current binocular image;
the device comprises a rock depth map extraction module, an image storage module, a rock segmentation module, a camera depth conversion module, a display event display module and a parallax map, wherein the rock depth map extraction module is connected with the image storage module, one output end of the rock depth map extraction module is connected with the rock segmentation module, one input end of the rock depth map extraction module is connected with one output end of the camera depth conversion module, and one output end of the rock depth map extraction module is connected with the information display module and is used for providing a display event and displaying the parallax map obtained by current binocular image calculation;
the rock block segmentation module is connected with the image storage module, one input end of the rock block segmentation module is connected with one output end of the rock block depth map extraction module, one output end of the rock block segmentation module is connected with one input end of the single rock block actual size estimation module, the rock block segmentation module is connected with the rock block size statistics module, and one output end of the rock block segmentation module is connected with the information display module and is used for providing display events and displaying actual segmentation conditions of the rock blocks;
The single rock actual size estimation module is connected with the rock block size statistics module and the camera depth conversion module and is used for acquiring parameters of an image coordinate to a world coordinate; an input end of the single rock actual size estimation module is connected with an output end of the rock segmentation module and is used for receiving single rock information to carry out actual size estimation; an output end of the actual size estimation of the single rock mass is connected with the information display module and is used for providing a display event and displaying the actual size estimation condition of the current rock mass;
the rock block size statistics module is respectively connected with the image storage module and the rock block segmentation module and is used for receiving all rock block information to carry out size distribution statistics; the rock block size statistics module is connected with the single rock block actual size estimation module and is used for calculating the actual size information of the single rock block; the rock block size statistics module is connected with the information display module and is used for providing display events and displaying the actual size distribution condition of all the current rock blocks.
The public data storage area is respectively connected with the camera depth conversion module, the rock depth map extraction module, the rock segmentation module, the single rock actual size estimation module and the rock block size statistics module and is used for storing fracture characteristic data information packets submitted by the modules, and the public data storage area is also connected with the basic module and is used for storing and initializing default parameters of solidified equipment.
Further, the camera depth conversion module comprises a distortion correction unit, a parallax coordinate transformation unit and a world coordinate calculation unit; the distortion correcting unit is connected with the rock mass depth map extracting module,
the distortion correction unit is used for inputting image data, correcting an initial distortion image, and outputting corrected image data to the rock depth map module for pixel-level depth estimation;
the parallax coordinate transformation unit is connected with the world coordinate calculation unit and is used for inputting parallax and image coordinates; the world coordinate calculation unit is used for outputting world coordinates;
the rock depth map extraction module comprises a characteristic scanning unit, a characteristic matching unit and a parallax calculation unit; the characteristic scanning unit, the characteristic matching unit and the parallax calculating unit are sequentially connected and used for providing a data transmission channel; the characteristic scanning unit is used for inputting binocular image data and outputting a parallax image, and the parallax calculating unit is used for calculating a parallax value which is optimally matched; the feature matching unit is used for calculating the matching degree of the feature data and collecting the parallax value returned by the parallax calculation unit to construct a parallax map; the parallax calculating unit is connected with the feature matching unit and is used for calculating the parallax value of the best matching feature; the feature matching unit is connected with the rock segmentation module and transmits the extracted parallax image to the rock segmentation module for subsequent segmentation;
The rock block segmentation module comprises a rock block rough segmentation unit, a noise filtering unit and a rock block integration unit; the rock block rough segmentation unit is connected with the rock block integration unit through the noise filtering unit, the rock block rough segmentation unit is used for inputting depth image data, the noise filtering unit is used for filtering noise small areas of non-rock blocks, and the rock block integration unit is used for outputting all rock block information.
The single rock actual size estimation module comprises a contour depth statistics unit, a contour width statistics unit and an actual size calculation unit; the profile depth statistical unit is connected with the actual size calculation unit through the profile width statistical unit; the contour depth statistics unit is used for inputting single rock block information, and the actual size calculation unit is used for outputting the actual size of the rock block;
the rock block degree statistics module comprises a block degree extraction unit, a block degree distribution statistics unit and an information integration unit; the block degree extraction unit, the block degree distribution statistical unit and the information integration unit are sequentially connected, the block degree extraction unit is used for inputting single rock block information, and the information integration unit is used for outputting rock block distribution information.
An analysis method of a rock mass block degree automatic analysis device based on binocular vision comprises the following steps:
Step 1: according to the required parameters, equipment index configuration and equipment initialization are carried out, wherein the parameters comprise the size of an image acquired by a camera and a lens distortion correction coefficient; after configuration is finished, the camera, the display and other devices work normally according to the required parameters; starting a binocular image acquisition module, and acquiring rock images to an image storage module according to a corresponding format;
step 2: the image storage module transmits the rock block image acquired by the camera in the step 1 to the camera depth conversion module, and the camera depth conversion module acquires configuration information from the base module and corrects the image distortion;
step 3: the camera depth conversion module transmits the image after distortion correction in the step 2 to the rock depth map extraction module, the rock depth map extraction module carries out parallax map calculation on the rock, the original binocular images are fused into one image, and then the image copying module is utilized to backup the parallax map and the fused image;
step 4: the rock depth map extraction module transmits the parallax images in the step 3 to the rock segmentation module for rock segmentation, and the rock segmentation module backs up the segmentation map by using the image copying unit and packages the information of all the rock;
Step 5: the rock segmentation module inputs all the rock information packets obtained in the step 4 into a single rock actual size estimation module, the single rock actual size estimation module obtains configuration information from the basic module, the single rock is respectively subjected to actual size estimation, and the actual size information of each rock is extracted and packed into rock information;
step 6: the single rock actual size estimation module transmits all the rock information packets updated in the step 5 to the rock block statistics module, and the rock block statistics module calculates information of all the rock blocks and re-integrates the information to form a histogram of actual size distribution;
step 7: transmitting all rock information packages updated by the rock segmentation module in the step 5 to the rock block size statistics module, and marking the rock blocks with different sizes according to colors on the fused image by combining the fused image backup and the segmentation map backup obtained in the step 3.
Further, the step 1 includes the following steps:
step 1.1: default camera configuration parameters are imported from a public data storage area of the equipment, and parameter initialization is carried out on the cameras;
step 1.2: importing default configuration lens correction parameters and parallax transformation parameters from a common data storage area of the equipment, and initializing distortion correction and parallax parameters;
Step 1.3: importing display parameters from a public data storage area of the equipment, and initializing a display;
step 1.4: the information display module reports the parameter information, pops up configuration information and consults whether to modify;
step 1.5: if not, directly entering the system, if so, submitting the parameters to a public data storage area, and repeating the steps 1.1-1.4.
