CN109509200A - Checkerboard angle point detection process, device and computer readable storage medium based on contours extract - Google Patents
Checkerboard angle point detection process, device and computer readable storage medium based on contours extract Download PDFInfo
- Publication number
- CN109509200A CN109509200A CN201811601937.9A CN201811601937A CN109509200A CN 109509200 A CN109509200 A CN 109509200A CN 201811601937 A CN201811601937 A CN 201811601937A CN 109509200 A CN109509200 A CN 109509200A
- Authority
- CN
- China
- Prior art keywords
- pixel
- points
- point
- corner
- checkerboard
- 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.)
- Granted
Links
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
- G06T7/13—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
- G06T7/187—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
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- 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
- G06T2207/20112—Image segmentation details
- G06T2207/20164—Salient point detection; Corner detection
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
Checkerboard angle point detection process, X-comers detection device and computer readable storage medium of the application based on contours extract, method include: to carry out edge detection process to original image to obtain the marginal point in original image；The tessellated profile in original image is extracted according to marginal point；Corner recognition is carried out to the tessellated profile extracted；Tessellated internal angle point is filtered out according to the angle point recognized.The application is by extracting gridiron pattern profile, eliminate a large amount of non-corner pixels, on the one hand the calculation amount of corner recognition is greatly reduced, on the other hand X-comers can accurately be identified, the advantages of combining Corner Detection Algorithm based on edge and the Corner Detection Algorithm based on gray scale, the two disadvantage is improved simultaneously, can be greatly reduced data processing amount, be promoted processing speed and efficiency, and effectively improve anti-interference ability and accuracy rate.
Description
Technical Field
The present application relates to the field of image processing, and in particular, to a method and an apparatus for detecting checkerboard corners based on contour extraction, and a computer-readable storage medium.
Background
The camera calibration and pose measurement are basic and popular problems in the field of machine vision, and the processing process usually establishes a mapping relation between a world coordinate system and a pixel coordinate system through characteristic points on a target, so that internal parameters of the camera and pose parameters of the target are obtained by solving a PnP problem. Therefore, the feature point coordinate extraction is a very important step, wherein the corner point is the most common feature point, and the checkerboard corner point is widely used as a special corner point.
At present, the detection algorithm of the checkerboard corner point can be mainly divided into two types: edge-based corner detection algorithms and grayscale-based corner detection algorithms. The edge-based corner detection algorithm firstly segments and extracts edges of an image, and then detects corners according to the characteristic that the corners are edge inflection points or intersection points. The gray-scale-based corner detection algorithm considers that a corner is a maximum point of the image with gray scale and gradient change in a local range, and therefore the corner is mainly detected by calculating curvature and gradient.
The edge-based corner detection algorithm needs to extract edges of an image firstly, and then the subsequent steps are all detection according to characteristics of inflection points or intersection points on the basis, when the edges of the image are interrupted, the corners cannot be extracted well, so that the quality requirement on edge detection is high, the processing steps of the algorithm are complex, the calculation amount is large, and the consumed time is long. The gray-scale-based corner detection algorithm mainly detects corners by calculating curvatures and gradients, which are commonly known as Harris operators and Susan operators, and in the case of a complex background, please refer to fig. 1 to 3, where fig. 1 is an original image, pixels of the original image are 1280 × 1024, fig. 2 is an identification effect after Harris operators are used, fig. 3 is an identification effect after Susan operators are used, and it can be seen that fig. 2 and 3 identify many non-corner pixels. Therefore, although the principle of the gray-scale-based angular point detection method is simple, high in stability and easy to implement, the accuracy is low, false detection is easy, and the anti-interference capability is weak.
Disclosure of Invention
The application provides a checkerboard corner detection method based on contour extraction, a checkerboard corner detection device and a computer readable storage medium, which can solve the problems of complex processing steps, large calculation amount and long consumed time of an edge-based corner detection algorithm in the existing checkerboard corner detection algorithm and the problems of low accuracy, easiness in false detection and weak anti-interference capability of a gray-level-based corner detection algorithm.
According to a first aspect of the present application, the present application provides a checkerboard corner detection method based on contour extraction, the method comprising: carrying out edge detection processing on the original image to obtain edge points in the original image; extracting the contour of the checkerboard in the original image according to the edge points; performing corner point identification on the extracted contour of the checkerboard; and screening out the inner corner points of the checkerboard according to the identified corner points.
Preferably, in the step of performing edge detection processing on the original image to obtain edge points in the original image, the method includes: by a transverse convolution factor SxAnd a longitudinal convolution factor SyPerforming plane convolution with the original image to obtain a horizontal gray difference approximate value G of the pixel point of the original imagexAnd longitudinal gray difference approximate value GyWherein the original image is set as A, and the transverse convolution factorLongitudinal convolution factorAccording to the formula: the lateral gray difference approximation is Gx＝SxA, longitudinal gray difference approximation Gy＝SyA; approximate value G of transverse gray difference of each pixel point in original image AxAnd a longitudinal gray difference approximation GyObtaining the gray scale weighting difference G of the pixel point according to the following formula (1) or formula (2), wherein the formula (1) is as follows:the formula (2) is: g ═ Gx|+|GyL, |; and when the gray weighting difference G of the pixel point is greater than a first set threshold value, the pixel point is an edge point, otherwise, the pixel point is marked to be 0.
Preferably, the step of extracting the contour of the checkerboard in the original image according to the edge points includes: scanning pixel points of an original image according to a set sequence, when the scanned pixel points in the original image are effective points, giving a set marking value to the effective points in the original image according to a first set rule, representing the marking value of the effective points in the same connected domain in an equivalent chain mode, wherein the equivalent chain comprises marking values with equivalent relations, storing the equivalent chain into an equivalent array in an equivalent pair mode, updating the marking value of the same equivalent chain in the equivalent array into a uniform marking value according to a second set rule, and then, the pixel points with the same marking value are the pixel points in the same connected domain; and judging whether the pixel points of the connected domain meet the set conditions or not, and if the parameters of the pixel points of the connected domain meet the set conditions, extracting the connected domain as the outline of the checkerboard.
Preferably, the step of assigning the set mark value to the effective point in the original image according to the first setting rule includes the steps of: judging whether the mark values of the pixel points in the neighborhood of the current effective point are all 0, wherein the pixel points in the neighborhood comprise a first pixel point, a second pixel point, a third pixel point and a fourth pixel point, the first pixel point is a pixel point adjacent to the left side of the current effective point, the second pixel point is a pixel point adjacent to the upper part of the first pixel point, the third pixel point is a pixel point adjacent to the left side of the second pixel point, the fourth pixel point is a pixel point adjacent to the right side of the second pixel point, when the mark values of the first pixel point, the second pixel point, the third pixel point and the fourth pixel point are all 0, the valid point is given a marker value that is distinct from the previously marked valid point and the marker value of the current valid point is stored in the equivalence pair array, otherwise, selecting a marking value of a pixel point with a numerical value not being 0 from the four pixel points according to the sequence of the first pixel point, the second pixel point, the third pixel point and the fourth pixel point to be endowed with an effective point; and when the marking values of the first pixel point and the fourth pixel point are further judged to be not 0 and unequal at the same time, storing the marking values of the first pixel point and the fourth pixel point in an equivalent array as an equivalent pair, otherwise, further judging whether the marking values of the third pixel point and the fourth pixel point are not 0 and unequal at the same time, and if the marking values of the third pixel point and the fourth pixel point are judged to be not 0 and unequal at the same time, storing the third pixel point and the fourth pixel point in the equivalent array as the equivalent pair.
Preferably, the step of updating the flag value of the same equivalence chain in the equivalence array to a uniform flag value according to a second set rule includes: and updating the marking values of the pixel points corresponding to the marking values in the equivalent chain in the equivalent array to the final value of the equivalent chain so as to ensure that the updated marking values of the same connected domain are the same.
