CN110246123B - Tile paving regularity detection method based on machine vision - Google Patents
Tile paving regularity detection method based on machine vision Download PDFInfo
- Publication number
- CN110246123B CN110246123B CN201910424955.2A CN201910424955A CN110246123B CN 110246123 B CN110246123 B CN 110246123B CN 201910424955 A CN201910424955 A CN 201910424955A CN 110246123 B CN110246123 B CN 110246123B
- Authority
- CN
- China
- Prior art keywords
- image
- tile
- function
- judging
- regularity
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
Abstract
The invention relates to a tile paving regularity detection method based on machine vision, which belongs to the technical field of tile detection and comprises the following steps: acquiring an image after the tile is laid and pasted, and generating a grid matrix image; performing dimensionality reduction on the grid matrix image to generate a monochrome image; coordinate calibration is carried out on four corner points and a central point of a measured ceramic tile in a monochrome image, the points of each calibrated row and each calibrated column are respectively connected, a curve function is respectively fitted and established, and deviation information is obtained according to curve function analysis; judging deviation information, and judging that the information is qualified if the deviation information is less than 5; otherwise, the product is judged to be unqualified. The method has the beneficial effect of being capable of rapidly and accurately detecting the regularity of the tile paving.
Description
Technical Field
The invention belongs to the technical field of tile detection, and particularly relates to a tile paving regularity detection method based on machine vision.
Background
The ceramic tile is mainly a building material for decorating and protecting the outer wall, and the outer wall in general indoor and outdoor at present can be paved with the ceramic tile, so that the outer wall is protected from being corroded by rainwater, the service life of a building is longer, and the appearance of the building can be beautified.
The tiles are in block shapes and are regularly paved on an outer wall or an indoor wall by workers, but the paved tiles are generally positioned by positioning lines before paving so as to ensure that the paved tiles are regular and the paved tiles are neat and beautiful; however, due to the reason of observing the viewing angle during manual paving, the paved ceramic tiles often incline or are not flat, so that the attractiveness of the paved ceramic tiles is greatly affected, and therefore the regularity of the paved ceramic tiles needs to be detected.
Disclosure of Invention
The invention provides a method for detecting the tile paving regularity based on machine vision, which aims to solve the technical problems and can quickly and accurately detect the tile paving regularity.
The technical scheme for solving the technical problems is as follows: a tile paving regularity detection method based on machine vision acquires an image after tile paving and generates a grid matrix image; performing dimensionality reduction on the grid matrix image to generate a monochrome image; coordinate calibration is carried out on four corner points and a central point of a measured ceramic tile in a monochrome image, the points of each row and each column which are calibrated are respectively connected, a straight line function is respectively fitted and established, and deviation information is obtained according to the analysis of the straight line function; judging deviation information, and judging that the information is qualified if the deviation information is less than 5; otherwise, the product is judged to be unqualified.
The invention has the beneficial effects that: the tile that can spread under to different scenes pastes detects, acquires the wall picture of pasting the ceramic tile through the shooting mode, can not need the manual work to detect, has saved a large amount of manpower and materials, and the effect that detects simultaneously is more accurate than the manual work detection, improves work efficiency.
On the basis of the technical scheme, the invention can be further improved as follows.
Further, the specific processing formula for generating the monochrome image by performing the dimensionality reduction processing on the grid matrix image is as follows:
I′(x,y)＝α·IR(x,y)+β·IG(x,y)+λ·IB(x,y)
where α denotes a red chromaticity coefficient in the case of photographing, β denotes a green chromaticity coefficient in the case of photographing, λ denotes a blue chromaticity coefficient in the case of photographing, and α + β + λ ≦ 1, x ═ 1,2, … h and h denote the image height, y ═ 1,2, … w and w denote the image width.
The beneficial effect of adopting the further scheme is that: by analyzing the three primary colors of the image, a more accurate monochromatic image is obtained according to different indoor and outdoor brightness and different color differences, so that the subsequent further processing is facilitated, and the transportation efficiency can be improved.
Further, the calibration method for the central point of the measured tile in the monochrome image comprises the following steps: firstly, determining each edge of a ceramic tile to be detected; and calculating the coordinates of the central point of the ceramic tile to be measured according to the function of each edge.
The beneficial effect of adopting the further scheme is that: the offset information amount of the ceramic tile can be measured by determining the central point, the offset information of the ceramic tile in the horizontal direction and the vertical direction can be accurately acquired, and if the central point is offset in the horizontal direction and the vertical direction, the offset information of the central point is larger.
