CN109712199B - Simple camera calibration method and device based on A4 paper point extraction - Google Patents
Simple camera calibration method and device based on A4 paper point extraction Download PDFInfo
- Publication number
- CN109712199B CN109712199B CN201811618007.4A CN201811618007A CN109712199B CN 109712199 B CN109712199 B CN 109712199B CN 201811618007 A CN201811618007 A CN 201811618007A CN 109712199 B CN109712199 B CN 109712199B
- Authority
- CN
- China
- Prior art keywords
- image
- paper
- point
- camera
- vanishing
- 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
Abstract
The invention discloses a simple camera calibration method and a simple camera calibration device based on A4 paper extraction binary vanishing points, wherein the method comprises the following steps: acquiring an image which is acquired by a camera and contains A4 paper, and preprocessing the image to obtain a filtered and de-noised gray scale image; performing edge extraction on the gray level image to obtain an edge image; scanning the edge image, searching a 0-to-1 sudden change or a 1-to-0 sudden change as a boundary starting point, searching from the found starting point, and taking the longest boundary which can be searched as an A4 paper outline point set; dividing A4 paper outline point sets into 4 types by adopting a spectrum multi-manifold clustering method, and respectively fitting straight lines on the 4 point sets by adopting a least square method to obtain 4 straight lines; 6 intersection points of the obtained 4 straight lines are obtained, and 2 intersection points of which the coordinates are not in the image range are regarded as vanishing points; and calculating the intrinsic parameters of the camera based on the camera calibration method of the second vanishing point. The method is simple and easy to implement, has better noise resistance, and can be applied to popular non-measuring cameras.
Description
Technical Field
The invention relates to camera calibration, in particular to a simple camera calibration method and device based on A4 paper point extraction, and belongs to the field of computer vision.
Background
Popular non-metrology camera applications typically require calibration of camera parameters. The camera calibration has gone through the development process from the traditional calibration method requiring precise markers to the active visual calibration method controlling the camera to make specific motions, and then to the self-calibration method based on the self-characteristics of the image. In the traditional calibration method, a checkerboard-based method is widely applied, but the calibration result is influenced by the detection precision of the checkerboard angular points, so that the defects of inconvenience in carrying, easiness in interference, complex template manufacturing and the like exist; the calibration method based on active vision does not need to place a calibration object, but requires to control a camera to do precise special motion, and is not suitable for occasions where the camera cannot be precisely controlled; the camera self-calibration method does not need to set specific control conditions, and directly solves the camera parameters only according to the relation between corresponding points of a plurality of images, and mainly comprises camera self-calibration based on vanishing points, camera self-calibration based on a plane or a three-dimensional template and self-calibration based on a natural scene.
Most camera calibration methods can only be implemented by professional people, and the common public has the problem of easy use. Therefore, the camera simple calibration method which can be used for local materials, is user-friendly and has strong noise resistance undoubtedly has important significance for popular non-measuring camera application.
The self-calibration method of the camera is concerned by the characteristics of simplicity, flexibility, strong applicability and the like, wherein the self-calibration method based on vanishing point calibration is widely regarded and deeply researched due to less limiting conditions. The common vanishing point-based self-calibration method needs to obtain vanishing points in three directions in a scene, which is often difficult to meet the requirements in practical application; the camera calibration method is carried out by calculating vanishing points in two orthogonal directions in a plurality of images, so that constraint conditions are reduced, and the calculation result is sensitive to the accuracy of vanishing point extraction. Therefore, the ease of acquisition of markers and the accuracy of vanishing point extraction are key issues that need to be addressed heavily in vanishing point-based methods.
The A4 paper is easy to obtain and has two pairs of parallel edges, so that the method is suitable for being applied to a vanishing point-based self-calibration method, but is easily influenced by illumination and background noise when edge parallel lines are extracted according to an A4 paper image and then vanishing points are extracted in a complex environment. Therefore, the research on the simple camera calibration method based on the A4 paper with strong noise resistance undoubtedly has important theoretical significance and application value for popular non-measurement camera application.
Disclosure of Invention
The purpose of the invention is as follows: aiming at the defects of the prior art, the invention aims to provide a simple camera calibration method and device based on A4 paper extraction binary vanishing point, which can be applied to popular non-measuring cameras and has the advantages of simplicity, easiness in implementation, strong noise resistance and the like.