Further, the step 2 includes the steps of:
step 2.1: initializing an image acquisition unit, controlling the image acquisition unit to acquire a frame of binocular image, stopping working, storing the image in a storage unit, and submitting the image to a distortion correction unit;
step 2.2: the distortion correction unit obtains the distortion correction parameters initialized in the step 1.2 from the equipment index configuration module; carrying out coordinate distortion correction on all coordinates of the image obtained in the step 2.1, wherein the correction formula is as follows, k1, k2 and k3 are correction parameters, x and y are original coordinates, and r is the pixel Euclidean distance r from the coordinates to the center of the image 2 ＝x 2 +y 2 ：
Step 2.3: the distortion correction unit acquires the corrected coordinates in the step 2.3, performs coordinate remapping on the image obtained in the step 2.1, and stores the corrected image in the memory;
Step 2.4: and the distortion correction unit outputs the corrected image and submits the corrected image to the rock mass depth map extraction module.
Further, the step 3 includes the following steps:
step 3.1: the feature scanning unit obtains a corrected frame of rock block binocular image from the image storage module, and performs edge feature detection on the left view and the right view respectively to serve as an initial feature map;
step 3.2: acquiring a suppression threshold value from the equipment index allocation module, removing all smaller edge features in the initial feature image through judgment of the threshold value, and taking the image after all the smaller edge features are removed as a transverse parallax feature matching guide image;
step 3.3: taking all point coordinates which are not removed in the guide graph obtained in the step 3.2 as effective point coordinates, packaging the effective point coordinates according to rows, submitting the effective point coordinates to an information packet and transmitting the information packet to a feature matching unit;
step 3.4: acquiring the coordinates of each row of points obtained in the step 3.3, inquiring the values of the guide graph of the coordinates, carrying out gradient sequencing on each row, and submitting the sequence values to the information package;
step 3.5: obtaining the maximum detection window and the size of a feature description window from a public data area, extracting the coordinates of each row of detection points from the message packet obtained in the step 3.4, respectively extracting the feature descriptors thereof, carrying out gradient descending matching on the descriptors thereof in the detection window, searching the best matching point thereof, averaging the central color values of the two, putting the central color values into a fusion graph, and submitting the coordinates to a parallax calculation unit;
Step 3.6: making a difference between the best matching point coordinates matched in the step 3.5 and the matching source point coordinates, and putting the value into a parallax image;
step 3.7: after filling all the guide points, performing edge protection filtering, performing linear interpolation on points on the non-guide graph to calculate the depth of the points, putting the points into the parallax graph, and performing linear interpolation filling on the fusion graph;
step 3.8: submitting the parallax map and the fusion map to an image copying module and backing up the parallax map and the fusion map to a memory (8-3);
step 3.9: when the event of displaying the disparity map is triggered, displaying a backup disparity map; and when the event of displaying the fusion map is triggered, displaying the fused image.
Further, the step 4 includes the steps of:
step 4.1: the rock block rough segmentation unit obtains a parallax image from a memory, performs minimum value filtering, and then uses Gaussian filtering to smoothly filter out high-frequency signals and stores the high-frequency signals as a pre-segmentation image;
step 4.2: inputting the pre-segmentation image in the step 4.1 into a rock block rough segmentation unit, solving a gradient map of the rock block rough segmentation unit, acquiring a pre-segmentation threshold value from a public data storage area, performing binary segmentation on the gradient map, extracting a low-value region of the gradient map, and calculating a region mean value as an initial mean value to store the region mean value;
Step 4.3: performing boundary depth decreasing merging on the images extracted in the step 4.2, merging the smaller boundary into the region every time, and re-calculating the average value of the region until the images are completely segmented to obtain a rough segmentation binary image, and backing up the rough segmentation binary image into a memory through an image copying module;
step 4.4: comparing the final average value of each region with the initial average value, filtering the regions smaller than the rated change value, and extracting the regions larger than the rated change value as rock candidate regions;
step 4.5: carrying out area calculation on the candidate area obtained in the step 4.4, and removing the area less than the rated area as noise to be used as a final segmentation result of the rock mass;
step 4.6: storing the rock mass result obtained in the step 4.5, carrying out boundary tracking to extract contour coordinates and depth values, packaging the contour coordinates and depth values and the results into a rock mass information packet, and submitting the rock mass information packet to a public data storage area;
step 4.7: when the event of displaying the segmentation map is triggered, the backup segmentation map is displayed.
Further, the step 5 includes the steps of:
step 5.1: the outline depth statistics unit imports parallax transformation parameters from the equipment index configuration module, imports the outline coordinates and depth values extracted in the step 4.6 into the parallax coordinate transformation unit, and obtains the actual coordinates of the point under the camera coordinates;
Step 5.2: and 5.1, performing least square straight line fitting on the projection coordinates of the x-z plane obtained by calculation in the step, and storing a straight line equation, wherein the straight line equation is as follows:
z＝kx+b
step 5.3: the contour width statistical unit obtains the rough segmentation binary image obtained by calculation in the step 4.3 from a memory, obtains and records the external rectangular frame of each contour, obtains the occupied area of the interior of the contour, and obtains the depth value z on each depth image of the rock mass by using the image as a mask u The equation obtained in the step 5.2 is utilized to obtain the corresponding bottom depth z d Obtaining the relative depth z of the coordinate xy ＝z u -z d Rock volume was determined by integration:
wherein f x And f y Is the resolution at that location.
Step 5.4: and (3) taking the volume multiplied by twice obtained in the step (5.3) as the block degree, packaging the block degree, the external rectangular frame and the outline area into a rock information packet, and submitting the rock information packet to a public data storage area.
Further, the step 6 includes the steps of:
the block extraction unit obtains the volume of each rock block from the information packet, inputs the volume into the block distribution statistics unit, counts the frequency distribution histogram, normalizes the volume, inputs the volume into the information integration unit, integrates the information packet and the histogram into a new packet, and outputs the new packet.
Further, the step 7 includes the steps of:
step 7.1: the block degree extraction unit sends the volume of each rock block to the block degree distribution statistics unit, and statistics is carried out on the volume distribution histogram of all the rock blocks;
step 7.2: the block distribution statistical unit sends the histogram information to the information integration unit, and packages and outputs the information packet submitted in the step 5.4 together with the histogram;
step 7.3: clicking to display the block distribution histogram will display the block distribution histogram of the current image.