Preferably, in the step of determining whether the pixel point of the connected domain meets the set condition, the method includes: counting the number of pixel points of the connected domain, wherein when the number of the pixel points exceeds a set number level, the connected domain is the outline of the checkerboard; or recording the maximum value Xmax and the minimum value Xmin of the abscissa and the maximum value Ymax and the minimum value Ymin of the ordinate of the pixel point in the same connected domain, wherein according to the formula, G is (Xmax-Xmin) × (Ymax-Ymin), and the maximum value of the product G of the connected domain is determined as the contour of the checkerboard.
Preferably, in the step of performing corner point identification on the extracted contour of the checkerboard, the method includes: calculating the gradient I of pixel points I (X, Y) of the contour of the checkerboard in the X and Y directions by using a horizontal difference operator and a vertical difference operatorx、Iy, wherein ,calculating three products of gradients of two directions of the pixel points to obtain a matrix m: wherein , and performing Gaussian smoothing filtering on four elements of the matrix M to obtain a new matrix M: wherein ,calculating Harris response value R of each pixel point according to the formula of DetM- α (traceM)2Wherein detM ═ λ1λ2＝AC-B2，traceM＝λ1+λ2And (3) when the sum of A and C is α is 0.1, carrying out R value non-maximum value suppression in the 3 x 3 neighborhood of the pixel point to obtain a maximum value point which is the corner point.
Preferably, the step of screening out the inner corner points of the checkerboard according to the identified corner points comprises: traversing the angular points of the checkerboard from the set angular points, calculating the distances between the currently selected angular point and other angular points, and finding out four angular points closest to the current angular point by adopting a bubbling sequencing method; and calculating the variance of the distance values between the four corners and the current corner, wherein if the variance is smaller than a second set threshold, the current corner is the inner corner of the checkerboard.
According to a second aspect of the present application, the present application provides a checkerboard corner detection apparatus based on contour extraction, the apparatus comprising: the edge point acquisition module is used for carrying out edge detection processing on the original image so as to acquire edge points in the original image; the contour extraction module is used for extracting the contour of the checkerboard in the original image according to the edge points; the corner point identification module is used for carrying out corner point identification on the extracted contour of the checkerboard; and the internal corner screening module is used for screening out the internal corners of the checkerboard according to the identified corners.
According to a third aspect of the present application, there is provided a terminal comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to: carrying out edge detection processing on the original image to obtain edge points in the original image; extracting the contour of the checkerboard in the original image according to the edge points; performing corner point identification on the extracted contour of the checkerboard; and screening out the inner corner points of the checkerboard according to the identified corner points.
According to a fourth aspect of the present application, there is provided a computer readable storage medium having stored thereon a computer program which, when executed by a processor, carries out the steps of the method as described above.
The beneficial effect of this application lies in: according to the method, the edge points in the original image are obtained by performing edge detection on the original image, the contour of the checkerboard in the original image is extracted according to the edge points, then the extracted contour of the checkerboard is subjected to corner identification, and the internal corners of the checkerboard can be preferably selected according to the identified corners.
Drawings
FIG. 1 is an original image of the present application;
FIG. 2 is an identification effect of the gray-scale-based corner detection algorithm after the corner detection of FIG. 1 is performed using Harris operator;
FIG. 3 is an identification effect of the gray-based corner detection algorithm after the corner detection of FIG. 1 is performed using Susan operator;
FIG. 4 is a flow chart of a tessellation corner detection method based on contour extraction according to the present application;
FIG. 5 is an effect diagram of the present application, in which an original image is subjected to edge detection processing by a Sobel operator;
FIG. 6 is a schematic diagram of pixel points in the 3 × 3 neighborhood of the current active point of the present application;
FIG. 7 is a flowchart of the operation of step S1021 of the present application;
fig. 8 is an effect diagram of the present application extracting the contour of the checkerboard in the original image through step S102;
fig. 9 is an effect diagram of the present application on corner identification in the outline of the checkerboard through step S103;
fig. 10 is an effect diagram of the present application screening out an inner corner point for the identified corner point of the checkerboard through step S104; and
fig. 11 is a schematic diagram of a checkerboard corner detection apparatus based on contour extraction according to the present application.
Description of reference numerals: edge point acquisition module 111 contour extraction module 112 corner point identification module 113 internal corner point screening module 114.
Detailed Description
The present application will be described in further detail below with reference to the accompanying drawings by way of specific embodiments.
The conception of the application is as follows: a large number of non-angular point pixels are removed by extracting the checkerboard outline, so that the data processing amount can be greatly reduced, the processing speed and efficiency are improved, and the anti-interference capability and accuracy are effectively improved.
Referring to fig. 1 to 10, the present application provides a method for detecting a corner point of a checkerboard based on contour extraction, the method comprising:
step S101: and carrying out edge detection processing on the original image to acquire edge points in the original image.
In this embodiment, the edge detection processing is performed on the original image by using a Sobel operator. And the Sobel operator detects the edge when the edge reaches an extreme value according to the gray weighting difference of adjacent points in the neighborhood of the pixel point 8.
Therefore, in step S101, the method includes:
step S1011: by a transverse convolution factor SxAnd a longitudinal convolution factor SyPerforming planar convolution with the original image to obtain a horizontal gray difference approximate value G of the pixel points of the original imagexAnd longitudinal gray difference approximate value GyWherein the original image is set as A, and the transverse convolution factorLongitudinal convolution factorAccording to the formula: the lateral gray difference approximation is Gx＝SxA, longitudinal gray difference approximation Gy＝Sy*A。
Step S1012: approximate value G of transverse gray difference of each pixel point in original image AxAnd a longitudinal gray difference approximation GyObtaining the gray weighting difference G of the pixel point through a formula (1), wherein the formula (1) is as follows:to improve efficiency, it can also be obtained by equation (2) without squaring: g ═ Gx|+|Gy|。
Step S1013: and when the gray weighting difference G of the pixel point is greater than a first set threshold value, the pixel point is an edge point, otherwise, the pixel point is marked to be 0.
In this embodiment, the first setting threshold may be set according to an actual usage scenario, which is not limited herein.
Referring to fig. 5, fig. 5 is a diagram illustrating an effect of performing an edge detection process on an original image by using a Sobel operator. The Sobel operator has a simple principle and a small calculation amount, and is a commonly used edge detection method when the requirement on precision is not very high. According to the method, for the characteristic that the angular points of the checkerboard are edge inflection points or intersection points, a Sobel operator is selected for edge detection, most of non-angular points are eliminated, and the data volume of subsequent processing is greatly reduced.
Step S102: and extracting the contour of the checkerboard in the original image according to the edge points.
In this step, the contour of the checkerboard is extracted from the edge points obtained in step S101. The design idea of the step is to classify the outlines by adopting a connected domain marking method according to the connectivity of the checkerboard outlines, and then extract the outlines by combining the characteristics of the checkerboard outlines.
In step S102, the method specifically includes the following steps:
step S1021: scanning pixel points of an original image according to a set sequence, when the scanned pixel points in the original image are effective points, giving a set marking value to the effective points in the original image according to a first set rule, wherein the marking value of the effective points in the same connected domain is represented in an equivalent chain mode, the equivalent chain comprises marking values with equivalent relations, and the equivalent chain is stored in an equivalent array in an equivalent pair mode. And updating the marking value of the same equivalence chain in the equivalence array into a uniform marking value according to a second set rule, wherein the pixel points with the same marking value are the pixel points of the same connected domain.
In this embodiment, the related content of the connected domain is defined first.
The equivalent pair is as follows: and A and B are mark values, the equivalence pair (A and B) indicates that the pixels marked with the mark values A and B belong to the same connected domain, namely, the pixels marked with the mark values A and B have equivalence relation, the mark values with the equivalence relation are stored in an equivalence array or a labelpair array, and labelpair [ A ] ═ B.