Further, the method also comprises depth detection, and comprises the following specific steps:
and acquiring a side image after the tile is paved, constructing a plane function, judging that the tile is unqualified if the concave-convex condition occurs, and judging that the tile is qualified if the concave-convex condition does not occur.
The beneficial effect of adopting the further scheme is that: the purpose here is to determine whether the tile is convex or concave.
Further, the constructing plane function is a three-dimensional plane function, and a three-dimensional grid matrix B (i, j, h) is generated according to the acquired image, wherein i represents the number of rows, j represents the number of columns, and h represents the depth; fitting a three-dimensional stereo function f (i, j, h) as follows:
wherein the content of the first and second substances,
sigma is used as a correction parameter;
and solving the extreme points of each concave peak and each convex peak in the function f (i, j, h) through an evolution optimization algorithm, and calculating the distance between the extreme points and the horizontal depth value to obtain the concave-convex degree value, the concave-convex position and the number of the whole image to be measured.
The beneficial effect of adopting the further scheme is that: can detect out the unsmooth position of ceramic tile through calculating concave peak and convex peak, it is better to the roughness detection effect of ceramic tile.
Drawings
FIG. 1 is a schematic flow chart of the detection method of the present invention.
Detailed Description
The principles and features of this invention are described below in conjunction with the following drawings, which are set forth by way of illustration only and are not intended to limit the scope of the invention.
Examples
As shown in fig. 1, the present embodiment provides a tile paving regularity detecting method based on machine vision, including the following steps:
shooting and acquiring an image after the tile is laid by using shooting equipment, specifically, the shooting equipment can be a mobile phone or a camera, transmitting the shot image to a computer, and generating a grid matrix image by using the computer;
the grid matrix image is subjected to dimensionality reduction to generate a monochrome image, and the processing is specifically carried out by adopting the following formula:
I′(x,y)＝α·IR(x,y)+β·IG(x,y)+λ·IB(x,y)
where α denotes a red chromaticity coefficient in the case of photographing, β denotes a green chromaticity coefficient in the case of photographing, λ denotes a blue chromaticity coefficient in the case of photographing, and α + β + λ ≦ 1, x ═ 1,2, … h and h denote an image height, y ═ 1,2, … w and w denote an image width;
by analyzing the three primary colors of the image, a more accurate monochromatic image is obtained according to different brightness and different color differences indoors and outdoors, so that the subsequent further processing is facilitated.
To be noted, IR(x,y)，IG(x, y) and IB(x, y) respectively represents the values of the red, green and blue three primary colors corresponding to each pixel point in the image, wherein the value range is [0,255 ]]Alpha is the red chroma coefficient of the corresponding pixel in the shooting case, and the value is obtained by the pair IRThe normalization process of (x, y) yields:
beta represents the green chroma coefficient of the corresponding pixel in the shooting condition, and the value is obtained by the pair IGThe normalization process of (x, y) yields:
λ represents the blue chrominance coefficient of the corresponding pixel in the case of photographing, and the value is obtained by comparing IBThe normalization process of (x, y) yields:
α + β + λ ≦ 1, x ═ 1,2, … h, h denotes the mesh matrix image height, y ═ 1,2, … w, w denotes the mesh matrix image width; the monochrome image I' (x, y) after the dimension reduction is obtained by the processing of the above formula. And the transportation efficiency is improved after the dimension is reduced into the single-body image.
The calibration method for the central point of the measured ceramic tile in the monochrome image comprises the following steps:
determining each edge of the ceramic tile to be detected, wherein the determination formula is as follows:
wherein the content of the first and second substances,
σdand σrWhich represents the coefficient of the smoothing, is,
h 'and w' denote the height and width of the precision window respectively,
as a weight coefficient, is composed of
And (i, j) similarity;
where A (i, j) represents a function of each edge.
And calculating the coordinate of the central point of the ceramic tile to be measured according to the function of each edge.
And then carrying out coordinate calibration on the four corner points of the ceramic tile to be measured in the monochrome image, specifically, the calibration mode of the four corner points of the ceramic tile to be measured in the monochrome image is to traverse I' twice in a traversing mode, traverse from front to back for the first time, and traverse from back to front for the second time. Further traversal refers to scanning each point one by one.