The invention has the advantages and beneficial effects that:
the invention uses the basic module to control the hardware part and combine other modules to form an integral system frame, uses the equipment index configuration unit in the basic module to configure equipment information and parameters, organically combines the basic module, the camera depth conversion module, the rock depth map extraction module, the rock segmentation module, the actual size estimation module of the single rock and the rock block size statistics module, integrates the functional modules of each part, and is more convenient to realize the improvement and perfection of the functions of each part. After the system recognizes the rock mass image, a user can view and display the distribution information of rock mass blocks through the controller. According to the rock block analysis method, the binocular vision technology is adopted for rock block analysis, so that the stability of rock block analysis can be ensured. The invention adopts automatic monitoring equipment, wireless transmission technology, non-contact video measurement technology and image vision algorithm technology, and realizes real-time acquisition and feedback analysis of data and images. The invention provides the technology based on automatic monitoring, digital technology, binocular vision image processing and the like for real-time image data acquisition and rock mass distribution analysis, has good real-time performance and reliability, and is favorable for real-time detection of on-site rock mass distribution.
Drawings
FIG. 1 is a block diagram of a basic module structure;
FIG. 2 is a block diagram of a camera depth conversion module;
FIG. 3 is a block diagram of a block depth map extraction module;
FIG. 4 is a block diagram of a block segmentation module;
FIG. 5 is a block diagram of a block size estimation module;
FIG. 6 is a block diagram of a rock mass statistics module;
FIG. 7 is a block diagram of a data flow structure of an automatic analysis device for rock burst block sizes based on binocular vision;
FIG. 8 is a schematic diagram of a binocular vision-based rock mass block degree automatic analysis device;
fig. 9 is a schematic diagram of a disparity map calculation flow provided by the binocular vision-based rock mass blasting block size automatic analysis method;
fig. 10 is a flow chart of a rock segmentation algorithm provided by the automatic analysis method of rock burst block size based on binocular vision.
Detailed Description
In order to make the technical scheme and advantages of the present invention more clear, the technical scheme in the embodiment of the present invention is clearly and completely described below with reference to the accompanying drawings in the embodiment of the present invention:
in this embodiment, the volume unit is cubic centimeters (cm) 3 ) If other units are adopted, the equipment index configuration module 1-4 is required to be adopted for configuration and unit conversion.
In the present embodiment, the histogram distribution is made of percentage, and if other units such as the number are used, the device index configuration units 1 to 4 are used for configuration and unit conversion.
As shown in fig. 8, the automatic rock burst block size analysis device based on binocular vision of the invention comprises a camera 8-1, a processor 8-2, a memory 8-3, a display 8-5 and a controller 8-6, wherein the processor 8-2 comprises a base module 1, a camera depth conversion module 2, a rock block depth map extraction module 3, a rock block segmentation module 4, a single rock block actual size estimation module 5, a rock block size statistics module 6 and a public data storage area 7 as shown in fig. 7. As shown in fig. 1, the base module 1 includes a binocular image acquisition module 1-1, an image storage module 1-2, an image copying module 1-3, an information display module 1-5, an equipment index configuration module 1-4 and an equipment control module 1-6, and is used for image acquisition, caching, control and configuration of equipment base information; the binocular image acquisition module 1-1 receives an image obtained by the camera 8-1 and stores the image into the memory 8-3, the image storage module 1-2 is connected with the memory 8-3, the information display module 1-5 is connected with the display 8-5, the equipment control module 1-6 is connected with the controller 8-6, the image copying module 1-3 is connected with the base module, the camera depth conversion module, the rock mass depth map extraction module, the rock mass segmentation module, the single rock mass actual size estimation module and the rock mass statistics module and is connected with the memory 8-3 in the base module, and the equipment index configuration module 1-4 is respectively connected with the camera 8-1, the memory 8-3, the display 8-5 and the controller 8-6 and is used for controlling data modes input or output by equipment and system related parameters, units and indexes;
The image storage module is connected with the camera depth conversion module, the image storage module is connected with the rock depth map extraction module, the image storage module is connected with the rock segmentation module, and the image storage module and the equipment index configuration module are respectively connected with the single rock actual size estimation module; the information display module is respectively connected with the camera depth conversion module, the rock depth map extraction module, the rock segmentation module, the single rock actual size estimation module and the rock block size statistics module and is used for providing display events for the modules.
The distortion correction unit of the camera depth conversion module is connected with the image storage module and is used for correcting distortion of the acquired original image; the camera depth conversion module is connected with the rock block segmentation module; one output end of the camera depth conversion module is connected with the rock mass depth map extraction module, and one input end of the camera depth conversion module is connected with the single rock mass actual size estimation module; one output end of the camera depth conversion module is connected with the image copying module, and one output end of the rock depth map extraction module is connected with the information display module and is used for providing a display event and displaying a parallax map obtained by calculating a current binocular image;
The device comprises a rock depth map extraction module, an image storage module, a rock segmentation module, a camera depth conversion module, a parallax map and an information display module, wherein the rock depth map extraction module is connected with the image storage module, one output end of the rock depth map extraction module is connected with the rock segmentation module, one input end of the rock depth map extraction module is connected with one output end of the camera depth conversion module, and one output end of the rock depth map extraction module is connected with the information display module and is used for providing a display event and displaying the parallax map obtained by calculating the current binocular image.
The rock mass segmentation module is connected with the image storage module, one input end of the rock mass segmentation module is connected with one output end of the rock mass depth map extraction module, one output end of the rock mass segmentation module is connected with one input end of the single rock mass actual size estimation module, the rock mass segmentation module is connected with the rock mass statistics module, and one output end of the rock mass segmentation module is connected with the information display module and used for providing display events and displaying actual segmentation conditions of the rock mass.
The single rock actual size estimation module is connected with the rock block size statistics module and the camera depth conversion module and is used for acquiring parameters of an image coordinate to a world coordinate; an input end of the single rock actual size estimation module is connected with an output end of the rock segmentation module and is used for receiving single rock information to carry out actual size estimation; and an output end of the actual size estimation of the single rock mass is connected with the information display module and is used for providing a display event and displaying the actual size estimation condition of the current rock mass.
The rock block size statistics module is connected with the image storage module and is used for receiving all rock block information to carry out size distribution statistics; the rock block size statistics module is connected with the single rock block actual size estimation module and is used for calculating the actual size information of the single rock block; the rock block size statistics module is also connected with the information display module and is used for providing display events and displaying the actual size distribution condition of all the current rock blocks.
The public data storage area is respectively connected with the camera depth conversion module, the rock depth map extraction module, the rock segmentation module, the single rock actual size estimation module and the rock block size statistics module and is used for storing fracture characteristic data information packets submitted by the modules, and the public data storage area is also connected with the basic module and is used for storing and initializing default parameters of solidified equipment.
