CN116824483A - Maximum value sawing algorithm for timber based on optimal solution of permutation and combination - Google Patents
Maximum value sawing algorithm for timber based on optimal solution of permutation and combination Download PDFInfo
- Publication number
- CN116824483A CN116824483A CN202310619170.7A CN202310619170A CN116824483A CN 116824483 A CN116824483 A CN 116824483A CN 202310619170 A CN202310619170 A CN 202310619170A CN 116824483 A CN116824483 A CN 116824483A
- Authority
- CN
- China
- Prior art keywords
- wood
- image
- defect
- sawing
- algorithm
- 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.)
- Pending
Links
- 230000007547 defect Effects 0.000 claims abstract description 104
- 239000002023 wood Substances 0.000 claims abstract description 88
- 238000012549 training Methods 0.000 claims abstract description 13
- 238000009826 distribution Methods 0.000 claims abstract description 10
- 238000000034 method Methods 0.000 claims description 34
- 238000005259 measurement Methods 0.000 claims description 12
- 238000007781 pre-processing Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000012360 testing method Methods 0.000 claims description 8
- 230000009466 transformation Effects 0.000 claims description 8
- 238000012795 verification Methods 0.000 claims description 8
- 238000001914 filtration Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 7
- 238000004458 analytical method Methods 0.000 claims description 6
- 238000005286 illumination Methods 0.000 claims description 5
- 238000012512 characterization method Methods 0.000 claims description 4
- 238000013135 deep learning Methods 0.000 claims description 4
- 238000001514 detection method Methods 0.000 claims description 4
- 238000003708 edge detection Methods 0.000 claims description 4
- 230000009897 systematic effect Effects 0.000 claims description 4
- 238000013519 translation Methods 0.000 claims description 4
- 230000017105 transposition Effects 0.000 claims description 4
- 239000013598 vector Substances 0.000 claims description 4
- 238000004891 communication Methods 0.000 claims description 3
- 238000005520 cutting process Methods 0.000 abstract description 13
- 239000000463 material Substances 0.000 description 11
- 239000002699 waste material Substances 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 238000005498 polishing Methods 0.000 description 2
- 241001425476 Apiosporina morbosa Species 0.000 description 1
- 229920002531 Rubberwood Polymers 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000005336 cracking Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 210000002268 wool Anatomy 0.000 description 1
Landscapes
- Image Processing (AREA)
Abstract
The invention discloses a sawing algorithm of maximum value of wood based on an optimal solution of permutation and combination, which relates to the technical field of sawing algorithms of wood, and aims to solve the technical problems that the sawing algorithm cannot be applied to wood with different specifications, the maximization of the value of the wood cannot be ensured, and defects in the wood are not completely sawn in the prior art, and the sawing algorithm comprises the following steps: s1: collecting wood pictures, and establishing a plate defect sample picture library, wherein S2: inputting the defect image marked with the characteristics into a network for training, and S3: collecting an image of wood to be treated through a 16k industrial line scanning camera, and marking defects on the wood image, wherein S4: according to the defect distribution condition of the front side and the back side of the wood, sawing the wood, wherein the sawing algorithm does not limit the length and the shape of the wood to be detected, and efficient cutting is performed by using a mode of arranging and combining an optimal solution, so that the value of the wood obtained after cutting is maximized, and the defect in the wood is ensured to be completely sawed.
Description
Technical Field
The invention belongs to the technical field of wood sawing algorithms, and particularly relates to a wood maximum value sawing algorithm based on an optimal solution of permutation and combination.
Background
Batching is the first step of wood processing, sawing wood into wool of various specifications and shapes according to the size and specification requirements of the product, wood being a naturally growing material, which is somewhat defective in production, so enterprises need to perform batching procedures to optimize the wood before production and processing.