Equivalent chain: the tag values in the equivalence chains are all the same connected domain, which is stored in equivalence pairs. For example, (2, 3, 4, 5) in the equivalence chain indicates that the pixels with label values of 2, 3, 4, 5 are pixels of the same connected domain, and the equivalence chain is stored as equivalence pairs (2, 3), (3, 4), (4, 5), (5, 5), wherein the equivalence pair (5, 5) indicates the chain end of the equivalence chain, and 5 is the final value.
Equivalent array: the equivalent array may be a labelpari array, which is a one-dimensional array storing equivalent pairs, and for example, the equivalent pair (2, 3) is stored as labelpair [2] ═ 3. For example, in the array, labelpair [2] ═ 3, labelpair [3] ═ 4, labelpair [4] ═ 5, indicate that the equivalent pair (2, 3), (3, 4), (4, 5), (5, 5) belong to the same equivalent chain, (2, 3, 4, 5), labelpair [1] ═ 6, indicate that the equivalent pair (1, 6) belongs to the same equivalent chain, (1, 6), and it can be seen that two different connected domains are indicated by two different equivalent chains in one array.
In the two-step method for labeling the connected domain, the equivalent array is realized by a chain structure, the label belonging to the same connected domain is represented by one equivalent array, only one-dimensional array is needed for storage, compared with the two-dimensional array of a tree structure, the storage of the equivalent array is more convenient, the normalization and the traceability of complex contours are better, and the subsequent equivalent pair and the update of image labels are convenient.
In the present embodiment, scanning is performed in the order of column by column and row by row starting from the upper left corner of the image.
Therefore, referring to fig. 6 and 7, the step of assigning the set mark value to the effective point in the original image according to the first setting rule includes:
step S1021 a: and judging whether the marking values of the pixel points in the neighborhood of the current effective point are all 0. Referring to fig. 6, in the current effective point a, the pixels in the neighborhood of the current effective point 3 × 3 include a first pixel a1, a second pixel a2, a third pixel a3, and a fourth pixel a4, where the first pixel a1 is a pixel adjacent to the left of the current effective point, the second pixel a2 is a pixel adjacent to the top of the first pixel a1, the third pixel a3 is a pixel adjacent to the left of the second pixel, and the fourth pixel a4 is a pixel adjacent to the right of the second pixel. When the flag values of the first pixel point a1, the second pixel point a2, the third pixel point a3 and the fourth pixel point a4 are all 0, the step S1021b is skipped, otherwise, the step S1021c is skipped.
Step S1021 b: the valid point is given a marker value that is distinct from the previously marked valid point, and the marker value for the current valid point is stored in the equivalence pair array. For example, if the previous pixel has been given tag values 1, 2, and 3, a new tag value, for example, 4, is given to the previous pixel and stored in the equivalent array, that is, labelpair [4] ═ 4.
Step S1021 c: according to the sequence of the first pixel point a1, the second pixel point a2, the third pixel point a3 and the fourth pixel point a4, a marking value of a pixel point with the numerical value not being 0 is selected from the four pixel points to be endowed to an effective point.
Step S1021 d: and when the marking values of the first pixel point a1 and the fourth pixel point a4 are judged to be not 0 and not equal at the same time, if so, jumping to the step S1021e, and otherwise, jumping to the step S1021 f.
Step S1021 e: and storing the marking values of the first pixel point a1 and the fourth pixel point a4 as equivalent pairs in an equivalent array. Also, labelpair [ a1] ═ a 4.
Step S1021 f: and further judging whether the marking values of the third pixel point a3 and the fourth pixel point a4 are not 0 and are not equal at the same time, if so, jumping to the step S1021g, and if not, keeping the marking values of the third pixel point a3 and the fourth pixel point a4 unchanged.
Step S1021 g: and storing the third pixel point a3 and the fourth pixel point a4 as equivalent pairs in an equivalent array. Also, labelpair [ a4] ═ a 3.
In this embodiment, the second setting rule is to update the mark value of the pixel point corresponding to each mark value in the equivalence chain in the equivalence array to the last value of the equivalence chain, and the updated mark values of the same connected domain are the same. For example, in the equivalence chain (2, 3, 4, 5), the equivalence pair stored in the array by the equivalence chain 1 includes: labelpair [2] ═ 3, labelpair [3] ═ 4, labelpair [4] ═ 5, labelpair [5], since the end value of the chain is 5, the values of the equivalence pairs are all set to 5, or labelpair [2] ═ 5, labelpair [3] ═ 5, labelpair [4] ═ 5, labelpair [5], and thus the label values of the pixels corresponding to label values 2, 3, 4, and 5 are all set to 5. Therefore, the pixels with the same mark value are in the same connected region.
Step S1022: and judging whether the pixel points of the connected domain meet the set conditions or not, and if the parameters of the pixel points of the connected domain meet the set conditions, extracting the connected domain as the outline of the checkerboard.
The pixel points of the checkerboard connected domain, the interference points and the line segments are not in the same order of magnitude when the checkerboard features are seen from the connected domain, and the checkerboard occupies most of the image when seen from a use scene, so that the outline of the checkerboard can be judged by judging the pixel points of the connected domain or the occupied range of the connected domain in the image.
In this embodiment, in step S1022, it is determined whether the connected region is the checkerboard outline by counting the number of the pixels in the connected region, and when the number of the pixels exceeds the set number level, the connected region is the checkerboard outline. For example, if the number of the pixels in the connected domain exceeds 10000, the connected domain is considered as the checkerboard outline.
In other embodiments, determining whether the connected component is a checkerboard outline may also be determined by determining the range occupied by the connected component in the image: recording the maximum value Xmax and the minimum value Xmin of the abscissa and the maximum value Ymax and the minimum value Ymin of the ordinate of the pixel points in the same connected domain, wherein according to a formula, G is (Xmax-Xmin) × (Ymax-Ymin), and the maximum value of the product G of the connected domains is determined as the contour of the checkerboard.
Referring to fig. 8, fig. 8 is a diagram illustrating the effect of extracting the contour of the checkerboard in the original image through step S102.
Step S103: and performing corner point identification on the extracted contour of the checkerboard.
In this embodiment, the contour of the checkerboard is subjected to corner point identification by the Harris operator.
Therefore, in step S103, the method includes:
step S1031: calculating the gradient I of pixel points I (X, Y) of the contour of the checkerboard in the X and Y directions by using a horizontal difference operator and a vertical difference operatorx、Iy, wherein ,
step S1032: calculating three products of gradients of two directions of the pixel points to obtain a matrix m: wherein ,Ixy＝Ix·Iy。
step S1033: and performing Gaussian smoothing filtering on four elements of the matrix M to obtain a new matrix M: wherein ,
step S1034, calculating a Harris response value R of each pixel point according to a formula of R ═ detM- α (traceM)2Wherein detM ═ λ1λ2＝AC-B2，traceM＝λ1+λ2＝A+C，α＝0.1。
Step S1035: and carrying out R value non-maximum suppression in a3 x 3 neighborhood of the pixel points to obtain maximum points, namely the corner points.
It can be seen that compared with the Susan operator, the Harris operator has the advantages of simple principle, good stability and wider application when used for performing corner identification of the checkerboard outline.
With continuing reference to fig. 9, fig. 9 is a diagram illustrating the effect of identifying the corner points in the contour of the checkerboard by step S103. It can be seen that in the figure, the corner points of the squares of the checkerboard are all identified.
Step S104: and screening out the inner corner points of the checkerboard according to the identified corner points.