As shown in fig. 2, the camera depth conversion module includes a distortion correction unit 2-1, a parallax coordinate transformation unit 2-2, and a world coordinate calculation unit 2-3; the distortion correcting unit is connected with the rock mass depth map extracting module,
the distortion correction unit is used for inputting image data, correcting an initial distortion image, and outputting corrected data to the rock depth map module for pixel-level depth estimation;
The parallax coordinate transformation unit is connected with the world coordinate calculation unit and is used for inputting parallax and image coordinates; the world coordinate calculation unit is used for outputting world coordinates;
as shown in fig. 3, the rock depth map extraction module includes a feature scanning unit 3-1, a feature matching unit 3-2, and a parallax calculation unit 3-3; the characteristic scanning unit, the characteristic matching unit and the parallax calculating unit are sequentially connected and used for providing a data transmission channel; the characteristic scanning unit is used for inputting binocular image data and outputting a parallax image, and the parallax calculating unit is used for calculating a parallax value which is optimally matched; the feature matching unit is used for calculating the matching degree of the feature data and collecting the parallax value returned by the parallax calculation unit to construct a parallax map; the parallax calculating unit is connected with the feature matching unit and is used for calculating the parallax value of the best matching feature; the feature matching unit is connected with the rock segmentation module and transmits the extracted parallax image to the rock segmentation module for subsequent segmentation.
As shown in fig. 4, the rock mass segmentation module comprises a rock mass rough segmentation unit 4-1, a noise filtering unit 4-2 and a rock mass integration unit 4-3; the rock block rough segmentation unit is connected with the rock block integration unit through the noise filtering unit, the rock block rough segmentation unit is used for inputting depth image data, the noise filtering unit is used for filtering noise small areas of non-rock blocks, and the rock block integration unit is used for outputting all rock block information.
As shown in fig. 5, the block actual size estimation module includes a contour depth statistics unit 5-1, a contour width statistics unit 5-2, and an actual size calculation unit 5-3; the profile depth statistical unit is connected with the actual size calculation unit through the profile width statistical unit; the contour depth statistics unit is used for inputting single rock block information, and the actual size calculation unit is used for outputting the actual size of the rock block.
As shown in fig. 6, the rock block size statistics module includes a block size extraction unit 6-1, a block size distribution statistics unit 6-2, and an information integration unit 6-3; the block degree extraction unit, the block degree distribution statistical unit and the information integration unit are sequentially connected, the block degree extraction unit is used for inputting single rock block information, and the information integration unit is used for outputting rock block distribution information.
An analysis method of a rock mass block degree automatic analysis device based on binocular vision comprises the following steps:
step 1: according to the required parameters, equipment index configuration and equipment initialization are carried out, wherein the parameters comprise the size of an image acquired by a camera and a lens distortion correction coefficient; after configuration is finished, the camera, the display and other devices work normally according to the required parameters; starting a binocular image acquisition module, and acquiring rock images to an image storage module according to a corresponding format;
Step 2: the image storage module transmits the rock block image acquired by the camera in the step 1 to the camera depth conversion module, and the camera depth conversion module acquires configuration information from the base module and corrects the image distortion;
step 3: the camera depth conversion module transmits the image subjected to distortion correction in the step 2 to the rock depth map extraction module, the rock depth map extraction module carries out parallax map calculation on the rock, the original binocular images are fused into one image, and then the image copying module is utilized to back up the fused image;
step 4: the rock depth map extraction module transmits the parallax images after distortion correction in the step 3 to the rock segmentation module for rock segmentation, and the rock segmentation module backs up the segmentation map by using the image copying unit and packages the information of all the rock;
step 5: the rock segmentation module inputs all the rock information packets obtained in the step 4 into a single rock actual size estimation module, the single rock actual size estimation module obtains configuration information from the basic module, the single rock is respectively subjected to actual size estimation, and the actual size information of each rock is extracted and packed into rock information;
Step 6: the single rock actual size estimation module transmits all the rock information packets updated in the step 5 to the rock block statistics module, and the rock block statistics module calculates information of all the rock blocks and re-integrates the information to form a histogram of actual size distribution;
step 7: transmitting all rock information packages updated by the rock segmentation module in the step 5 to the rock block statistics module, and marking the rock blocks with different sizes according to colors on the fused image by combining the fused image backup and the segmentation map backup obtained in the step 3.
The step 1 comprises the following steps:
step 1.1: default camera configuration parameters are imported from a public data storage area of the equipment, and parameter initialization is carried out on the cameras;
step 1.2: importing default configuration lens correction parameters and parallax transformation parameters from a common data storage area of the equipment, and initializing distortion correction and parallax parameters;
step 1.3: importing display parameters from a public data storage area of the equipment, and initializing a display;
step 1.4: the information display module reports the parameter information, pops up configuration information and consults whether to modify;
Step 1.5: if not, directly entering the system, if so, submitting the parameters to a public data storage area, and repeating the steps 1.1-1.4.
The step 2 comprises the following steps:
step 2.1: initializing an image acquisition unit, controlling the image acquisition unit to acquire a frame of binocular image, stopping working, storing the image in a storage unit 1-2, and submitting the image to a distortion correction unit 2-1;
step 2.2: the distortion correction unit obtains the distortion correction parameters initialized in the step 1.2 from the equipment index configuration module; carrying out coordinate distortion correction on all coordinates of the image obtained in the step 2.1, wherein the correction formula is as follows, k1, k2 and k3 are correction parameters, x and y are original coordinates, and r is the pixel Euclidean distance r from the coordinates to the center of the image 2 ＝x 2 +y 2 ：
Step 2.3: the distortion correction unit acquires the corrected coordinates in the step 2.3, performs coordinate remapping on the image obtained in the step 2.1, and stores the corrected image in the memory 8-3;
step 2.4: and the distortion correction unit outputs the corrected image and submits the corrected image to the rock mass depth map extraction module.