At present, the invention patent with the patent number of CN201810961397.9 discloses an optimal cutting method aiming at wood surface defects, which specifically comprises the following steps: s1, firstly, after artificial blind cutting, transverse polishing and vertical polishing are carried out on rubber wood from raw materials, a wood block with the length of 300mm, the width of 80mm and the thickness of 25mm is cut, the wood is usually provided with defects such as black knots, white knots, tree cores, edge defects, cracking, oblique heads, breakage and the like, the wood is cut according to the distribution rule of the defects, and the cut wood is divided into four types: AA material, AB material, C material and waste material; s2, acquiring images of two surfaces of the wood by using a high-resolution industrial CCD, detecting defects of the two surfaces of the wood by using a convolutional neural network algorithm, and extracting the central position and the area of each defect; s3, performing Blob analysis on the defects, filtering out the defects with smaller areas, and solving out the minimum circumscribed rectangle of each defect, wherein if the two rectangles have overlapping areas or are close to each other, the two rectangles can be combined into a large rectangle for processing; s4, judging whether the two ends of the wood have defects or oblique heads, if so, cutting off the defect or oblique head areas, and generating more AA materials or AB materials at the same time although generating a part of waste materials; s5, after the oblique head and the two defects are removed, analyzing the middle defect distribution, firstly mapping the defects of two surfaces of the wood to one surface, and solving the distance from the uppermost defect to the upper edge of the wood and the distance from the lowermost defect to the lower edge of the wood on the surface; s6, dividing cutting schemes into three main categories according to whether two distances in S5 are larger than 100mm, wherein the first category is that: when the two distances are larger than 100mm, cutting a knife at the edges of the uppermost defect and the lowermost defect respectively, taking two AA materials and a short piece of waste material, and the second type: only one distance of about 100mm, for example, the distance from the uppermost defect to the upper edge of the wood exceeds 100mm, a knife is cut at the upper edge of the defect, an AA material and an AB or C material are taken, and the third is: the method comprises the steps of analyzing whether the distance from the defect at the most edge of a single surface to the edge at the same side of the wood is greater than 100mm or not, cutting a cutter at the edge of the defect if the distance from the defect at the most edge of the single surface to the edge at the same side of the wood is greater than 100mm, taking an AB material and a C material, performing mathematical modeling according to the distribution rule of the defects on the surface of the wood, extracting mathematical models from a plurality of defect distribution types, classifying the mathematical models into three categories, and performing subdivision analysis on each category, so that the cutting scheme is standardized, but the cutting method cannot be suitable for the wood with different specifications, has poor universality, cannot arrange the sawing modes of the wood board, cannot guarantee the maximization of the wood value, and does not completely saw the defects in the wood, and influences the quality of the wood.
Therefore, in order to solve the problem that the maximum value of the wood cannot be ensured after sawing, the problem needs to be solved so as to improve the use situation of the wood.
Disclosure of Invention
(1) Technical problem to be solved
Aiming at the defects of the prior art, the invention aims to provide a sawing algorithm with the maximum value of the wood based on an optimal solution of permutation and combination, which aims to solve the technical problems that the sawing algorithm cannot be applied to the wood with different specifications, the maximization of the value of the wood cannot be ensured, and the defects in the wood are not completely sawn in the prior art.