Because the inner corner features of the checkerboard are: in the four directions of the inner corner points, there are four corner points, namely, an upper corner point, a lower corner point, a left corner point, a right corner point, and a distance value which is closest to the four corner points, namely, the variance of the four distance values is small, and if the four distance values are not the inner corner points, the condition cannot be met. Therefore, the inner corner point can be found by judging whether four corner points with the same distance value exist simultaneously in the up, down, left and right directions of the corner point.
Therefore, in this embodiment, in step S104, the method specifically includes:
step S1041: traversing the angular points of the checkerboard from the set angular points, calculating the distances between the currently selected angular point and other angular points, and finding out four angular points closest to the current angular point by adopting a bubbling sequencing method;
step S1042: and calculating the variance of the distance values between the four corners and the current corner, wherein if the variance is smaller than a second set threshold, the current corner is the inner corner of the checkerboard.
Referring to fig. 10, fig. 10 is an effect diagram of screening out an inner corner point for the identified corner point of the checkerboard through step S104. It can be seen that the inner corners of the squares of the checkerboard are all identified in the figure.
The present application can be compared to a corner detection algorithm that uses Harris operators to directly identify corners. In the operation speed, the simulation is carried out through matlab to verify, the two adopt the same Harris code, the test is carried out under the same environment, four sample drawings are selected at will, the corner detection algorithm for directly identifying the corners by using the Harris operator and the simulation time data of the application are shown in the following table, and it can be seen that the operation speed of the application is far faster than the corner detection algorithm for directly identifying the corners by using the Harris operator.
| Unit: second of | Harris | This application |
| Sample FIG. 1 | 4.3554 | 0.3136 |
| Sample 2 | 4.3190 | 0.3464 |
| Same figure 3 | 4.3316 | 0.3319 |
| Sample 4 | 4.4103 | 0.3348 |
TABLE 1 comparison of recognition velocity using Harris operator corner detection algorithm with the present application
In terms of accuracy, the recognition effect of the corner detection algorithm for directly recognizing the corners by using the Harris operator is shown in fig. 2, the effect of recognizing the corners by using the algorithm is shown in fig. 9, and obviously, the recognition effect of the algorithm is better, and many results which are not the corners appear in the recognition result of fig. 2.
In conclusion, compared with the corner detection algorithm for directly identifying the corners by using the Harris operator, the speed and the accuracy of identification are greatly improved.
Correspondingly, the invention also provides a contour extraction-based checker corner detection device, which corresponds to the embodiments of contour extraction-based checker corner detection methods shown in fig. 4 and 7, according to the functional modular thinking of computer software. Referring to fig. 11, the modules included in the apparatus and the specific functions implemented by the modules are specifically disclosed below.
The checkerboard corner detection device based on contour extraction comprises: an edge point obtaining module 111, configured to perform edge detection processing on the original image to obtain edge points in the original image; the contour extraction module 112 is used for extracting the contour of the checkerboard in the original image according to the edge points; the corner identification module 113 is used for performing corner identification on the extracted contour of the checkerboard; an inner corner screening module 114, configured to screen out inner corners of the checkerboard according to the identified corners.
The edge point obtaining module 111 is further configured to: by a transverse convolution factor SxAnd a longitudinal convolution factor SyPerforming plane convolution with the original image to obtain a horizontal gray difference approximate value G of the pixel point of the original imagexAnd longitudinal gray difference approximate value GyWherein the original image is set toA, transverse convolution factorLongitudinal convolution factorAccording to the formula: the lateral gray-scale difference approximation is Gx ═ Sx a, and the longitudinal gray-scale difference approximation is Gy＝SyA; approximate value G of transverse gray difference of each pixel point in original image AxAnd a longitudinal gray difference approximation GyObtaining the gray scale weighting difference G of the pixel point according to the following formula (1) or formula (2), wherein the formula (1) is as follows:the formula (2) is: g ═ Gx|+|GyL, |; and when the gray weighting difference G of the pixel point is greater than a first set threshold value, the pixel point is an edge point, otherwise, the pixel point is marked to be 0.
The contour extraction module 112 includes:
the connected domain marking unit is used for scanning pixel points of an original image according to a set sequence, when the scanned pixel points in the original image are effective points, a set marking value is given to the effective points in the original image according to a first set rule, the marking values of the effective points in the same connected domain are expressed in a marking chain mode, the marking chain stores the marking values with an equivalent relationship in an equivalent array in an equivalent pair mode, the marking values of the same marking chain in the equivalent array are updated to be uniform marking values according to a second set rule, and the pixel points with the same marking values are the pixel points in the same connected domain;
and the extraction unit is used for judging whether the pixel points of the connected domain meet the set conditions or not, and if the parameters of the pixel points of the connected domain meet the set conditions, extracting the connected domain as the outline of the checkerboard.
The connected component marking unit is also used for: judging whether the mark values of the pixel points in the neighborhood of the current effective point are all 0, wherein the pixel points in the neighborhood comprise a first pixel point, a second pixel point, a third pixel point and a fourth pixel point, the first pixel point is a pixel point adjacent to the left side of the current effective point, the second pixel point is a pixel point adjacent to the upper part of the first pixel point, the third pixel point is a pixel point adjacent to the left side of the second pixel point, the fourth pixel point is a pixel point adjacent to the right side of the second pixel point, when the mark values of the first pixel point, the second pixel point, the third pixel point and the fourth pixel point are all 0, the valid point is given a marker value that is distinct from the previously marked valid point and the marker value of the current valid point is stored in the equivalence pair array, otherwise, selecting a marking value of a pixel point with a numerical value not being 0 from the four pixel points according to the sequence of the first pixel point, the second pixel point, the third pixel point and the fourth pixel point to be endowed with an effective point; and when the marking values of the first pixel point and the fourth pixel point are further judged to be not 0 and unequal at the same time, storing the marking values of the first pixel point and the fourth pixel point in an equivalent array as an equivalent pair, otherwise, further judging whether the marking values of the third pixel point and the fourth pixel point are not 0 and unequal at the same time, and if the marking values of the third pixel point and the fourth pixel point are judged to be not 0 and unequal at the same time, storing the third pixel point and the fourth pixel point in the equivalent array as the equivalent pair.
The connected component marking unit is also used for: and updating the marking value of the pixel point corresponding to each marking value in the marking chain in the equivalent array to the last value of the marking chain so as to ensure that the updated marking values of the same connected domain are the same.
The extraction unit is further configured to: counting the number of pixel points of the connected domain, wherein when the number of the pixel points exceeds a set number level, the connected domain is the outline of the checkerboard; or recording the maximum value Xmax and the minimum value Xmin of the abscissa and the maximum value Ymax and the minimum value Ymin of the ordinate of the pixel point in the same connected domain, wherein according to the formula, G is (Xmax-Xmin) × (Ymax-Ymin), and the maximum value of the product G of the connected domain is determined as the contour of the checkerboard.
The corner identification module 113 is further configured to: by using waterCalculating the gradient I of pixel points I (X, Y) of the contour of the checkerboard in the X and Y directions by using an average difference operator and a vertical difference operatorx、Iy, wherein ,calculating three products of gradients of two directions of the pixel points to obtain a matrix m: wherein ,Ixy＝Ix·Iy(ii) a And performing Gaussian smoothing filtering on four elements of the matrix M to obtain a new matrix M: wherein ,calculating Harris response value R of each pixel point according to the formula of DetM- α (traceM)2Wherein detM ═ λ1λ2＝AC-B2，traceM＝λ1+λ2And (3) when the sum of A and C is α is 0.1, carrying out R value non-maximum value suppression in the 3 x 3 neighborhood of the pixel point to obtain a maximum value point which is the corner point.
The internal corner screening module 114 is further configured to: traversing the angular points of the checkerboard from the set angular points, calculating the distances between the currently selected angular point and other angular points, and finding out four angular points closest to the current angular point by adopting a bubbling sequencing method; and calculating the variance of the distance values between the four corners and the current corner, wherein if the variance is smaller than a second set threshold, the current corner is the inner corner of the checkerboard.