As shown in fig. 9, the step 3 includes the steps of:
step 3.1: the feature scanning unit 3-1 obtains a frame of corrected rock binocular image from the image storage module 1-2, and performs edge feature detection on the left view and the right view respectively to serve as an initial feature map;
Step 3.2: acquiring a suppression threshold value from the equipment index allocation module, removing all smaller edge features in the initial feature image through judgment of the threshold value, and taking the image after all the smaller edge features are removed as a transverse parallax feature matching guide image;
step 3.3: taking all point coordinates which are not removed in the guide graph obtained in the step 3.2 as effective point coordinates, packaging the effective point coordinates according to rows, submitting the effective point coordinates to an information packet and transmitting the information packet to a feature matching unit;
step 3.4: acquiring the coordinates of each row of points obtained in the step 3.3, inquiring the values of the guide graph of the coordinates, carrying out gradient sequencing on each row, and submitting the sequence values to the information package;
step 3.5: obtaining the maximum detection window and the size of a feature description window from a public data area, extracting the coordinates of each row of detection points from the message packet obtained in the step 3.4, respectively extracting the feature descriptors thereof, carrying out gradient descending matching on the descriptors thereof in the detection window, searching the best matching point thereof, averaging the central color values of the two, putting the central color values into a fusion graph, and submitting the coordinates to a parallax calculation unit;
step 3.6: making a difference between the best matching point coordinates matched in the step 3.5 and the matching source point coordinates, and putting the value into a parallax image;
Step 3.7: after filling all the guide points, performing edge protection filtering, performing linear interpolation on points on the non-guide graph to calculate the depth of the points, putting the points into the parallax graph, and performing linear interpolation filling on the fusion graph;
step 3.8: submitting the parallax map and the fusion map to an image copying module and backing up the parallax map and the fusion map to a memory 8-3;
step 3.9: when the event of displaying the disparity map is triggered, displaying a backup disparity map; and when the event of displaying the fusion map is triggered, displaying the fused image.
As shown in fig. 10, the step 4 includes the steps of:
step 4.1: the rock block rough segmentation unit obtains a parallax image from the memory 8-3, performs minimum value filtering, and then uses Gaussian filtering to smoothly filter out high-frequency signals and stores the high-frequency signals as a pre-segmentation image;
step 4.2: inputting the pre-segmentation image in the step 4.1 into a rock block rough segmentation unit, solving a gradient map of the rock block rough segmentation unit, acquiring a pre-segmentation threshold value from a public data storage area, performing binary segmentation on the gradient map, extracting a low-value region of the gradient map, and calculating a region mean value as an initial mean value to store the region mean value;
step 4.3: performing boundary depth decreasing merging on the images extracted in the step 4.2, merging the smaller boundary into the region every time, and re-calculating the average value of the region until the images are completely segmented to obtain a rough segmentation binary image, and backing up the rough segmentation binary image into a memory 8-3 through an image copying module;
Step 4.4: comparing the final average value of each region with the initial average value, filtering the regions smaller than the rated change value, and extracting the regions larger than the rated change value as rock candidate regions;
step 4.5: and (4) carrying out area calculation on the candidate region obtained in the step (4.4), and removing the region with less than the rated area as noise to obtain a rock final segmentation result.
Step 4.6: and (3) saving the rock mass result obtained in the step (4.5), carrying out boundary tracking on the rock mass result, extracting contour coordinates and depth values, packaging the rock mass result and the result into a rock mass information package, and submitting the rock mass information package to a public data storage area.
Step 4.7: when the event of displaying the segmentation map is triggered, displaying a backup segmentation map;
said step 5 comprises the steps of:
step 5.1: the outline depth statistics unit imports parallax transformation parameters from the equipment index configuration module, imports the outline coordinates and depth values extracted in the step 4.6 into the parallax coordinate transformation unit, and obtains the actual coordinates of the point under the camera coordinates;
step 5.2: and 5.1, performing least square straight line fitting on the projection coordinates of the x-z plane obtained by calculation in the step, and storing a straight line equation, wherein the straight line equation is as follows:
z＝kx+b
step 5.3: the contour width statistical unit obtains the rough segmentation binary image obtained by calculation in the step 4.3 from the memory 8-3, and obtains the external connection of each contour Rectangular frames are recorded, the occupied area of the inner part of the outline is obtained, and the depth value z of each depth map of the rock mass is obtained by using the map as a mask u The equation obtained in the step 5.2 is utilized to obtain the corresponding bottom depth z d Obtaining the relative depth z of the coordinate xy ＝z u -z d Rock volume was determined by integration:
wherein f x And f y Is the resolution at that location.
Step 5.4: and (3) taking the volume multiplied by twice obtained in the step (5.3) as the block degree, packaging the block degree, the external rectangular frame and the outline area into a rock information packet, and submitting the rock information packet to a public data storage area.
The step 6 further comprises the following steps:
the block extraction unit (5-1) obtains the volume of each rock block from the information packet, inputs the volume into the block distribution statistics unit (2-2), performs frequency distribution histogram statistics, normalizes the volume, inputs the volume into the information integration unit (6-3), integrates the information packet and the histogram into a new packet, and outputs the new packet.
The step 7 comprises the following steps:
step 7.1: the block degree extraction unit sends the volume of each rock block to the block degree distribution statistics unit, and statistics is carried out on the volume distribution histogram of all the rock blocks;
step 7.2: the block distribution statistical unit sends the histogram information to the information integration unit, and packages and outputs the information packet submitted in the step 5.4 together with the histogram;
Step 7.3: clicking to display the block distribution histogram will display the block distribution histogram of the current image.
The foregoing is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art, who is within the scope of the present invention, should make equivalent substitutions or modifications according to the technical scheme of the present invention and the inventive concept thereof, and should be covered by the scope of the present invention.