Connecting the calibrated points and fitting to establish a linear function, specifically comprising connecting the coordinates of the calibrated points in each row and fitting to establish a linear function yi＝f(xi)。
Connecting the coordinate of the calibration point of each column, and fitting to establish a straight line function y'i＝f′(xi)。
And obtaining deviation information of each line by using a straight line function of each line, judging the deviation information of each line to be qualified if the deviation information of each line is larger than or equal to xi, i is 1, … and w, and judging the deviation information of each line to be unqualified if the deviation information of each line is not smaller than xi, wherein xi represents the error of a horizontal direction fitting value and an average value, the xi is set to be 5 pixel points, and xi is 5 which is a value range acceptable by vision.
And obtaining deviation information of each column by using a straight line function of each column, judging the deviation information to be qualified if the deviation information of each column is larger than or equal to xi, i is 1, … and wh, and judging the deviation information to be unqualified if the deviation information of each column is not smaller than xi, wherein xi represents the error of a vertical direction fitting value and an average value, the error is set to be 5 pixel points, and xi is 5 which is a value range acceptable by vision.
Preferably, the tile paving regularity detection method based on machine vision in this embodiment further includes depth detection, and includes the following specific steps:
and acquiring a side image after the tile is paved, constructing a plane function, judging that the tile is unqualified if the concave-convex condition occurs, and judging that the tile is qualified if the concave-convex condition does not occur.
Specifically, the constructed plane function is a constructed three-dimensional plane function, and a three-dimensional grid matrix B (i, j, h) is generated according to the acquired image, wherein i represents a row number, j represents a column number, and h represents a depth; fitting a three-dimensional stereo function f (i, j, h) as follows:
wherein the content of the first and second substances,
sigma is used as a correction parameter;
and solving the extreme points of each concave peak and each convex peak in the function f (i, j, h) through an evolution optimization algorithm, and calculating the distance between the extreme points and the horizontal depth value to obtain the concave-convex degree value, the concave-convex position and the number of the whole image to be measured.
Specifically, coordinate positions corresponding to i, j and h are taken as any solution in the three-dimensional stereo function, n solutions (n is the number of tiles in the image) are randomly acquired, and mutation operation is performed on the acquired n solutions: f. ofθ′(i,j,h)＝fθ(i,j,h)+rand(0,1)(max(f1...θ(i,j,h))-min(f1...θ(i, j, h))), θ ═ 1, …, n, where fθ(i, j, h) represents the current solution, rand (0,1) represents a value randomly generated in the interval 0 to 1, max (f)1...θ(i, j, h)) represents the solution of the maximum value of the current solution median, min (f)1...θ(i, j, h)) represents the solution whose current solution median is the smallest; performing mutation operation on the n solutions through the formula to generate n new solutions
Generating a new round solution by which to iterate operations when f'θ(i,j,h)-fθStopping iteration when (i, j, h) | is 0 to obtain fθEach of (i, j, h) a concave peak and a convex peak extreme point. Horizontal depth value is 0, so fθ(i, j, h) > 0 is a bump, fθ(i, j, h) < 0 is a pit, and specifically fθThe (i, j, h) value is the degree of unevenness. Wherein the coordinate point (i, j, h) is a concave-convex position.
It should be noted that the three-dimensional stereo function is a three-dimensional fitting function established according to depth information of an acquired image, a normal tile plane is a plane with a fitting function depth of 0, if a tile recess condition occurs, a concave peak occurs in the fitting function, and if a tile protrusion condition occurs, a convex peak occurs in the fitting function. And judging whether the attached ceramic tile is qualified or not according to the concave-convex degree.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (5)
1. A tile paving regularity detection method based on machine vision is characterized by comprising the following steps:
acquiring an image after the tile is laid and pasted, and generating a grid matrix image;
performing dimensionality reduction on the grid matrix image to generate a monochrome image;
coordinate calibration is carried out on four corner points and a central point of a measured ceramic tile in a monochrome image, the points of each row and each column which are calibrated are respectively connected, a straight line function is respectively fitted and established, and deviation information is obtained according to the analysis of the straight line function;
judging deviation information, and judging that the information is qualified if the deviation information is less than 5; otherwise, the product is judged to be unqualified.
2. The method for detecting tile paving regularity based on machine vision according to claim 1, wherein the specific processing formula for generating the monochrome image by the grid matrix image dimension reduction processing is as follows:
I′(x,y)＝α·IR(x,y)+β·IG(x,y)+λ·IB(x,y)
where α denotes a red chromaticity coefficient in the case of photographing, β denotes a green chromaticity coefficient in the case of photographing, λ denotes a blue chromaticity coefficient in the case of photographing, and α + β + λ ≦ 1, x ═ 1,2, … h and h denote the image height, y ═ 1,2, … w and w denote the image width.