The working principle of the contour extraction-based checkerboard corner detection method of the present application is described below with reference to fig. 1 to 11.
Firstly, edge detection is carried out on an original image through a Sobel operator to obtain edge points in the original image.
Then, the edge points are used as effective points, and the edge points are labeled. The method comprises the following steps: scanning is carried out from the upper left corner of the image according to a column-by-column line-by-line sequence, whether the marking values of the pixel points in the 3 x 3 neighborhood of the current effective point are all 0 is judged, if all 0, a new marking value is given to the effective point, the new marking value is stored in a new equivalence chain, and otherwise, the marking value of a pixel point with the numerical value not 0 is selected from the four pixel points according to the sequence of a first pixel point a1, a second pixel point a2, a third pixel point a3 and a fourth pixel point a4 and is given to the effective point. And when the marking values of the first pixel point a1 and the fourth pixel point a4 are judged to be not 0 and not equal at the same time, storing the marking values of the first pixel point a1 and the fourth pixel point a4 in an equivalent array as an equivalent pair. And then, when judging whether the marking values of the third pixel point a3 and the fourth pixel point a4 are not 0 and are not equal at the same time, storing the third pixel point a3 and the fourth pixel point a4 as an equivalent pair in an equivalent array.
And after the scanning is finished, updating all the marking values in the equivalent chain into the tail values of the equivalent chain, so that the pixel points with the same marking value belong to the same connected region. Because the labeled values of the same equivalence chain belong to the same connected region, after the scanning is completed, a plurality of equivalence chains are possible, that is, a plurality of connected regions are represented.
According to the characteristics of the connected region, if the number of the pixel points of the connected region exceeds a set number, such as 10000, the connected region is considered as the outline of the checkerboard. And identifying the corner points of the contour of the checkerboard, and screening out the inner corner points from the identified corner points.
The present disclosure further provides a terminal, including: a processor; a memory for storing processor-executable instructions; the processor is configured to perform edge detection processing on the original image to acquire edge points in the original image; extracting the contour of the checkerboard in the original image according to the edge points; performing corner point identification on the extracted contour of the checkerboard; and screening out the inner corner points of the checkerboard according to the identified corner points.
The steps in the various method embodiments described above, such as the steps shown in fig. 4, are implemented when the computer program is executed by a processor. Alternatively, the processor, when executing the computer program, implements the functions of the modules/units in the above-mentioned device embodiments, such as the functional modules in fig. 11.
Illustratively, the computer program may be partitioned into one or more modules/units that are stored in the memory and executed by the processor to implement the invention. The one or more modules/units may be a series of instruction segments of a computer program capable of performing a specific function, and the instruction segments are used for describing the execution process of the computer program in the contour extraction based checkerboard corner detection device. For example, the computer program may be divided into modules as shown in fig. 11, and the specific functions of each module are as described above.
The checkerboard angular point detection device based on contour extraction can be computing equipment such as a desktop computer, a notebook computer, a palm computer and a cloud server, and can also be shooting equipment such as a vehicle event data recorder and a motion camera. The contour extraction based checkerboard corner detection apparatus/terminal device may include, but is not limited to, a processor, and a memory. It will be understood by those skilled in the art that the schematic diagram is merely an example of the contour extraction based tessellation corner detection apparatus, and does not constitute a limitation of the contour extraction based tessellation corner detection apparatus, and may include more or less components than those shown, or combine some components, or different components, for example, the contour extraction based tessellation corner detection apparatus may further include an input-output device, a network access device, a bus, and the like.
The Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable gate array (FPGA) or other Programmable logic device, discrete gate or transistor logic device, discrete hardware component, or the like. The general processor may be a microprocessor or the processor may be any conventional processor, etc., and the processor is a control center of the contour extraction based checkerboard corner detection apparatus, and various interfaces and lines are used to connect various parts of the contour extraction based checkerboard corner detection apparatus.
The memory may be configured to store the computer program and/or the module, and the processor may implement various functions of the contour extraction based checkerboard corner detection apparatus by running or executing the computer program and/or the module stored in the memory and calling data stored in the memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data (such as audio data, a phonebook, etc.) created according to the use of the cellular phone, and the like. In addition, the memory may include high speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.
The present disclosure proposes a computer-readable storage medium, in which a computer program is stored, which, when being executed by a processor, implements the steps of the method for tessellation corner detection based on contour extraction as described above.
The modules/units of the contour extraction-based checkerboard corner detection device can be stored in a computer-readable storage medium if the modules/units are implemented in the form of software functional units and sold or used as independent products. Based on such understanding, all or part of the flow of the method according to the embodiments of the present invention may also be implemented by a computer program, which may be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the method embodiments may be implemented. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, etc. It should be noted that the computer readable medium may contain content that is subject to appropriate increase or decrease as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable media does not include electrical carrier signals and telecommunications signals as is required by legislation and patent practice.
The beneficial effect of this application lies in: according to the method, the edge points in the original image are obtained by performing edge detection on the original image, the contour of the checkerboard in the original image is extracted according to the edge points, then the extracted contour of the checkerboard is subjected to corner identification, and the internal corners of the checkerboard can be preferably selected according to the identified corners.
The foregoing is a more detailed description of the present application in connection with specific embodiments thereof, and it is not intended that the present application be limited to the specific embodiments thereof. It will be apparent to those skilled in the art from this disclosure that many more simple derivations or substitutions can be made without departing from the inventive concepts herein.
Claims (10)
1. A checkerboard corner detection method based on contour extraction is characterized by comprising the following steps:
carrying out edge detection processing on an original image to obtain edge points in the original image;
extracting the contour of the checkerboard in the original image according to the edge points;
performing corner point identification on the extracted contour of the checkerboard;
and screening out the inner corner points of the checkerboard according to the identified corner points.
2. The checkerboard corner detection method as claimed in claim 1, wherein in said step of performing edge detection processing on the original image to obtain edge points in the original image, the method comprises:
by a transverse convolution factor SxAnd a longitudinal convolution factor SyPerforming plane convolution with the original image to obtain a horizontal gray difference approximate value G of a pixel point of the original imagexAnd longitudinal gray difference approximate value GyWherein the original image is set as A, a transverse convolution factorLongitudinal convolution factorAccording to the formula: the lateral gray difference approximation is Gx＝SxA, longitudinal gray difference approximation Gy＝Sy*A；
Obtaining a horizontal gray difference approximate value G of each pixel point in the original image AxAnd a longitudinal gray difference approximation GyObtaining the gray scale weighting difference G of the pixel point according to the following formula (1) or formula (2), wherein the formula (1) is as follows:the formula (2) is: g ═ Gx|+|Gy|；
And when the gray weighting difference G of the pixel point is greater than a first set threshold value, the pixel point is the edge point, otherwise, the pixel point is marked as 0.
3. The method for detecting angular points of checkerboards as claimed in claim 1, wherein said step of extracting the contour of the checkerboard in the original image according to the edge points comprises:
scanning pixel points of the original image according to a set sequence, when the scanned pixel points in the original image are effective points, namely edge points, giving set marking values to the effective points in the original image according to a first set rule, wherein the marking values of the effective points in the same connected domain are represented in an equivalence chain mode, the equivalence chain comprises marking values with the equivalence relation, the equivalence chain is stored in an equivalence array in an equivalence pair mode, the marking values of the same equivalence chain in the equivalence array are updated to be uniform marking values according to a second set rule, and then the pixel points with the same marking values are the pixel points of the same connected domain;
and judging whether the pixel points of the connected domain meet set conditions or not, and if the parameters of the pixel points of the connected domain meet the set conditions, extracting the connected domain as the outline of the checkerboard.