Claims (10)
1. The utility model provides a rock burst block degree automatic analysis device based on binocular vision, has camera (8-1), processor (8-2), memory (8-3), display (8-5) and controller (8-6), its characterized in that: the processor (8-2) comprises a basic module (1), a camera depth conversion module (2), a rock depth map extraction module (3), a rock segmentation module (4), a single rock actual size estimation module (5), a rock block size statistics module (6) and a public data storage area (7); the basic module (1) comprises a binocular image acquisition module (1-1), an image storage module (1-2), an image copying module (1-3), an information display module (1-5), an equipment index configuration module (1-4) and an equipment control module (1-6); the binocular image acquisition module (1-1) receives an image obtained by the camera (8-1) and stores the image into the memory (8-3), the image storage module (1-2) is connected with the memory (8-3), the information display module (1-5) is connected with the display (8-5), the equipment control module (1-6) is connected with the controller (8-6), the image copying module (1-3) is connected with the base module, the camera depth conversion module, the rock mass depth map extraction module, the rock mass segmentation module, the single rock mass actual size estimation module and the rock mass statistics module, and is connected with the memory (8-3) in the base module, and the equipment index configuration module (1-4) is respectively connected with the camera (8-1), the memory (8-3), the display (8-5) and the controller (8-6) and is used for controlling data modes input or output by equipment and relevant parameters, units and indexes of a system; the image storage module (1-2) is respectively connected with the camera depth conversion module, the rock depth map extraction module and the rock segmentation module, and the image storage module and the equipment index configuration module are respectively connected with the single rock actual size estimation module; the information display modules (1-5) are respectively connected with the camera depth conversion module, the rock depth map extraction module, the rock segmentation module, the single rock actual size estimation module and the rock block size statistics module and are used for providing display events for the modules;
The distortion correction unit of the camera depth conversion module is connected with the image storage module and is used for correcting distortion of the acquired original image; the camera depth conversion module is connected with the rock block segmentation module; one output end of the camera depth conversion module is connected with the rock mass depth map extraction module, and one input end of the camera depth conversion module is connected with the single rock mass actual size estimation module; one output end of the camera depth conversion module is connected with the image copying module, and one output end of the rock depth map extraction module is connected with the information display module and is used for providing a display event and displaying a parallax map obtained by calculating a current binocular image;
the device comprises a rock depth map extraction module, an image storage module, a rock segmentation module, a camera depth conversion module, a display event display module and a parallax map, wherein the rock depth map extraction module is connected with the image storage module, one output end of the rock depth map extraction module is connected with the rock segmentation module, one input end of the rock depth map extraction module is connected with one output end of the camera depth conversion module, and one output end of the rock depth map extraction module is connected with the information display module and is used for providing a display event and displaying the parallax map obtained by current binocular image calculation;
the rock block segmentation module is connected with the image storage module, one input end of the rock block segmentation module is connected with one output end of the rock block depth map extraction module, one output end of the rock block segmentation module is connected with one input end of the single rock block actual size estimation module, the rock block segmentation module is connected with the rock block size statistics module, and one output end of the rock block segmentation module is connected with the information display module and is used for providing display events and displaying actual segmentation conditions of the rock blocks;
The single rock actual size estimation module is connected with the image storage module and the equipment index configuration module respectively, is connected with the rock block size statistics module and is connected with the camera depth conversion module and is used for acquiring parameters from image coordinates to world coordinates; an input end of the single rock actual size estimation module is connected with an output end of the rock segmentation module and is used for receiving single rock information to carry out actual size estimation; an output end of the single rock mass actual size estimation module is connected with the information display module and is used for providing a display event and displaying the actual size estimation condition of the current rock mass;
the rock block size statistics module is respectively connected with the image storage module and the rock block segmentation module and is used for receiving all rock block information to carry out size distribution statistics; the rock block size statistics module is connected with the single rock block actual size estimation module and is used for calculating the actual size information of the single rock block; the rock block size statistics module is connected with the information display module and is used for providing display events and displaying the actual size distribution conditions of all the current rock blocks;
The public data storage area is respectively connected with the camera depth conversion module, the rock depth map extraction module, the rock segmentation module, the single rock actual size estimation module and the rock block size statistics module and is used for storing fracture characteristic data information packets submitted by the modules, and the public data storage area is also connected with the basic module and is used for storing and initializing default parameters of solidified equipment.
2. The binocular vision-based rock mass blasting block automatic analysis device of claim 1, wherein: the camera depth conversion module comprises a distortion correction unit (2-1), a parallax coordinate transformation unit (2-2) and a world coordinate calculation unit (2-3); the distortion correcting unit is connected with the rock mass depth map extracting module,
the distortion correction unit is used for inputting image data, correcting an initial distortion image, and outputting corrected image data to the rock depth map module for pixel-level depth estimation;
the parallax coordinate transformation unit is connected with the world coordinate calculation unit and is used for inputting parallax and image coordinates; the world coordinate calculation unit is used for outputting world coordinates;
The rock depth map extraction module comprises a characteristic scanning unit (3-1), a characteristic matching unit (3-2) and a parallax calculation unit (3-3); the characteristic scanning unit, the characteristic matching unit and the parallax calculating unit are sequentially connected and used for providing a data transmission channel; the characteristic scanning unit is used for inputting binocular image data and outputting a parallax image, and the parallax calculating unit is used for calculating a parallax value which is optimally matched; the feature matching unit is used for calculating the matching degree of the feature data and collecting the parallax value returned by the parallax calculation unit to construct a parallax map; the parallax calculating unit is connected with the feature matching unit and is used for calculating the parallax value of the best matching feature; the feature matching unit is connected with the rock segmentation module and transmits the extracted parallax image to the rock segmentation module for subsequent segmentation;
the rock block segmentation module comprises a rock block rough segmentation unit (4-1), a noise filtering unit (4-2) and a rock block integration unit (4-3); the rock block rough segmentation unit is connected with the rock block integration unit through the noise filtering unit, the rock block rough segmentation unit is used for inputting depth image data, the noise filtering unit is used for filtering noise small areas of non-rock blocks, and the rock block integration unit is used for outputting all rock block information;
The single rock actual size estimation module comprises a contour depth statistics unit (5-1), a contour width statistics unit (5-2) and an actual size calculation unit (5-3); the profile depth statistical unit is connected with the actual size calculation unit through the profile width statistical unit; the contour depth statistics unit is used for inputting single rock block information, and the actual size calculation unit is used for outputting the actual size of the rock block;
the rock block size statistics module comprises a block size extraction unit (6-1), a block size distribution statistics unit (6-2) and an information integration unit (6-3); the block degree extraction unit, the block degree distribution statistical unit and the information integration unit are sequentially connected, the block degree extraction unit is used for inputting single rock block information, and the information integration unit is used for outputting rock block distribution information.
3. The analysis method of a binocular vision-based rock mass block automatic analysis device according to claim 1, characterized by comprising the steps of:
step 1: according to the required parameters, equipment index configuration and equipment initialization are carried out, wherein the parameters comprise the size of an image acquired by a camera and a lens distortion correction coefficient; after configuration is finished, the camera, the display and other devices work normally according to the required parameters; starting a binocular image acquisition module, and acquiring rock images to an image storage module according to a corresponding format;
Step 2: the image storage module transmits the rock block image acquired by the camera in the step 1 to the camera depth conversion module, and the camera depth conversion module acquires configuration information from the base module and corrects the image distortion;
step 3: the camera depth conversion module transmits the image after distortion correction in the step 2 to the rock depth map extraction module, the rock depth map extraction module carries out parallax map calculation on the rock, the original binocular images are fused into one image, and then the image copying module is utilized to backup the parallax map and the fused image;
step 4: the rock depth map extraction module transmits the parallax images in the step 3 to the rock segmentation module for rock segmentation, and the rock segmentation module backs up the segmentation map by using the image copying unit and packages the information of all the rock;
step 5: the rock segmentation module inputs all the rock information packets obtained in the step 4 into a single rock actual size estimation module, the single rock actual size estimation module obtains configuration information from the basic module, the single rock is respectively subjected to actual size estimation, and the actual size information of each rock is extracted and packed into rock information;
Step 6: the single rock actual size estimation module transmits all the rock information packets updated in the step 5 to the rock block statistics module, and the rock block statistics module calculates information of all the rock blocks and re-integrates the information to form a histogram of actual size distribution;
step 7: transmitting all rock information packages updated by the rock segmentation module in the step 5 to the rock block size statistics module, and marking the rock blocks with different sizes according to colors on the fused image by combining the fused image backup and the segmentation map backup obtained in the step 3.