3. The machine-vision-based tile paving regularity detection method according to claim 1, wherein said calibration of the central points of the measured tiles in a monochromatic image comprises the steps of:
firstly, determining each edge of a ceramic tile to be detected;
and calculating the coordinates of the central point of the ceramic tile to be measured according to the function of each edge.
4. The machine vision-based tile paving regularity detection method according to claim 1, further comprising depth detection, the specific steps being as follows:
and acquiring a side image after the tile is paved, constructing a plane function, judging that the tile is unqualified if the concave-convex condition occurs, and judging that the tile is qualified if the concave-convex condition does not occur.
5. The machine-vision-based tile paving regularity detection method according to claim 4, wherein said building plane function is a building three-dimensional plane function, and a three-dimensional grid matrix B (i, j, h) is generated from the acquired image, wherein i represents a number of rows, j represents a number of columns, and h represents a depth; fitting a three-dimensional stereo function f (i, j, h) as follows:
wherein the content of the first and second substances,
sigma is used as a correction parameter;
and solving the extreme points of each concave peak and each convex peak in the function f (i, j, h) through an evolution optimization algorithm, and calculating the distance between the extreme points and the horizontal depth value to obtain the concave-convex degree value, the concave-convex position and the number of the whole image to be measured.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910424955.2A CN110246123B (en) | 2019-05-21 | 2019-05-21 | Tile paving regularity detection method based on machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910424955.2A CN110246123B (en) | 2019-05-21 | 2019-05-21 | Tile paving regularity detection method based on machine vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110246123A CN110246123A (en) | 2019-09-17 |
| CN110246123B true CN110246123B (en) | 2021-08-03 |
Family
ID=67884748
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910424955.2A Active CN110246123B (en) | 2019-05-21 | 2019-05-21 | Tile paving regularity detection method based on machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110246123B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110853014B (en) * | 2019-11-12 | 2020-06-26 | 王文学 | Failure rate analysis device for cancer operating room |
| CN111894247B (en) * | 2020-08-04 | 2022-05-10 | 广东博智林机器人有限公司 | Decorative plane materiel paving method, device, system and storage medium |
| CN112348795A (en) * | 2020-11-05 | 2021-02-09 | 深圳市海腾建设工程有限公司 | Tile paving quality detection method, detection equipment and storage medium |
| CN113375568B (en) * | 2021-05-12 | 2023-03-31 | 苏州阿普奇物联网科技有限公司 | Metal wiredrawing polishing defect detection method based on laser scanning |
| CN114778547A (en) * | 2022-04-01 | 2022-07-22 | 西南交通大学 | Method for detecting surface pasting quality of curved surface part based on image processing |
| CN117274249B (en) * | 2023-11-20 | 2024-03-01 | 江西省中鼐科技服务有限公司 | Ceramic tile appearance detection method and system based on artificial intelligent image technology |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107505948A (en) * | 2017-07-20 | 2017-12-22 | 航天东方红卫星有限公司 | It is a kind of to be used for the motor-driven middle attitude adjusting method being imaged along curvilinear bands of quick satellite |
| CN107945136A (en) * | 2017-12-04 | 2018-04-20 | 深圳市路畅科技股份有限公司 | A kind of fish eye images antidote, system, equipment and computer-readable storage medium |
| CN109082984A (en) * | 2018-06-29 | 2018-12-25 | 浙江大学城市学院 | A kind of road abnormality detection model based on window division and dynamic time warping |
| CN208547588U (en) * | 2018-07-17 | 2019-02-26 | 中山泰维电子有限公司 | Crystal-vibration-chip detection machine video testing agency |
| CN109470165A (en) * | 2018-12-07 | 2019-03-15 | 广州大学 | A kind of extensometer measurement method of the measurement material bending deformation based on machine vision |
| CN109506569A (en) * | 2019-01-08 | 2019-03-22 | 大连理工大学 | A method of crystallization process square block and column crystal three-dimensional dimension are monitored based on binocular vision |
| CN109520450A (en) * | 2017-09-17 | 2019-03-26 | 王海 | A kind of device for wall plastering flatness and measuring for verticality |
| CN109523553A (en) * | 2018-11-13 | 2019-03-26 | 华际科工（北京）卫星通信科技有限公司 | A kind of container unusual fluctuation monitoring method based on LSD straight-line detection partitioning algorithm |
| CN208671928U (en) * | 2018-08-01 | 2019-03-29 | 天津市康耀玻璃有限公司 | A kind of glass flatness detection device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6769766B2 (en) * | 2016-07-19 | 2020-10-14 | 株式会社ニューフレアテクノロジー | Pattern inspection device |
-
2019
- 2019-05-21 CN CN201910424955.2A patent/CN110246123B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107505948A (en) * | 2017-07-20 | 2017-12-22 | 航天东方红卫星有限公司 | It is a kind of to be used for the motor-driven middle attitude adjusting method being imaged along curvilinear bands of quick satellite |
| CN109520450A (en) * | 2017-09-17 | 2019-03-26 | 王海 | A kind of device for wall plastering flatness and measuring for verticality |
| CN107945136A (en) * | 2017-12-04 | 2018-04-20 | 深圳市路畅科技股份有限公司 | A kind of fish eye images antidote, system, equipment and computer-readable storage medium |
| CN109082984A (en) * | 2018-06-29 | 2018-12-25 | 浙江大学城市学院 | A kind of road abnormality detection model based on window division and dynamic time warping |
| CN208547588U (en) * | 2018-07-17 | 2019-02-26 | 中山泰维电子有限公司 | Crystal-vibration-chip detection machine video testing agency |
| CN208671928U (en) * | 2018-08-01 | 2019-03-29 | 天津市康耀玻璃有限公司 | A kind of glass flatness detection device |
| CN109523553A (en) * | 2018-11-13 | 2019-03-26 | 华际科工（北京）卫星通信科技有限公司 | A kind of container unusual fluctuation monitoring method based on LSD straight-line detection partitioning algorithm |
| CN109470165A (en) * | 2018-12-07 | 2019-03-15 | 广州大学 | A kind of extensometer measurement method of the measurement material bending deformation based on machine vision |
| CN109506569A (en) * | 2019-01-08 | 2019-03-22 | 大连理工大学 | A method of crystallization process square block and column crystal three-dimensional dimension are monitored based on binocular vision |
Non-Patent Citations (2)
| Title |
|---|
| "基于机器视觉的表面平整度检测方法研究";谢剑;《中国优秀硕士学位论文全文数据库 信息科技辑》;20150315(第3期);I138-1944 * |
| "基于角点检测的二维图像边缘平直化算法研究";赵瑞鹏等;《机械工程师》;20181031(第10期);44-46 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110246123A (en) | 2019-09-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110246123B (en) | Tile paving regularity detection method based on machine vision | |
| US8724893B2 (en) | Method and system for color look up table generation | |
| US8249304B2 (en) | System and method for measuring gaps between object parts | |
| CN108535321A (en) | A kind of building thermal technique method for testing performance based on three-dimensional infrared thermal imaging technique | |
| US20100021068A1 (en) | System and method for extracting boundary elements of an object | |
| CN111369659B (en) | Texture mapping method, device and equipment based on three-dimensional model | |
| CN104732578B (en) | A kind of building texture optimization method based on oblique photograph technology | |
| CN108364292B (en) | Illumination estimation method based on multiple visual angle images | |
| CN103389042A (en) | Ground automatic detecting and scene height calculating method based on depth image | |
| CN106780589A (en) | A kind of method for obtaining target depth image | |
| KR101941801B1 (en) | Image processing method and device for led display screen | |
| CN109523544B (en) | Building outer wall quality defect detection system and method thereof | |
| TWI500904B (en) | Stereo camera and automatic range finding method for measuring a distance between stereo camera and reference plane | |
| CN115049658B (en) | RGB-D camera quality detection method, electronic device and storage medium | |
| WO2022048617A1 (en) | Method, device, and system for recognizing projection position, and storage medium | |
| CN115345844A (en) | Spraying method and device based on intelligent robot, electronic equipment and medium | |
| CN109631766A (en) | A kind of wood plank dimension measurement method based on image | |
| CN102414722B (en) | Display of effect coatings on electronic display devices | |
| CN109658372B (en) | Image uniformity evaluation method and device | |
| CN109348219B (en) | Method for image conversion under different structures and image conversion device | |
| US20170365067A1 (en) | Baseball game system | |
| CN113295617B (en) | Multi-target offset detection method without reference point | |
| KR101621201B1 (en) | Wearable projection apparatus and projection method | |
| KR101507642B1 (en) | Method for compensating fogs image based on estimated transmission value | |
| CN105509643B (en) | A kind of measurement method and device of sub-pixel unit line width |
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 |