(2) Technical proposal
In order to solve the technical problems, the invention provides a timber maximum value sawing algorithm based on an optimal solution of permutation and combination, which comprises the following steps:
s1: collecting wood pictures, preprocessing the pictures and extracting the characteristics, establishing a plate defect sample picture library, dividing the sample pictures, expanding the data of the pictures through data enhancement, marking the expanded pictures, and dividing the pictures in the plate defect sample picture library into a training set, a testing set and a verification set, wherein the steps of preprocessing the pictures and extracting the characteristics are as follows:
s11: the image preprocessing process comprises the following steps: graying, geometric transformation and image enhancement, wherein the graying adopts a weighted average method to obtain a gray image, the geometric transformation processes the acquired image through translation, transposition, mirroring, rotation and scaling, the systematic error of an image acquisition system and the random error of an instrument position are corrected, then a gray interpolation algorithm is used for calculation, and the image enhancement algorithm comprises a spatial domain method and a frequency domain method;
s12: dividing an image into n coordinate grids, counting gradient histograms of each coordinate grid to form characteristics of each coordinate grid, forming a block by each m coordinate grids, connecting the characteristics of all coordinate grids in each block in series to obtain the characteristics of the block, and connecting the characteristics of all blocks in the image in series to obtain characteristic vectors of the image for classification;
s2: inputting the defect images marked with the features into a network for training, and adopting a full-supervision learning model:
s21: the characterization study takes the defect detection problem as a classification task in computer vision, including coarse-grained image label classification, region classification and pixel classification;
s22: the measurement learning directly learns the input similarity measurement by using deep learning, in a defect classification task, a twin network is adopted to perform measurement learning, two or more paired images are input into the twin network, the similarity of the input images is learned through the network, and whether the input images belong to the same class is judged;
s3: collecting an image of wood to be treated by a 16k industrial line scanning camera, and marking the defect of the wood image:
s31: changing the original image of the wood picture to be processed into a gray image;
s32: performing Gaussian filtering on the picture;
s33: carrying out uniform illumination treatment on the filtered picture;
s34: equalizing the image;
s35: performing binarization processing on the image;
s36: a communication region;
s37: edge detection, which marks points with obvious brightness change in a digital image, reduces data volume, retains important structural attributes of the image, and can be divided into two types: search-based and zero-crossing-based;
s38: marking, namely positioning coordinates of the defect, wherein the initial position of the defect area is represented by X, and DeltaX represents the width of the defect area;
s4: dividing the upper and lower surfaces of the wood into a section with a defect and a section without a defect in a system according to the defect distribution condition of the front and back surfaces of the wood, setting the length of the minimum effective wood as D, calculating the distance between the first section with the defect and the second section with the defect, combining the first section with the defect and the second section with the defect for sawing when the distance is smaller than D, arranging all defect areas in the same way, generating sawing images in the system, and then sawing the wood.
Further, the number of sample pictures in the sample picture library in S1 is not less than 2000, wherein the ratio of the number of sample pictures in the training set, the test set and the verification set is 8:1:1.
Further, in the step S11, each pixel of the gray level image stores a gray level value by one byte, the gray level range is 0-255, and the gray level interpolation algorithm adopts the methods of nearest neighbor interpolation, bilinear interpolation and bicubic interpolation.
Further, the size of each coordinate grid in S12 is 6*6 pixels.
Further, the 16k industrial line scanning camera in the step S3 adopts an up-down double line scanning camera correlation architecture and is matched with a low-distortion quasi-lens.
Further, in the step S32, the current pixel is used as a kernel center, and the convolution kernel is used to perform weighted average on the neighboring pixels, and the value is used as a new value of the current pixel.
Further, in S36, the connected region refers to a region composed of pixels having the same pixel value and adjacent to each other in the image, and the connected region analysis refers to finding out the connected regions independent of each other in the image and marking them.
Further, the step S37 detects a boundary by finding a maximum value in the first derivative of the image based on the search, then estimates the local direction of the edge using the calculation result, adopts the direction of the gradient, and finds the maximum value of the local gradient mode in this direction.
Further, the step S37 searches for a boundary by searching for zero crossings of the second derivative of the image based on zero crossings.
(3) Advantageous effects
Compared with the prior art, the invention has the beneficial effects that: the sawing algorithm of the invention is not limited to the length and the shape of the wood to be detected, the universality is better, the cutting modes are various according to the defect distribution condition of the front and the back of the wood, the high-efficiency cutting is carried out by utilizing the mode of arranging and combining the optimal solutions, so that the value of the wood obtained after the cutting is maximized, the defect in the wood is ensured to be completely sawed, the latest 16k industrial line scanning camera is adopted, compared with the 4k industrial line scanning camera used by the conventional similar equipment, the unilateral resolution is improved by about 4 times, the line scanning camera is used for completely and efficiently obtaining the whole image, the problems of time and precision brought by the graph acquisition and the graph splicing of a plurality of area array cameras are avoided, and the ideal image with low distortion and uniform illumination can be obtained by matching with the low distortion quasi-lens of the latest technology.
Detailed Description
The specific implementation mode is a wood maximum value sawing algorithm based on an optimal solution of permutation and combination, and the method comprises the following steps:
s1: collecting wood pictures, preprocessing the pictures and extracting the characteristics, establishing a plate defect sample picture library, dividing the sample pictures, expanding the data of the pictures through data enhancement, marking the expanded pictures, and dividing the pictures in the plate defect sample picture library into a training set, a testing set and a verification set, wherein the steps of preprocessing the pictures and extracting the characteristics are as follows:
s11: the image preprocessing process comprises the following steps: graying, geometric transformation and image enhancement, wherein the graying adopts a weighted average method to obtain a gray image, the geometric transformation processes the acquired image through translation, transposition, mirroring, rotation and scaling, the systematic error of an image acquisition system and the random error of an instrument position are corrected, then a gray interpolation algorithm is used for calculation, and the image enhancement algorithm comprises a spatial domain method and a frequency domain method;
s12: dividing an image into n coordinate grids, counting gradient histograms of each coordinate grid to form characteristics of each coordinate grid, forming a block by each m coordinate grids, connecting the characteristics of all coordinate grids in each block in series to obtain the characteristics of the block, and connecting the characteristics of all blocks in the image in series to obtain characteristic vectors of the image for classification;
s2: inputting the defect images marked with the features into a network for training, and adopting a full-supervision learning model:
s21: the characterization study takes the defect detection problem as a classification task in computer vision, including coarse-grained image label classification, region classification and pixel classification;
s22: the measurement learning directly learns the input similarity measurement by using deep learning, in a defect classification task, a twin network is adopted to perform measurement learning, two or more paired images are input into the twin network, the similarity of the input images is learned through the network, and whether the input images belong to the same class is judged;
s3: collecting an image of wood to be treated by a 16k industrial line scanning camera, and marking the defect of the wood image:
s31: changing the original image of the wood picture to be processed into a gray image;
s32: performing Gaussian filtering on the picture;
s33: carrying out uniform illumination treatment on the filtered picture;
s34: equalizing the image;
s35: performing binarization processing on the image;
s36: a communication region;
s37: edge detection, which marks points with obvious brightness change in a digital image, reduces data volume, retains important structural attributes of the image, and can be divided into two types: search-based and zero-crossing-based;
s38: marking, namely positioning coordinates of the defect, wherein the initial position of the defect area is represented by X, and DeltaX represents the width of the defect area;
s4: dividing the upper and lower surfaces of the wood into a section with a defect and a section without a defect in a system according to the defect distribution condition of the front and back surfaces of the wood, setting the length of the minimum effective wood as D, calculating the distance between the first section with the defect and the second section with the defect, combining the first section with the defect and the second section with the defect for sawing when the distance is smaller than D, arranging all defect areas in the same way, generating sawing images in the system, and then sawing the wood.
Further, the number of sample pictures in the sample picture library in S1 is not less than 2000, wherein the ratio of the number of sample pictures in the training set, the test set and the verification set is 8:1:1.
Further, in the step S11, each pixel of the gray level image stores a gray level value by one byte, the gray level range is 0-255, and the gray level interpolation algorithm adopts the methods of nearest neighbor interpolation, bilinear interpolation and bicubic interpolation.
Further, the size of each coordinate grid in S12 is 6*6 pixels.
Further, the 16k industrial line scanning camera in the step S3 adopts an up-down double line scanning camera correlation architecture and is matched with a low-distortion quasi-lens.
Further, in the step S32, the current pixel is used as a kernel center, and the convolution kernel is used to perform weighted average on the neighboring pixels, and the value is used as a new value of the current pixel.
Further, in S36, the connected region refers to a region composed of pixels having the same pixel value and adjacent to each other in the image, and the connected region analysis refers to finding out the connected regions independent of each other in the image and marking them.
Further, the step S37 detects a boundary by finding a maximum value in the first derivative of the image based on the search, then estimates the local direction of the edge using the calculation result, adopts the direction of the gradient, and finds the maximum value of the local gradient mode in this direction.
Further, the step S37 searches for a boundary by searching for zero crossings of the second derivative of the image based on zero crossings.
When the sawing algorithm of the technical scheme is used, the steps are as follows:
s1: collecting wood pictures, preprocessing the images and extracting the characteristics, establishing a plate defect sample picture library, wherein the number of sample pictures in the sample picture library is not less than 2000, the sample picture number ratio of a training set, a test set and a verification set is 8:1:1, dividing the sample pictures, expanding the data of the pictures through data enhancement, labeling the expanded images, dividing the pictures in the plate defect sample picture library into the training set, the test set and the verification set, and preprocessing and extracting the characteristics of the images as follows:
s11: the image preprocessing process comprises the following steps: graying, geometric transformation and image enhancement, wherein the graying adopts a weighted average method to obtain a gray image, the geometric transformation processes the acquired image through translation, transposition, mirroring, rotation and scaling, the systematic error of an image acquisition system and the random error of an instrument position are corrected, then a gray interpolation algorithm is used for calculation, the image enhancement algorithm comprises a spatial domain method and a frequency domain method, each pixel of the gray image stores a gray value by one byte, the gray range is 0-255, and the gray interpolation algorithm adopts the methods of nearest neighbor interpolation, bilinear interpolation and bicubic interpolation;
s12: dividing an image into n coordinate grids, wherein the size of each coordinate grid is 6*6 pixels, counting gradient histograms of each coordinate grid to form characteristics of each coordinate grid, forming a block by each m coordinate grids, connecting the characteristics of all coordinate grids in each block in series to obtain the characteristics of the block, and connecting the characteristics of all blocks in the image in series to obtain characteristic vectors of the image for classification;
s2: inputting the defect images marked with the features into a network for training, and adopting a full-supervision learning model:
s21: the characterization study takes the defect detection problem as a classification task in computer vision, including coarse-grained image label classification, region classification and pixel classification;
s22: the measurement learning directly learns the input similarity measurement by using deep learning, in a defect classification task, a twin network is adopted to perform measurement learning, two or more paired images are input into the twin network, the similarity of the input images is learned through the network, and whether the input images belong to the same class is judged;
s3: the method comprises the steps that an image of wood to be processed is acquired through a 16k industrial line scanning camera, the 16k industrial line scanning camera adopts an up-down double line scanning camera correlation architecture, a low-distortion quasi-lens is matched, and defect marking is carried out on the wood image:
s31: changing the original image of the wood picture to be processed into a gray image;
s32: performing Gaussian filtering on the picture, taking a current pixel as a kernel center when performing Gaussian filtering, and performing weighted average on surrounding neighborhood pixels by using a convolution check, wherein the value of the weighted average is taken as a new value of the current pixel;
s33: carrying out uniform illumination treatment on the filtered picture;
s34: equalizing the image;
s35: performing binarization processing on the image;
s36: the connected region refers to a region which is formed by pixels with the same pixel value and adjacent positions in the image, and the connected region analysis refers to finding out mutually independent connected regions in the image and marking the mutually independent connected regions;
s37: edge detection, which marks points with obvious brightness change in a digital image, reduces data volume, retains important structural attributes of the image, and can be divided into two types: detecting a boundary by searching a maximum value in a first derivative of an image based on searching and zero crossing, estimating a local direction of the edge by using a calculation result, adopting a gradient direction, finding a maximum value of a local gradient mode in the direction, and searching the boundary by searching a second derivative zero crossing of the image based on zero crossing;
s38: marking, namely positioning coordinates of the defect, wherein the initial position of the defect area is represented by X, and DeltaX represents the width of the defect area;
s4: dividing the upper and lower surfaces of the wood into a section with a defect and a section without a defect in a system according to the defect distribution condition of the front and back surfaces of the wood, setting the length of the minimum effective wood as D, calculating the distance between the first section with the defect and the second section with the defect, combining the first section with the defect and the second section with the defect for sawing when the distance is smaller than D, arranging all defect areas in the same way, generating sawing images in the system, and then sawing the wood.
Claims (9)
1. The maximum value sawing algorithm for the timber based on the optimal solution of the permutation and combination is characterized by comprising the following steps:
s1: collecting wood pictures, preprocessing the pictures and extracting the characteristics, establishing a plate defect sample picture library, dividing the sample pictures, expanding the data of the pictures through data enhancement, marking the expanded pictures, and dividing the pictures in the plate defect sample picture library into a training set, a testing set and a verification set, wherein the steps of preprocessing the pictures and extracting the characteristics are as follows:
s11: the image preprocessing process comprises the following steps: graying, geometric transformation and image enhancement, wherein the graying adopts a weighted average method to obtain a gray image, the geometric transformation processes the acquired image through translation, transposition, mirroring, rotation and scaling, the systematic error of an image acquisition system and the random error of an instrument position are corrected, then a gray interpolation algorithm is used for calculation, and the image enhancement algorithm comprises a spatial domain method and a frequency domain method;
s12: dividing an image into n coordinate grids, counting gradient histograms of each coordinate grid to form characteristics of each coordinate grid, forming a block by each m coordinate grids, connecting the characteristics of all coordinate grids in each block in series to obtain the characteristics of the block, and connecting the characteristics of all blocks in the image in series to obtain characteristic vectors of the image for classification;
s2: inputting the defect images marked with the features into a network for training, and adopting a full-supervision learning model:
s21: the characterization study takes the defect detection problem as a classification task in computer vision, including coarse-grained image label classification, region classification and pixel classification;
s22: the measurement learning directly learns the input similarity measurement by using deep learning, in a defect classification task, a twin network is adopted to perform measurement learning, two or more paired images are input into the twin network, the similarity of the input images is learned through the network, and whether the input images belong to the same class is judged;
s3: collecting an image of wood to be treated by a 16k industrial line scanning camera, and marking the defect of the wood image:
s31: changing the original image of the wood picture to be processed into a gray image;
s32: performing Gaussian filtering on the picture;
s33: carrying out uniform illumination treatment on the filtered picture;
s34: equalizing the image;
s35: performing binarization processing on the image;
s36: a communication region;
s37: edge detection, which marks points with obvious brightness change in a digital image, reduces data volume, retains important structural attributes of the image, and can be divided into two types: search-based and zero-crossing-based;
s38: marking, namely positioning coordinates of the defect, wherein the initial position of the defect area is represented by X, and DeltaX represents the width of the defect area;
s4: dividing the upper and lower surfaces of the wood into a section with a defect and a section without a defect in a system according to the defect distribution condition of the front and back surfaces of the wood, setting the length of the minimum effective wood as D, calculating the distance between the first section with the defect and the second section with the defect, combining the first section with the defect and the second section with the defect for sawing when the distance is smaller than D, arranging all defect areas in the same way, generating sawing images in the system, and then sawing the wood.
2. The wood maximum value sawing algorithm based on the permutation and combination optimal solution according to claim 1, wherein the number of sample pictures in the sample picture library in S1 is not less than 2000, and the ratio of the number of sample pictures in the training set, the test set and the verification set is 8:1:1.
3. The method for sawing the maximum value of the wood based on the optimal solution of the permutation and combination according to claim 1, wherein each pixel of the gray level image in the step S11 stores a gray level value with one byte, the gray level range is 0-255, and the method adopted by the gray level interpolation algorithm is nearest neighbor interpolation, bilinear interpolation and bicubic interpolation.
4. The wood maximum value sawing algorithm based on the permutation and combination optimal solution according to claim 1, wherein the size of each coordinate grid in S12 is 6*6 pixels.
5. The method for sawing the maximum value of the wood based on the optimal solution of the permutation and combination according to claim 1, wherein the 16k industrial line scanning camera in the step S3 adopts an up-down double line scanning camera correlation architecture and is matched with a low-distortion quasi-lens.
6. The method according to claim 1, wherein the gaussian filtering in S32 uses the current pixel as a kernel center, and uses a convolution kernel to perform weighted average on neighboring pixels, and the value is used as a new value of the current pixel.
7. The wood maximum value sawing algorithm based on the permutation and combination optimal solution according to claim 1, wherein the connected region in S36 refers to a region composed of pixels having the same pixel value and adjacent to each other in the image, and the connected region analysis refers to finding out connected regions independent of each other in the image and marking the connected regions.
8. The method according to claim 1, wherein the step S37 is based on searching to detect boundaries by finding maximum values in the first derivative of the image, estimating local directions of the edges using the calculation result, taking the direction of the gradient, and finding the maximum value of the local gradient mode in the direction.
9. The wood maximum value sawing algorithm based on permutation and combination optimal solutions according to claim 1, wherein the boundary is found by finding the second derivative zero crossing of the image based on zero crossing in S37.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310619170.7A CN116824483A (en) | 2023-05-30 | 2023-05-30 | Maximum value sawing algorithm for timber based on optimal solution of permutation and combination |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310619170.7A CN116824483A (en) | 2023-05-30 | 2023-05-30 | Maximum value sawing algorithm for timber based on optimal solution of permutation and combination |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116824483A true CN116824483A (en) | 2023-09-29 |
Family
ID=88117663
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310619170.7A Pending CN116824483A (en) | 2023-05-30 | 2023-05-30 | Maximum value sawing algorithm for timber based on optimal solution of permutation and combination |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116824483A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117152161A (en) * | 2023-11-01 | 2023-12-01 | 山东迪特智联信息科技有限责任公司 | Shaving board quality detection method and system based on image recognition |
-
2023
- 2023-05-30 CN CN202310619170.7A patent/CN116824483A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117152161A (en) * | 2023-11-01 | 2023-12-01 | 山东迪特智联信息科技有限责任公司 | Shaving board quality detection method and system based on image recognition |
| CN117152161B (en) * | 2023-11-01 | 2024-03-01 | 山东迪特智联信息科技有限责任公司 | Shaving board quality detection method and system based on image recognition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105067638B (en) | Tire fetal membrane face character defect inspection method based on machine vision | |
| CN107301637B (en) | Method for detecting surface flaws of approximately rectangular planar industrial product | |
| CN109658381B (en) | Method for detecting copper surface defects of flexible IC packaging substrate based on super-pixels | |
| CN114910480B (en) | Wafer surface defect detection method based on machine vision | |
| CN108181316B (en) | Bamboo strip defect detection method based on machine vision | |
| CN109767445B (en) | High-precision PCB defect intelligent detection method | |
| CN107543828A (en) | A kind of Surface Flaw Detection method and system | |
| CN106600600A (en) | Wafer defect detection method based on characteristic matching | |
| CN108562589A (en) | A method of magnetic circuit material surface defect is detected | |
| CN108596880A (en) | Weld defect feature extraction based on image procossing and welding quality analysis method | |
| CN112465809B (en) | Mold defect detection method based on image recognition, computer readable storage medium and device | |
| CN113706490B (en) | Wafer defect detection method | |
| CN110751604A (en) | Machine vision-based steel pipe weld defect online detection method | |
| CN112489042B (en) | Method for detecting metal printing defects and surface damage based on super-resolution reconstruction | |
| CN116824483A (en) | Maximum value sawing algorithm for timber based on optimal solution of permutation and combination | |
| CN115345876B (en) | Bolt thread defect detection method based on ultrasonic image | |
| CN117576100B (en) | Surface defect grading detection and evaluation method for FPC connector | |
| CN110781913B (en) | Zipper cloth belt defect detection method | |
| CN105931246A (en) | Fabric flaw detection method based on wavelet transformation and genetic algorithm | |
| CN114187267B (en) | Stamping part defect detection method based on machine vision | |
| CN118096722A (en) | Printing defect detection method based on machine vision and improved template matching | |
| CN109934800A (en) | Positioning method and system for cigarette packet paperboard | |
| CN111144513A (en) | Industrial product spraying defect data enhancement method combining generation countermeasure network and traditional image processing technology | |
| CN116434054A (en) | Intensive remote sensing ground object extraction method based on line-plane combination | |
| CN112862826B (en) | Normal sample nondestructive generation method for surface defect detection task |
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 |