4. A method of analyzing a rock mass block automatic analyzer based on binocular vision according to claim 3, wherein the step 1 comprises the steps of:
step 1.1: default camera configuration parameters are imported from a public data storage area of the equipment, and parameter initialization is carried out on the cameras;
step 1.2: importing default configuration lens correction parameters and parallax transformation parameters from a common data storage area of the equipment, and initializing distortion correction and parallax parameters;
step 1.3: importing display parameters from a public data storage area of the equipment, and initializing a display;
Step 1.4: reporting the parameter information to an information display module, popping up configuration information and consulting whether to modify;
step 1.5: if not, directly entering the system, if so, submitting the parameters to a public data storage area, and repeating the steps 1.1-1.4.
5. The method of analyzing a rock mass block automatic analyzer based on binocular vision according to claim 4, wherein the step 2 comprises the steps of:
step 2.1: initializing an image acquisition unit, controlling the image acquisition unit to acquire a frame of binocular image, stopping working, storing the image in an image storage module (1-2), and submitting the image to a distortion correction unit (2-1);
step 2.2: the distortion correction unit obtains the distortion correction parameters initialized in the step 1.2 from the equipment index configuration module; carrying out coordinate distortion correction on all coordinates of the image obtained in the step 2.1, wherein the correction formula is as follows, k1, k2 and k3 are correction parameters, x and y are original coordinates, and r is the pixel Euclidean distance r from the coordinates to the center of the image 2 ＝x 2 +y 2 ：
Step 2.3: the distortion correction unit acquires the corrected coordinates in the step 2.2, performs coordinate remapping on the image obtained in the step 2.1, and stores the corrected image in the memory (8-3);
Step 2.4: and the distortion correction unit outputs the corrected image and submits the corrected image to the rock mass depth map extraction module.
6. A method of analyzing a binocular vision rock mass block automatic analyzing apparatus according to claim 3, wherein the step 3 comprises the steps of:
step 3.1: the feature scanning unit (3-1) obtains a frame of corrected rock binocular image from the image storage module (1-2), and performs edge feature detection on the left view and the right view respectively to serve as an initial feature map;
step 3.2: acquiring a suppression threshold value from the equipment index allocation module, removing all smaller edge features in the initial feature image through judgment of the threshold value, and taking the image after all the smaller edge features are removed as a transverse parallax feature matching guide image;
step 3.3: taking all point coordinates which are not removed in the guide graph obtained in the step 3.2 as effective point coordinates, packaging the effective point coordinates according to rows, submitting the effective point coordinates to an information packet and transmitting the information packet to a feature matching unit;
step 3.4: acquiring the coordinates of each row of points obtained in the step 3.3, inquiring the values of the guide graph, sequencing the gradient of each row, and submitting the sequence values to the information packet;
step 3.5: obtaining the maximum detection window and the size of a feature description window from a public data area, extracting the coordinates of each row of detection points from the message packet obtained in the step 3.4, respectively extracting the feature descriptors thereof, carrying out gradient descending matching on the descriptors thereof in the detection window, searching the best matching point thereof, averaging the central color values of the two, putting the central color values into a fusion graph, and submitting the coordinates to a parallax calculation unit;
Step 3.6: making a difference between the best matching point coordinates matched in the step 3.5 and the matching source point coordinates, and putting the value into a parallax image;
step 3.7: after the filling of the guide points is finished, edge protection filtering is carried out, linear interpolation is carried out on the points on the non-guide graph to calculate the depth of the points, the points are put into the parallax graph, and linear interpolation filling is also carried out on the fusion graph of the points;
step 3.8: submitting the parallax map and the fusion map to an image copying module and backing up the parallax map and the fusion map to a memory (8-3);
step 3.9: when the event of displaying the disparity map is triggered, displaying a backup disparity map; and when the event of displaying the fusion map is triggered, displaying the fused image.
7. A method of analyzing a binocular vision type rock mass block automatic analyzing apparatus according to claim 3, wherein the step 4 comprises the steps of:
step 4.1: the rock block rough segmentation unit obtains a parallax image from a memory (8-3), performs minimum value filtering, and then uses Gaussian filtering to smoothly filter out high-frequency signals and stores the high-frequency signals as a pre-segmentation image;
step 4.2: inputting the pre-segmentation image in the step 4.1 into a rock block rough segmentation unit, solving a gradient map of the rock block rough segmentation unit, acquiring a pre-segmentation threshold value from a public data storage area, performing binary segmentation on the gradient map, extracting a low-value region of the gradient map, and calculating a region mean value as an initial mean value to store the region mean value;
Step 4.3: performing boundary depth decreasing combination on the low-value region of the gradient map extracted in the step 4.2, taking out the smaller one of the boundaries each time, combining the smaller one of the boundaries on the region, and recalculating the average value of the region until the image is completely segmented to obtain a rough segmentation binary map, and backing up the rough segmentation binary map into a memory (8-3) through an image copying module;
step 4.4: comparing the final average value of each region with the initial average value, filtering the regions smaller than the rated change value, and extracting the regions larger than the rated change value as rock candidate regions;
step 4.5: carrying out area calculation on the candidate area obtained in the step 4.4, and removing the area less than the rated area as noise to be used as a final segmentation result of the rock mass;
step 4.6: storing the rock mass result obtained in the step 4.5, carrying out boundary tracking to extract contour coordinates and depth values, packaging the contour coordinates and depth values and the results into a rock mass information packet, and submitting the rock mass information packet to a public data storage area;
step 4.7: when the event of displaying the segmentation map is triggered, the backup segmentation map is displayed.
8. The method of analyzing a binocular vision type rock mass block automatic analyzing apparatus according to claim 7, wherein the step 5 comprises the steps of:
Step 5.1: the outline depth statistics unit imports parallax transformation parameters from the equipment index configuration module, imports the outline coordinates and depth values extracted in the step 4.6 into the parallax coordinate transformation unit, and obtains the actual coordinates of the point under the camera coordinates;
step 5.2: and 5.1, performing least square straight line fitting on the projection coordinates of the x-z plane obtained by calculation in the step, and storing a straight line equation, wherein the straight line equation is as follows:
z＝kx+b
step 5.3: the contour width statistical unit obtains the rough segmentation binary image obtained by calculation in the step 4.3 from a memory (8-3), obtains and records the external rectangular frame of each contour, obtains the occupied area of the inside of the contour, and obtains the depth value z on each depth image of the rock mass by using the image as a mask u The equation obtained in the step 5.2 is utilized to obtain the corresponding bottom depth z d Obtaining the relative depth z of the coordinate xy ＝z u -z d Rock volume was determined by integration:
wherein f x And f y Is the resolution at that location;
step 5.4: and (3) taking the volume multiplied by twice obtained in the step (5.3) as the block degree, packaging the block degree, the external rectangular frame and the outline area into a rock information packet, and submitting the rock information packet to a public data storage area.