4. The method for detecting a checkerboard corner point of claim 3, wherein said step of assigning a set mark value to a valid point in the original image according to a first set rule comprises the steps of:
judging whether the marking values of the pixels in the neighborhood of the current effective point are all 0, wherein the pixels in the neighborhood comprise a first pixel, a second pixel, a third pixel and a fourth pixel, the first pixel is a pixel adjacent to the left side of the current effective point, the second pixel is a pixel adjacent to the upper side of the first pixel, the third pixel is a pixel adjacent to the left side of the second pixel, and the fourth pixel is a pixel adjacent to the right side of the second pixel, when the marking values of the first pixel, the second pixel, the third pixel and the fourth pixel are all 0, the effective point is endowed with the marking value different from the effective point marked before, and the marking value of the current effective point is stored in the equivalence pair array, otherwise, according to the first pixel, the marking values of the effective point are all 0, Selecting a marking value of a pixel point with a numerical value not being 0 from the four pixel points according to the sequence of the second pixel point, the third pixel point and the fourth pixel point, and endowing the marking value to the effective point;
and when the marking values of the first pixel point and the fourth pixel point are further judged to be not 0 and not equal at the same time, storing the marking values of the first pixel point and the fourth pixel point in the equivalent array as an equivalent pair, otherwise, further judging whether the marking values of the third pixel point and the fourth pixel point are not 0 and not equal at the same time, and if the marking values of the third pixel point and the fourth pixel point are judged to be not 0 and not equal at the same time, storing the third pixel point and the fourth pixel point in the equivalent array as an equivalent pair.
5. The method as claimed in claim 3, wherein said step of updating the flag values of the same equivalence chain in the equivalence array to uniform flag values according to a second set rule comprises:
and updating the mark values of the pixel points corresponding to the mark values in the equivalent chain in the equivalent array to the last value of the equivalent chain so as to ensure that the updated mark values of the same connected domain are the same.
6. The checkerboard corner detection method of claim 3, wherein in said step of determining whether a pixel point of a connected domain satisfies a set condition, comprising:
counting the number of the pixel points of the connected domain, wherein when the number of the pixel points exceeds a set number level, the connected domain is the outline of the checkerboard; or,
recording the maximum value Xmax and the minimum value Xmin of the abscissa and the maximum value Ymax and the minimum value Ymin of the ordinate of the pixel points in the same connected domain, wherein according to a formula, G is (Xmax-Xmin) × (Ymax-Ymin), and the maximum value of the product G of the connected domain is determined as the outline of the checkerboard.
7. The method for detecting the corner points of a checkerboard as claimed in claim 1, wherein said step of performing the corner point identification on the extracted contour of said checkerboard comprises:
calculating the gradient I of the pixel point I (X, Y) of the contour of the checkerboard in the X direction and the Y direction by using a horizontal difference operator and a vertical difference operatorx、Iy, wherein ,
calculating three products of gradients of the two directions of the pixel points to obtain a matrix m: wherein ,Ixy＝Ix·Iy；
and performing Gaussian smoothing filtering on four elements of the matrix M to obtain a new matrix M: wherein ,
calculating Harris response value R of each pixel point according to the formula of R ═ detM- α (traceM)2Wherein detM ═ λ1λ2＝AC-B2，traceM＝λ1+λ2＝A+C，α＝0.1；
And performing R value non-maximum suppression in a3 x 3 neighborhood of the pixel point to obtain a maximum point, namely the corner point.
8. The method for detecting angular points of a checkerboard as claimed in claim 1, wherein said step of screening out the internal angular points of said checkerboard based on the identified angular points comprises:
traversing the corner points of the chessboard from the set corner points, calculating the distances between the currently selected corner points and other corner points, and finding out the four corner points closest to the current corner points by adopting a bubbling sequencing method;
and calculating the variance of the distance values between the four corner points and the current corner point, wherein if the variance is smaller than a second set threshold, the current corner point is the inner corner point of the checkerboard.
9. A checkerboard corner detection device based on contour extraction is characterized in that the device comprises:
the edge point acquisition module is used for carrying out edge detection processing on an original image so as to acquire edge points in the original image;
the contour extraction module is used for extracting the contour of the checkerboard in the original image according to the edge points;
the corner identification module is used for carrying out corner identification on the extracted contour of the checkerboard;
and the internal corner screening module is used for screening out the internal corners of the checkerboard according to the identified corners.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811601937.9A CN109509200B (en) | 2018-12-26 | 2018-12-26 | Checkerboard corner detection method based on contour extraction and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811601937.9A CN109509200B (en) | 2018-12-26 | 2018-12-26 | Checkerboard corner detection method based on contour extraction and computer readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109509200A true CN109509200A (en) | 2019-03-22 |
| CN109509200B CN109509200B (en) | 2023-09-29 |
Family
ID=65755323
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811601937.9A Active CN109509200B (en) | 2018-12-26 | 2018-12-26 | Checkerboard corner detection method based on contour extraction and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109509200B (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110348263A (en) * | 2019-06-24 | 2019-10-18 | 西安理工大学 | A kind of two-dimensional random code image recognition and extracting method based on image recognition |
| CN110648368A (en) * | 2019-08-30 | 2020-01-03 | 广东奥普特科技股份有限公司 | Calibration board corner point discrimination method based on edge features |
| CN111553927A (en) * | 2020-04-24 | 2020-08-18 | 厦门云感科技有限公司 | Checkerboard corner detection method, checkerboard corner detection system, computer device and storage medium |
| CN111681284A (en) * | 2020-06-09 | 2020-09-18 | 商汤集团有限公司 | A corner detection method, device, electronic device and storage medium |
| CN112017218A (en) * | 2020-09-09 | 2020-12-01 | 杭州海康威视数字技术股份有限公司 | Image registration method and device, electronic equipment and storage medium |
| WO2021147755A1 (en) * | 2020-01-20 | 2021-07-29 | 北京芯海视界三维科技有限公司 | Checkerboard corner detection method and device |
| CN113283416A (en) * | 2020-12-29 | 2021-08-20 | 深圳怡化电脑股份有限公司 | Character outline recognition method and device, electronic equipment and machine readable medium |
| CN113744177A (en) * | 2020-05-28 | 2021-12-03 | 中科寒武纪科技股份有限公司 | Method and device for detecting angular point of image and storage medium |
| CN113989384A (en) * | 2021-10-27 | 2022-01-28 | 杭州爱科科技股份有限公司 | Camera calibration method, cropping data generation method, device, device and medium |
| CN114445499A (en) * | 2020-10-19 | 2022-05-06 | 深圳市光鉴科技有限公司 | Checkerboard angular point automatic extraction method, system, equipment and medium |
| CN116030450A (en) * | 2023-03-23 | 2023-04-28 | 摩尔线程智能科技(北京)有限责任公司 | Checkerboard corner recognition method, device, equipment and medium |
| WO2024016686A1 (en) * | 2022-07-18 | 2024-01-25 | 宁德时代新能源科技股份有限公司 | Corner detection method and apparatus |
| CN118212210A (en) * | 2024-04-02 | 2024-06-18 | 常州信息职业技术学院 | Point symmetry-based automobile dipped beam type feature point detection method and system |
Citations (31)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009021089A (en) * | 2007-07-11 | 2009-01-29 | Nippon Zeon Co Ltd | Organic electroluminescence device and method for producing the same |
| CN101667288A (en) * | 2008-09-02 | 2010-03-10 | 新奥特（北京）视频技术有限公司 | Method for detecting corner points of communicated regions in binary symbol images |
| CN101667287A (en) * | 2008-09-02 | 2010-03-10 | 新奥特（北京）视频技术有限公司 | Method for detecting corner points of outermost frames of symbols in symbol images |
| CN103093451A (en) * | 2011-11-03 | 2013-05-08 | 北京理工大学 | Checkerboard intersection recognition algorithm |
| CN103177439A (en) * | 2012-11-26 | 2013-06-26 | 惠州华阳通用电子有限公司 | Automatically calibration method based on black and white grid corner matching |
| CN103345755A (en) * | 2013-07-11 | 2013-10-09 | 北京理工大学 | Chessboard angular point sub-pixel extraction method based on Harris operator |
| WO2013188980A1 (en) * | 2012-06-20 | 2013-12-27 | Cfs Concrete Forming Systems Inc. | Formwork apparatus having resilient standoff braces and methods related thereto |
| CN103489192A (en) * | 2013-09-30 | 2014-01-01 | 北京林业大学 | Method for detecting number of Arabidopsis leaves and distance between cusp and center of mass of each leaf |
| CN103593840A (en) * | 2013-09-30 | 2014-02-19 | 北京林业大学 | Method for detecting phenotype of Arabidopsis |
| CN103606127A (en) * | 2013-12-03 | 2014-02-26 | 中国科学院大学 | Anti-copying image watermarking method based on optical microstructure |
| CN103996191A (en) * | 2014-05-09 | 2014-08-20 | 东北大学 | Detection method for black and white checkerboard image corners based on least square optimization |
| CN104122271A (en) * | 2014-07-09 | 2014-10-29 | 宁波摩视光电科技有限公司 | Automated optical inspection (AOI)-based bullet apparent defect detection method |
| US20150145862A1 (en) * | 2013-11-27 | 2015-05-28 | Adobe Systems Incorporated | Texture Modeling of Image Data |
| CN104866856A (en) * | 2015-05-17 | 2015-08-26 | 西南石油大学 | Imaging log image solution cave information picking method based on connected domain equivalence pair processing |
| CN105006022A (en) * | 2015-08-11 | 2015-10-28 | 中山大学 | Simplified method and device for edge collapse of 3D geometry graphics |
| CN106097281A (en) * | 2016-06-27 | 2016-11-09 | 安徽慧视金瞳科技有限公司 | A kind of calibration maps for projecting interactive system and demarcate detection method |
| CN106204570A (en) * | 2016-07-05 | 2016-12-07 | 安徽工业大学 | A kind of angular-point detection method based on non-causal fractional order gradient operator |
| CN106340010A (en) * | 2016-08-22 | 2017-01-18 | 电子科技大学 | Corner detection method based on second-order contour difference |
| CN106378514A (en) * | 2016-11-22 | 2017-02-08 | 上海大学 | Stainless steel non-uniform tiny multi-weld-joint visual inspection system and method based on machine vision |
| CN106780537A (en) * | 2017-01-11 | 2017-05-31 | 山东农业大学 | A kind of paper cocooning frame silk cocoon screening plant and method based on image procossing |
| JP2017135587A (en) * | 2016-01-28 | 2017-08-03 | 東芝メディカルシステムズ株式会社 | Image processing device, image processing method, medical equipment and test method |
| CN107123146A (en) * | 2017-03-20 | 2017-09-01 | 深圳市华汉伟业科技有限公司 | The mark localization method and system of a kind of scaling board image |
| CN107749071A (en) * | 2017-09-12 | 2018-03-02 | 深圳市易成自动驾驶技术有限公司 | Big distortion gridiron pattern image angular-point detection method and device |
| CN108062759A (en) * | 2018-01-25 | 2018-05-22 | 华中科技大学 | A kind of more pixel-parallel labeling methods and system for being used to mark bianry image |
| US20180174328A1 (en) * | 2016-05-12 | 2018-06-21 | Huizhou University | Turning radius-based corner detection algorithm |
| CN108447095A (en) * | 2018-01-31 | 2018-08-24 | 潍坊歌尔电子有限公司 | A kind of fisheye camera scaling method and device |
| CN108734743A (en) * | 2018-04-13 | 2018-11-02 | 深圳市商汤科技有限公司 | Method, apparatus, medium and electronic equipment for demarcating photographic device |
| CN108895959A (en) * | 2018-04-27 | 2018-11-27 | 电子科技大学 | A kind of camera calibration plate angle point calculating method based on sub-pix |
| CN108898147A (en) * | 2018-06-27 | 2018-11-27 | 清华大学 | A kind of two dimensional image edge straightened method, apparatus based on Corner Detection |
| CN108986721A (en) * | 2018-06-14 | 2018-12-11 | 武汉精测电子集团股份有限公司 | A kind of test pattern generation method for display panel detection |
| CN109035320A (en) * | 2018-08-12 | 2018-12-18 | 浙江农林大学 | Depth extraction method based on monocular vision |
-
2018
- 2018-12-26 CN CN201811601937.9A patent/CN109509200B/en active Active
Patent Citations (31)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009021089A (en) * | 2007-07-11 | 2009-01-29 | Nippon Zeon Co Ltd | Organic electroluminescence device and method for producing the same |
| CN101667288A (en) * | 2008-09-02 | 2010-03-10 | 新奥特（北京）视频技术有限公司 | Method for detecting corner points of communicated regions in binary symbol images |
| CN101667287A (en) * | 2008-09-02 | 2010-03-10 | 新奥特（北京）视频技术有限公司 | Method for detecting corner points of outermost frames of symbols in symbol images |
| CN103093451A (en) * | 2011-11-03 | 2013-05-08 | 北京理工大学 | Checkerboard intersection recognition algorithm |
| WO2013188980A1 (en) * | 2012-06-20 | 2013-12-27 | Cfs Concrete Forming Systems Inc. | Formwork apparatus having resilient standoff braces and methods related thereto |
| CN103177439A (en) * | 2012-11-26 | 2013-06-26 | 惠州华阳通用电子有限公司 | Automatically calibration method based on black and white grid corner matching |
| CN103345755A (en) * | 2013-07-11 | 2013-10-09 | 北京理工大学 | Chessboard angular point sub-pixel extraction method based on Harris operator |
| CN103489192A (en) * | 2013-09-30 | 2014-01-01 | 北京林业大学 | Method for detecting number of Arabidopsis leaves and distance between cusp and center of mass of each leaf |
| CN103593840A (en) * | 2013-09-30 | 2014-02-19 | 北京林业大学 | Method for detecting phenotype of Arabidopsis |
| US20150145862A1 (en) * | 2013-11-27 | 2015-05-28 | Adobe Systems Incorporated | Texture Modeling of Image Data |
| CN103606127A (en) * | 2013-12-03 | 2014-02-26 | 中国科学院大学 | Anti-copying image watermarking method based on optical microstructure |
| CN103996191A (en) * | 2014-05-09 | 2014-08-20 | 东北大学 | Detection method for black and white checkerboard image corners based on least square optimization |
| CN104122271A (en) * | 2014-07-09 | 2014-10-29 | 宁波摩视光电科技有限公司 | Automated optical inspection (AOI)-based bullet apparent defect detection method |
| CN104866856A (en) * | 2015-05-17 | 2015-08-26 | 西南石油大学 | Imaging log image solution cave information picking method based on connected domain equivalence pair processing |
| CN105006022A (en) * | 2015-08-11 | 2015-10-28 | 中山大学 | Simplified method and device for edge collapse of 3D geometry graphics |
| JP2017135587A (en) * | 2016-01-28 | 2017-08-03 | 東芝メディカルシステムズ株式会社 | Image processing device, image processing method, medical equipment and test method |
| US20180174328A1 (en) * | 2016-05-12 | 2018-06-21 | Huizhou University | Turning radius-based corner detection algorithm |
| CN106097281A (en) * | 2016-06-27 | 2016-11-09 | 安徽慧视金瞳科技有限公司 | A kind of calibration maps for projecting interactive system and demarcate detection method |
| CN106204570A (en) * | 2016-07-05 | 2016-12-07 | 安徽工业大学 | A kind of angular-point detection method based on non-causal fractional order gradient operator |
| CN106340010A (en) * | 2016-08-22 | 2017-01-18 | 电子科技大学 | Corner detection method based on second-order contour difference |
| CN106378514A (en) * | 2016-11-22 | 2017-02-08 | 上海大学 | Stainless steel non-uniform tiny multi-weld-joint visual inspection system and method based on machine vision |
| CN106780537A (en) * | 2017-01-11 | 2017-05-31 | 山东农业大学 | A kind of paper cocooning frame silk cocoon screening plant and method based on image procossing |
| CN107123146A (en) * | 2017-03-20 | 2017-09-01 | 深圳市华汉伟业科技有限公司 | The mark localization method and system of a kind of scaling board image |
| CN107749071A (en) * | 2017-09-12 | 2018-03-02 | 深圳市易成自动驾驶技术有限公司 | Big distortion gridiron pattern image angular-point detection method and device |
| CN108062759A (en) * | 2018-01-25 | 2018-05-22 | 华中科技大学 | A kind of more pixel-parallel labeling methods and system for being used to mark bianry image |
| CN108447095A (en) * | 2018-01-31 | 2018-08-24 | 潍坊歌尔电子有限公司 | A kind of fisheye camera scaling method and device |
| CN108734743A (en) * | 2018-04-13 | 2018-11-02 | 深圳市商汤科技有限公司 | Method, apparatus, medium and electronic equipment for demarcating photographic device |
| CN108895959A (en) * | 2018-04-27 | 2018-11-27 | 电子科技大学 | A kind of camera calibration plate angle point calculating method based on sub-pix |
| CN108986721A (en) * | 2018-06-14 | 2018-12-11 | 武汉精测电子集团股份有限公司 | A kind of test pattern generation method for display panel detection |
| CN108898147A (en) * | 2018-06-27 | 2018-11-27 | 清华大学 | A kind of two dimensional image edge straightened method, apparatus based on Corner Detection |
| CN109035320A (en) * | 2018-08-12 | 2018-12-18 | 浙江农林大学 | Depth extraction method based on monocular vision |
Non-Patent Citations (2)
| Title |
|---|
| QILIN BI等: "An automatic camera calibration method based on checkerboard", 《TRAITEMENT DU SIGNAL》, vol. 34, no. 3, pages 209 - 226 * |
| 罗钧: "摄像机标定的棋盘格亚像素角点检测", 《重庆大学学报》, vol. 31, no. 6, pages 615 - 618 * |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110348263B (en) * | 2019-06-24 | 2022-09-30 | 西安理工大学 | Two-dimensional random code image identification and extraction method based on image identification |
| CN110348263A (en) * | 2019-06-24 | 2019-10-18 | 西安理工大学 | A kind of two-dimensional random code image recognition and extracting method based on image recognition |
| CN110648368A (en) * | 2019-08-30 | 2020-01-03 | 广东奥普特科技股份有限公司 | Calibration board corner point discrimination method based on edge features |
| CN110648368B (en) * | 2019-08-30 | 2022-05-17 | 广东奥普特科技股份有限公司 | A method for discriminating corner points of calibration board based on edge features |
| WO2021147755A1 (en) * | 2020-01-20 | 2021-07-29 | 北京芯海视界三维科技有限公司 | Checkerboard corner detection method and device |
| CN111553927A (en) * | 2020-04-24 | 2020-08-18 | 厦门云感科技有限公司 | Checkerboard corner detection method, checkerboard corner detection system, computer device and storage medium |
| CN111553927B (en) * | 2020-04-24 | 2023-05-16 | 厦门云感科技有限公司 | Checkerboard corner detection method, detection system, computer device and storage medium |
| CN113744177B (en) * | 2020-05-28 | 2024-08-23 | 中科寒武纪科技股份有限公司 | Corner detection method, device and storage medium of image |
| CN113744177A (en) * | 2020-05-28 | 2021-12-03 | 中科寒武纪科技股份有限公司 | Method and device for detecting angular point of image and storage medium |
| CN111681284A (en) * | 2020-06-09 | 2020-09-18 | 商汤集团有限公司 | A corner detection method, device, electronic device and storage medium |
| CN112017218A (en) * | 2020-09-09 | 2020-12-01 | 杭州海康威视数字技术股份有限公司 | Image registration method and device, electronic equipment and storage medium |
| CN114445499A (en) * | 2020-10-19 | 2022-05-06 | 深圳市光鉴科技有限公司 | Checkerboard angular point automatic extraction method, system, equipment and medium |
| CN113283416A (en) * | 2020-12-29 | 2021-08-20 | 深圳怡化电脑股份有限公司 | Character outline recognition method and device, electronic equipment and machine readable medium |
| CN113989384A (en) * | 2021-10-27 | 2022-01-28 | 杭州爱科科技股份有限公司 | Camera calibration method, cropping data generation method, device, device and medium |
| WO2024016686A1 (en) * | 2022-07-18 | 2024-01-25 | 宁德时代新能源科技股份有限公司 | Corner detection method and apparatus |
| CN116030450B (en) * | 2023-03-23 | 2023-12-19 | 摩尔线程智能科技(北京)有限责任公司 | Checkerboard corner point identification method, device, equipment and medium |
| CN116030450A (en) * | 2023-03-23 | 2023-04-28 | 摩尔线程智能科技(北京)有限责任公司 | Checkerboard corner recognition method, device, equipment and medium |
| CN118212210A (en) * | 2024-04-02 | 2024-06-18 | 常州信息职业技术学院 | Point symmetry-based automobile dipped beam type feature point detection method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109509200B (en) | 2023-09-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109509200B (en) | Checkerboard corner detection method based on contour extraction and computer readable storage medium | |
| CN101425142B (en) | Method and device for determining page angle of slope | |
| CN111275730A (en) | Method, device and equipment for determining map area and storage medium | |
| CN107577979B (en) | Method and device for quickly identifying DataMatrix type two-dimensional code and electronic equipment | |
| CN108334879B (en) | Region extraction method, system and terminal equipment | |
| CN110751620B (en) | Method for estimating volume and weight, electronic device, and computer-readable storage medium | |
| CN109300104B (en) | Angular point detection method and device | |
| CN107862235B (en) | Two-dimensional code position positioning method and device and terminal equipment | |
| CN112634301A (en) | Equipment area image extraction method and device | |
| CN107356213B (en) | Optical filter concentricity measuring method and terminal equipment | |
| CN112560847B (en) | Image text area positioning method and device, storage medium and electronic device | |
| CN115457063A (en) | Method, device and equipment for extracting edge of circular hole of PCB (printed Circuit Board) and storage medium | |
| CN111325164B (en) | Pointer representation number identification method and device and electronic equipment | |
| CN109784328B (en) | Method for positioning bar code, terminal and computer readable storage medium | |
| CN110969046A (en) | Face recognition method, device and computer-readable storage medium | |
| CN108960012B (en) | Feature point detection method and device and electronic equipment | |
| CN113744200B (en) | Camera dirt detection method, device and equipment | |
| CN112364807A (en) | Image recognition method and device, terminal equipment and computer readable storage medium | |
| CN113780278A (en) | Method and device for identifying license plate content, electronic equipment and storage medium | |
| CN110298835B (en) | A leather surface damage detection method, system and related device | |
| CN111340040A (en) | Paper character recognition method and device, electronic equipment and storage medium | |
| CN110751156A (en) | Method, system, device and medium for table line bulk interference removal | |
| CN108665439B (en) | Substrate detection method and terminal equipment | |
| CN111178366A (en) | Mobile robot positioning method and mobile robot | |
| CN113554695A (en) | Intelligent part hole site identification and positioning method |
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 |