The technical scheme is as follows: in order to achieve the purpose, the invention provides a simple camera calibration method based on A4 paper extraction binary vanishing point, which comprises the following steps:
(1) acquiring an image which is acquired by a camera and contains A4 paper, and preprocessing the image to obtain a filtered and de-noised gray scale image;
(2) performing edge extraction on the gray level image to obtain an edge image;
(3) scanning the edge image, searching a 0-to-1 sudden change or a 1-to-0 sudden change as a boundary starting point, searching from the found starting point, and taking the longest boundary which can be searched as an A4 paper outline point set;
(4) dividing A4 paper outline point sets into 4 types by adopting a spectrum multi-manifold clustering method, and respectively fitting straight lines on the 4 point sets by adopting a least square method to obtain 4 straight lines;
(5) 6 intersection points of the obtained 4 straight lines are obtained, and 2 intersection points of which the coordinates are not in the image range are regarded as vanishing points;
(6) and (5) repeating the steps (1) to (5) to extract vanishing points in a plurality of paper images with different shooting angles A4, and calculating intrinsic parameters of the camera based on the vanishing points.
In a preferred embodiment, the preprocessing of the image in step (1) includes converting the acquired RGB image into a gray-scale map, and performing a median filtering process on the gray-scale map.
In a preferred embodiment, the edge extraction of the gray-scale map by using a Canny operator in the step (2) specifically includes:
(2.1) carrying out Gaussian blur processing on the image;
(2.2) calculating the gradient strength and the direction of each pixel point in the image;
(2.3) comparing the gradient strength of each pixel point in the image with the gradient of points in the positive and negative gradient directions, and if the gradient strength of the current point is maximum, reserving the current point as an edge point;
and (2.4) setting a high threshold and a low threshold, if the gradient amplitude of the pixel exceeds the high threshold, reserving as an edge point, if the gradient amplitude of the pixel is lower than the low threshold, excluding, if the gradient amplitude of the pixel is between the high threshold and the low threshold, checking 8 pixel points around the pixel, and if the gradient amplitude of the pixel is higher than the high threshold, reserving as the edge point.
In a preferred embodiment, the method for extracting the outline of the a4 paper in the step (3) specifically comprises the following steps:
(3.1) scanning the binary image, and searching a pixel with a pixel value suddenly changing from 0 to 1 as a starting point of an outer boundary, wherein the pixel with the pixel value suddenly changing from 1 to 0 is used as a starting point of a hole boundary;
(3.2) continuing to search the boundary from the found starting point, marking the pixels on the boundary, assigning a unique ID to the newly found boundary in the process, namely NBD, setting the NBD to be 1 initially, adding 1 to each new boundary found, and extracting the contour point set of A4 paper based on the principle that the contour of A4 paper is the longest.
In a preferred embodiment, the step (5) calculates the camera reference matrix by extracting vanishing points of at least 5 a4 paper images, specifically: note that the extracted vanishing point is expressed as:
establishing a linear equation set according to vanishing point theory
Solving the system of equations to obtain a matrix
Then by the square root method
Is decomposed into
Then, the result of the decomposition is inverted to obtain
Finally obtaining the internal parameter matrix of the camera
Where the superscript T denotes transpose and the subscript 33 denotes row 3, column 3 in the matrix.
On the other hand, the simple camera calibration device for extracting the second vanishing point based on the A4 paper comprises the following components:
the preprocessing unit is used for acquiring an image which is acquired by a camera and contains A4 paper, and preprocessing the image to obtain a filtered and denoised gray image;
the edge extraction unit is used for extracting edges of the gray level image to obtain an edge image;
the contour extraction unit is used for scanning the edge image, searching a 0-to-1 sudden change or a 1-to-0 sudden change as a boundary starting point, searching from the found starting point, and taking the longest boundary which can be searched as an A4 paper contour point set;
the line fitting unit is used for dividing the A4 paper outline point set into 4 types by adopting a spectrum multi-manifold clustering method, and respectively fitting the 4 point sets with lines by adopting a least square method to obtain 4 lines;
a vanishing point extraction unit, which is used for solving 6 intersection points of the obtained 4 straight lines and taking 2 intersection points of which the coordinates are not in the image range as vanishing points;
and the camera calibration unit is used for calculating the intrinsic parameters of the camera based on the vanishing points in the extracted multiple paper images with different shooting angles A4.
Based on the same conception, the invention also discloses a computing device which comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the computer program realizes the camera simple calibration method based on the A4 paper extraction binary vanishing point when being loaded to the processor
Has the advantages that: according to the method for extracting the vanishing points based on the A4 paper and simply calibrating the camera, provided by the invention, the vanishing points in two directions of an image plane are calculated by performing edge detection, contour extraction and straight line fitting on an image containing the A4 paper by utilizing the characteristic that the edge of the A4 paper is mutually orthogonal parallel lines, and further the internal parameters of the camera are calculated by the camera calibration method based on the vanishing points. The method is simple and easy to implement, has good noise immunity, can effectively eliminate the influence of image background noise, accurately extracts parallel line information in the image and obtains an accurate vanishing point. The calculated camera internal parameters meet the precision requirement and can be used for popular non-measuring camera application.
Drawings
Fig. 1 is a flowchart of vanishing point extraction in the embodiment of the present invention.
Fig. 2 is a diagram illustrating the effect of edge detection in the embodiment of the present invention, in which (a) is an original image and (b) is a diagram illustrating the result of edge detection.
Fig. 3 is a graph showing the edge effect of a4 paper extracted in the embodiment of the present invention.
Fig. 4 is a diagram of the clustering effect of the SMMC algorithm in the embodiment of the present invention.
FIG. 5 is a diagram illustrating the effect of fitting a straight line according to an embodiment of the present invention.
Detailed Description
The invention is further described with reference to the following figures and specific examples.
As shown in fig. 1, the simple camera calibration method based on a4 paper extraction binary vanishing point disclosed in the embodiment of the present invention mainly includes the following steps:
step 1: image pre-processing
(a) And (3) converting into a gray-scale image: the RGB image containing A4 paper acquired by a camera is firstly converted into a gray-scale image, so that the subsequent image processing is facilitated.
(b) Median filtering treatment: 3 x 3 median filtering is carried out on the obtained gray level image, so that background noise is weakened, and the influence of noise on an extraction result is reduced.
Step 2: canny operator A4 paper edge extraction
(a) Gaussian blur processing: the image is convolved by a 5 x 5 Gaussian kernel, so as to achieve the purposes of smoothing the image, filtering noise and improving the edge detection performance.
(b) Gradient magnitude and direction are calculated: and calculating the gradient strength and the direction of each pixel point in the image.
(c) Non-maxima suppression: and for each pixel point in the image, comparing the gradient strength of the current point with the gradient of the point in the positive and negative gradient directions, and if the gradient strength of the current point is the maximum, keeping the current point as an edge point.
(d) Dual threshold detection and hysteresis threshold: setting the ratio as 2: 1 or 3: the high and low thresholds of 1 further extract edges. If the gradient amplitude of a certain pixel point exceeds a high threshold value, the pixel is reserved as an edge pixel; if the pixel is lower than the low threshold value, the pixel is suppressed and excluded; if the gradient amplitude of the pixel point in the 8-neighborhood range is higher than the high threshold value, the pixel point is reserved as an edge pixel, and if the gradient amplitude of the pixel point in the 8-neighborhood range is not higher than the high threshold value, the pixel point is inhibited from being eliminated. The final detection effect is shown in fig. 2.
And step 3: contour extraction
(a) Finding a boundary starting point: the input binary image has only two values of 0 and 1, the outer boundary represents a connected region with a pixel value of 1, and the hole boundary represents a connected region with a pixel value of 0. And scanning the binary image, and searching for a pixel with a pixel value suddenly changing from 0 to 1 as a starting point of an outer boundary, wherein the pixel with the pixel value suddenly changing from 1 to 0 is used as a starting point of a hole boundary.
(b) Searching for a boundary: the search continues from the start point found, marking the pixels on the boundary, assigning a unique ID to the newly found boundary, called NBD, which is initially set to 1 and incremented by 1 each time a new boundary is found. Based on the principle that the A4 paper has the longest outline, an outline point set of the A4 paper is extracted. The extracted profile is shown in fig. 3.
And 4, step 4: straight line fitting
(a) Spectral multi-manifold clustering: by adopting a spectrum multi-manifold clustering (SMMC) method, the mixed point set of 4 edges of the A4 paper is divided into four classes, and the classification effect is shown in figure 4.
(b) And (3) straight line fitting: after the contour point sets are clustered according to edges, each straight line corresponds to one point set, the straight lines are fitted to the point sets by adopting a least square method, and fig. 5 shows the fitting result.
And 5: extraction vanishing point
(a) And (3) linear intersection: and 6 intersection points can be obtained by calculating the intersection of the straight lines according to the fitted 4 straight lines.
(b) And (3) judging coordinates: 4 of the 6 intersection points are A4 paper corner points, and 2 are vanishing points, and since the image used for calibration generally does not have a large inclination angle, vanishing points are generally unlikely to appear in the image range, so that which two intersection points are vanishing points is determined according to whether the coordinates of the points are in the image range.
Step 6: camera calibration
At least 5A 4 paper images with different visual angles are shot by a camera, and the vanishing points of the images are respectively extracted and expressed as:
establishment of linear equation set based on vanishing point theory
Solving the system of equations to obtain a matrix
Then by the square root method
Is decomposed into
Then, the result of the decomposition is inverted to obtain
At this time
The difference of the parameter matrix K and the intra-camera parameter matrix K is a constant factor, and the last element of K is 1
Based on the same inventive concept, another embodiment of the invention discloses a camera simple calibration device based on a4 paper extraction binary vanishing point, which comprises: the preprocessing unit is used for acquiring an image which is acquired by a camera and contains A4 paper, and preprocessing the image to obtain a filtered and denoised gray image; the edge extraction unit is used for extracting edges of the gray level image to obtain an edge image; the contour extraction unit is used for scanning the edge image, searching a 0-to-1 sudden change or a 1-to-0 sudden change as a boundary starting point, searching from the found starting point, and taking the longest boundary which can be searched as an A4 paper contour point set; the straight line calculating unit is used for dividing the A4 paper outline point set into 4 types by adopting a spectrum multi-manifold clustering method, and respectively fitting straight lines on the 4 point sets by adopting a least square method to obtain 4 straight lines; a vanishing point extraction unit, which is used for solving 6 intersection points of the obtained 4 straight lines and taking 2 intersection points of which the coordinates are not in the image range as vanishing points; and the camera calibration unit is used for calculating the intrinsic parameters of the camera based on the vanishing points in the extracted multiple paper images with different shooting angles A4.
Based on the same inventive concept, the invention also discloses a computing device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the computer program realizes the camera simple calibration method based on the A4 paper extraction binary vanishing point when being loaded to the processor.
Claims (7)
1. A camera simple calibration method for extracting binary vanishing points based on A4 paper is characterized by comprising the following steps:
(1) acquiring an image which is acquired by a camera and contains A4 paper, and preprocessing the image to obtain a filtered and de-noised gray scale image;
(2) performing edge extraction on the gray level image to obtain an edge image;
(3) scanning the edge image, searching a 0-to-1 sudden change or a 1-to-0 sudden change as a boundary starting point, searching from the found starting point, and taking the longest boundary which can be searched as an A4 paper outline point set;
(4) dividing A4 paper outline point sets into 4 types by adopting a spectrum multi-manifold clustering method, and respectively fitting straight lines on the 4 point sets by adopting a least square method to obtain 4 straight lines;
(5) 6 intersection points of the obtained 4 straight lines are obtained, and 2 intersection points of which the coordinates are not in the image range are regarded as vanishing points;
(6) and (5) repeating the steps (1) to (5) to extract vanishing points in a plurality of paper images with different shooting angles A4, and calculating intrinsic parameters of the camera based on the vanishing points.
2. The camera easy calibration method for extracting binary points based on A4 paper as claimed in claim 1, wherein the preprocessing of the image in step (1) includes converting the acquired RGB image into gray scale map and performing median filtering process on the gray scale map.
3. The simple camera calibration method for extracting binary vanishing points based on A4 paper as claimed in claim 1, wherein the edge extraction of the gray scale map by Canny operator in the step (2) comprises:
(2.1) carrying out Gaussian blur processing on the image;
(2.2) calculating the gradient strength and the direction of each pixel point in the image;
(2.3) comparing the gradient strength of each pixel point in the image with the gradient of points in the positive and negative gradient directions, and if the gradient strength of the current point is maximum, reserving the current point as an edge point;
and (2.4) setting a high threshold and a low threshold, if the gradient amplitude of the pixel exceeds the high threshold, reserving as an edge point, if the gradient amplitude of the pixel is lower than the low threshold, excluding, if the gradient amplitude of the pixel is between the high threshold and the low threshold, checking 8 pixel points around the pixel, and if the gradient amplitude of the pixel is higher than the high threshold, reserving as the edge point.
4. The simple camera calibration method based on A4 paper extraction binary vanishing point of claim 1, wherein the method for extracting A4 paper contour in step (3) comprises:
(3.1) scanning the binary image, and searching a pixel with a pixel value suddenly changing from 0 to 1 as a starting point of an outer boundary, wherein the pixel with the pixel value suddenly changing from 1 to 0 is used as a starting point of a hole boundary;
(3.2) continuing to search the boundary from the found starting point, marking the pixels on the boundary, assigning a unique ID to the newly found boundary in the process, namely NBD, setting the NBD to be 1 initially, adding 1 to each new boundary found, and extracting the contour point set of A4 paper based on the principle that the contour of A4 paper is the longest.
5. The camera easy calibration method for extracting binary vanishing points based on A4 paper as claimed in claim 1, wherein the step (5) is performed by calculating an intra-camera parameter matrix by extracting vanishing points of at least 5A 4 paper images, specifically: note that the extracted vanishing point is expressed as:
establishing a linear equation set according to vanishing point theory
Solving the system of equations to obtain a matrix
Then by the square root method
Is decomposed into
Then, the result of the decomposition is inverted to obtain
Finally obtaining the internal parameter matrix of the camera
Where the superscript T denotes transpose and the subscript 33 denotes row 3, column 3 in the matrix.
6. A camera simple calibration device based on A4 paper extraction binary vanishing point is characterized by comprising:
the preprocessing unit is used for acquiring an image which is acquired by a camera and contains A4 paper, and preprocessing the image to obtain a filtered and denoised gray image;
the edge extraction unit is used for extracting edges of the gray level image to obtain an edge image;
the contour extraction unit is used for scanning the edge image, searching a 0-to-1 sudden change or a 1-to-0 sudden change as a boundary starting point, searching from the found starting point, and taking the longest boundary which can be searched as an A4 paper contour point set;
the straight line calculating unit is used for dividing the A4 paper outline point set into 4 types by adopting a spectrum multi-manifold clustering method, and respectively fitting straight lines on the 4 point sets by adopting a least square method to obtain 4 straight lines;
a vanishing point extraction unit, which is used for solving 6 intersection points of the obtained 4 straight lines and taking 2 intersection points of which the coordinates are not in the image range as vanishing points;
and the camera calibration unit is used for calculating the intrinsic parameters of the camera based on the vanishing points in the extracted multiple paper images with different shooting angles A4.
7. A computing device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the computer program, when loaded into the processor, implements the camera easy calibration method based on a4 paper extraction binary vanishing point according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811618007.4A CN109712199B (en) | 2018-12-28 | 2018-12-28 | Simple camera calibration method and device based on A4 paper point extraction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811618007.4A CN109712199B (en) | 2018-12-28 | 2018-12-28 | Simple camera calibration method and device based on A4 paper point extraction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109712199A CN109712199A (en) | 2019-05-03 |
| CN109712199B true CN109712199B (en) | 2020-09-08 |
Family
ID=66258895
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811618007.4A Active CN109712199B (en) | 2018-12-28 | 2018-12-28 | Simple camera calibration method and device based on A4 paper point extraction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109712199B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8368762B1 (en) * | 2010-04-12 | 2013-02-05 | Adobe Systems Incorporated | Methods and apparatus for camera calibration based on multiview image geometry |
| CN103729837A (en) * | 2013-06-25 | 2014-04-16 | 长沙理工大学 | Rapid calibration method of single road condition video camera |
| CN103735269A (en) * | 2013-11-14 | 2014-04-23 | 大连民族学院 | Height measurement method based on video multi-target tracking |
| CN104809755A (en) * | 2015-04-09 | 2015-07-29 | 福州大学 | Single-image-based cultural relic three-dimensional reconstruction method |
| CN207173127U (en) * | 2017-05-06 | 2018-04-03 | 北京七视野文化创意发展有限公司 | Perspective view drawing instrument |
| CN108961182A (en) * | 2018-06-25 | 2018-12-07 | 北京大学 | Vertical direction vanishing point detection method and video positive twist method for video image |
-
2018
- 2018-12-28 CN CN201811618007.4A patent/CN109712199B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8368762B1 (en) * | 2010-04-12 | 2013-02-05 | Adobe Systems Incorporated | Methods and apparatus for camera calibration based on multiview image geometry |
| CN103729837A (en) * | 2013-06-25 | 2014-04-16 | 长沙理工大学 | Rapid calibration method of single road condition video camera |
| CN103735269A (en) * | 2013-11-14 | 2014-04-23 | 大连民族学院 | Height measurement method based on video multi-target tracking |
| CN104809755A (en) * | 2015-04-09 | 2015-07-29 | 福州大学 | Single-image-based cultural relic three-dimensional reconstruction method |
| CN207173127U (en) * | 2017-05-06 | 2018-04-03 | 北京七视野文化创意发展有限公司 | Perspective view drawing instrument |
| CN108961182A (en) * | 2018-06-25 | 2018-12-07 | 北京大学 | Vertical direction vanishing point detection method and video positive twist method for video image |
Non-Patent Citations (3)
| Title |
|---|
| Improved Calibration Algorithm;Xu Renjie 等;《Spring》;20131231;第363卷;全文 * |
| 基于两灭点法的摄像机标定方法研究;胡桂廷 等;《电子测量技术》;20120731;第35卷(第7期);全文 * |
| 基于矩形的摄像机自标定几何方法;徐嵩 等;《光学学报》;20141130;第34卷(第11期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109712199A (en) | 2019-05-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110443836B (en) | Point cloud data automatic registration method and device based on plane features | |
| CN107301661B (en) | High-resolution remote sensing image registration method based on edge point features | |
| CN110232389B (en) | Stereoscopic vision navigation method based on invariance of green crop feature extraction | |
| CN106981077B (en) | Infrared image and visible light image registration method based on DCE and LSS | |
| CN107452030B (en) | Image registration method based on contour detection and feature matching | |
| CN110334762B (en) | Feature matching method based on quad tree combined with ORB and SIFT | |
| KR102073468B1 (en) | System and method for scoring color candidate poses against a color image in a vision system | |
| CN104966285B (en) | A kind of detection method of salient region | |
| TWI497450B (en) | Visual object tracking method | |
| CN104778701A (en) | Local image describing method based on RGB-D sensor | |
| CN109919960B (en) | Image continuous edge detection method based on multi-scale Gabor filter | |
| CN105787943B (en) | SAR image registration method based on multi-scale image block feature and rarefaction representation | |
| CN110472521B (en) | Pupil positioning calibration method and system | |
| CN108269274B (en) | Image registration method based on Fourier transform and Hough transform | |
| Wang et al. | Noise-robust color edge detector using gradient matrix and anisotropic Gaussian directional derivative matrix | |
| CN112634125B (en) | Automatic face replacement method based on off-line face database | |
| CN106940782B (en) | High-resolution SAR newly-added construction land extraction software based on variation function | |
| Srinivas et al. | Remote sensing image segmentation using OTSU algorithm | |
| Han et al. | Research and implementation of an improved canny edge detection algorithm | |
| CN103077528A (en) | Rapid image matching method based on DCCD (Digital Current Coupling)-Laplace and SIFT (Scale Invariant Feature Transform) descriptors | |
| CN106683125B (en) | RGB-D image registration method based on 2D/3D mode switching | |
| CN109767442B (en) | Remote sensing image airplane target detection method based on rotation invariant features | |
| CN109712199B (en) | Simple camera calibration method and device based on A4 paper point extraction | |
| CN112861870A (en) | Pointer instrument image correction method, system and storage medium | |
| CN110599407B (en) | Human body noise reduction method and system based on multiple TOF cameras in downward inclination angle direction |
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 |