9. The method of analyzing a binocular vision type rock mass block automatic analyzing apparatus according to claim 7, wherein the step 6 comprises the steps of:
The block extraction unit (6-1) obtains the volume of each rock block from the information packet, inputs the volume into the block distribution statistics unit (6-2), counts the frequency distribution histogram, normalizes the frequency distribution histogram, inputs the normalized frequency distribution histogram into the information integration unit (6-3), integrates the information packet and the histogram into a new packet, and outputs the new packet.
10. The method of analyzing a binocular vision type rock mass block automatic analyzing apparatus according to claim 8, wherein the step 7 comprises the steps of:
step 7.1: the block degree extraction unit sends the volume of each rock block to the block degree distribution statistics unit, and statistics is carried out on the volume distribution histogram of all the rock blocks;
step 7.2: the block distribution statistical unit sends the histogram information to the information integration unit, and packages and outputs the information packet submitted in the step 5.4 together with the histogram;
step 7.3: clicking to display the block distribution histogram will display the block distribution histogram of the current image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110600753.6A CN113344945B (en) | 2021-05-31 | 2021-05-31 | Automatic rock mass blasting block size analysis device and method based on binocular vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110600753.6A CN113344945B (en) | 2021-05-31 | 2021-05-31 | Automatic rock mass blasting block size analysis device and method based on binocular vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113344945A CN113344945A (en) | 2021-09-03 |
| CN113344945B true CN113344945B (en) | 2024-04-09 |
Family
ID=77472670
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110600753.6A Active CN113344945B (en) | 2021-05-31 | 2021-05-31 | Automatic rock mass blasting block size analysis device and method based on binocular vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113344945B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114202569B (en) * | 2021-11-29 | 2024-12-10 | 内蒙古工业大学 | Method, device, equipment and medium for determining original rock mass fragmentation size |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018086348A1 (en) * | 2016-11-09 | 2018-05-17 | 人加智能机器人技术（北京）有限公司 | Binocular stereo vision system and depth measurement method |
| CN108073774A (en) * | 2017-12-29 | 2018-05-25 | 辽宁科技大学 | A kind of reliable method for verifying quick-fried heap LUMPINESS DISTRIBUTION |
| CN108305263A (en) * | 2018-01-31 | 2018-07-20 | 中国科学院武汉岩土力学研究所 | A kind of image statistics method of rock fragmentation after explosion |
| CN108645299A (en) * | 2018-05-03 | 2018-10-12 | 中国葛洲坝集团易普力股份有限公司 | Rock Blasting Fragmentation analysis method based on Particle Vibration Velocity |
| CN109544624A (en) * | 2018-11-09 | 2019-03-29 | 北方爆破科技有限公司 | A kind of rock fragmentation image analysis system |
-
2021
- 2021-05-31 CN CN202110600753.6A patent/CN113344945B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018086348A1 (en) * | 2016-11-09 | 2018-05-17 | 人加智能机器人技术（北京）有限公司 | Binocular stereo vision system and depth measurement method |
| CN108073774A (en) * | 2017-12-29 | 2018-05-25 | 辽宁科技大学 | A kind of reliable method for verifying quick-fried heap LUMPINESS DISTRIBUTION |
| CN108305263A (en) * | 2018-01-31 | 2018-07-20 | 中国科学院武汉岩土力学研究所 | A kind of image statistics method of rock fragmentation after explosion |
| CN108645299A (en) * | 2018-05-03 | 2018-10-12 | 中国葛洲坝集团易普力股份有限公司 | Rock Blasting Fragmentation analysis method based on Particle Vibration Velocity |
| CN109544624A (en) * | 2018-11-09 | 2019-03-29 | 北方爆破科技有限公司 | A kind of rock fragmentation image analysis system |
Non-Patent Citations (1)
| Title |
|---|
| 矿山爆破质量定量评价系统研究;伊志宣;郭连军;李超亮;张大宁;赵楠楠;石文东;李科;;工程爆破(第03期);25-28 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113344945A (en) | 2021-09-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107578390B (en) | Method and device for correcting image white balance by using neural network | |
| CN109584170B (en) | Underwater image restoration method based on convolutional neural network | |
| CN107316326B (en) | Edge-based disparity map calculation method and device applied to binocular stereo vision | |
| EP2983131A1 (en) | Method and device for camera calibration | |
| CN112200771B (en) | Height measurement method, device, equipment and medium | |
| CN111582072A (en) | Substation picture bird's nest detection method combined with ResNet50+FPN+DCN | |
| CN102867295B (en) | A kind of color correction method for color image | |
| CN110763306B (en) | Monocular vision-based liquid level measurement system and method | |
| CN103679737B (en) | Method for color image edge detection on basis of multichannel information selection | |
| CN105224908A (en) | A kind of roadmarking acquisition method based on orthogonal projection and device | |
| CN104182757A (en) | Method of acquiring actual coverage area of measured target and device | |
| CN107256421B (en) | A rapid counting method for rice and wheat grains | |
| CN115661187B (en) | Image enhancement method for analysis of traditional Chinese medicine preparation | |
| CN110310301B (en) | A method and device for detecting a target object | |
| CN113344945B (en) | Automatic rock mass blasting block size analysis device and method based on binocular vision | |
| CN102333234B (en) | A monitoring method and device for binocular stereoscopic video state information | |
| CN110838148A (en) | Binocular camera self-calibration method | |
| CN106599776A (en) | People counting method based on trajectory analysis | |
| CN111630569B (en) | Binocular matching method, visual imaging device and device with storage function | |
| CN115690107B (en) | High-throughput counting method, system and device for pod fruit grains based on image processing | |
| CN107886540B (en) | Object identification and positioning method in refrigeration equipment and refrigeration equipment | |
| CN101729739A (en) | Method for rectifying deviation of image | |
| CN112819882A (en) | Real-time earth volume calculation method based on binocular vision | |
| CN116524000A (en) | Rose size measurement method and system based on deep learning and stereoscopic vision | |
| CN115546716A (en) | Binocular vision-based method for positioning fire source around power transmission line |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |