CN113837206A - Image corner detection method based on machine learning SVM - Google Patents
Image corner detection method based on machine learning SVM Download PDFInfo
- Publication number
- CN113837206A CN113837206A CN202111148810.8A CN202111148810A CN113837206A CN 113837206 A CN113837206 A CN 113837206A CN 202111148810 A CN202111148810 A CN 202111148810A CN 113837206 A CN113837206 A CN 113837206A
- Authority
- CN
- China
- Prior art keywords
- corner
- image
- points
- point
- svm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an image corner detection method based on a machine learning SVM, which is characterized in that extracted corner candidate points are subjected to relevant judgment, and are classified by means of a support vector machine, so that reasonable corners are selected finally.
Description
Technical Field
The invention relates to the technical field of automatic visual detection, in particular to an image corner point detection method based on a machine learning SVM.
Background
The support vector machine method in machine vision is used for corner detection by means of an OpenCV simulation tool. For corner detection, it represents key points in the classification. Corner points play an important role in image processing. It is a method used in computer vision systems to extract certain features and understand the content of images. Therefore, the detection of the corner points in the image processing process is very important, and an efficient method for detecting the corner points is urgently needed.
For technical reasons, most researchers in the prior art still adopt Harris corner detection and Shi-Thomasi corner detection methods, and the detection methods have the following problems: the number of detected corner points is small, and robustness is lacked, so that feature extraction and identification of subsequent images are influenced.
Aiming at the defects, the invention provides the image corner detection method based on the machine learning SVM, and the method can increase the number of detected corners and improve the detection robustness, thereby improving the capability of extracting and identifying the features of the image so as to acquire more useful image information.
Disclosure of Invention
The invention aims to provide an image corner detection method based on a machine learning SVM (support vector machine) so as to solve the technical problems in the background technology. In order to achieve the purpose, the invention provides the following technical scheme: an image corner detection method based on a machine learning SVM comprises the following steps:
s1, preprocessing the collected corner image;
s2, setting a threshold t, acquiring pixel points and gray level change intensity of an angular point image, and predicting the angular point position;
and S3, measuring the strength of the predicted corner points, classifying the corner points by using an SVM (support vector machine), and displaying the finally determined corner points on the corner point image.
Preferably, the step S1 specifically includes: and converting the collected corner point image into a gray image.
Preferably, the step S3 specifically includes:
s31, selecting a pixel point p of a predicted corner position, simultaneously obtaining the positions of a plurality of pixel points k adjacent to the point p, and storing the position coordinates of the plurality of pixel points k adjacent to the point p in a vector;
s32, performing corner detection operation, and marking a pixel point p as 1 when k is less than p-t; when k is greater than p + t, marking the pixel point p as 2; wherein k-p is [ -255,255], and a lookup table with the size of 512 is constructed;
and S33, performing corner filtering operation, measuring the strength of the corner by using a response function, classifying the corner by using an SVM (support vector machine), and displaying the finally determined corner on the corner image.
Preferably, the response function is the sum of the absolute differences of adjacent arcs and intermediate pixel points.
Compared with the prior art, the invention has the beneficial effects that:
the invention discloses an image corner detection method based on a machine learning SVM, which is characterized in that extracted corner candidate points are subjected to relevant judgment, and are classified by means of a support vector machine, so that reasonable corners are selected finally.
Drawings
Fig. 1 is a schematic overall flow chart of an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention discloses an image corner detection method based on a machine learning SVM, which is characterized in that extracted corner candidate points are subjected to relevant judgment, and are classified by means of a support vector machine, so that reasonable corners are selected finally, specifically, a threshold value is set, pixel points of an image and the intensity of gray level change of the pixel points are obtained, then the candidate points are compared with surrounding pixel points by using the threshold value, the threshold value is automatically estimated according to an image global threshold value, finally, the obtained corners are subjected to filtering operation and classification by the support vector machine, and the obtained corners can be displayed finally in the image. Referring to fig. 1, the present invention provides a technical solution: an image corner detection method based on a machine learning SVM comprises the following steps:
s1, preprocessing the collected corner image;
s2, setting a threshold value t, acquiring pixel points of the corner image and gray level change intensity, and predicting the corner position, wherein the threshold value is an image threshold value, is selected from 0 to t, is in the range of 0 to 255, and the gray level change intensity refers to a variable of a gray level value;
and S3, measuring the strength of the predicted corner points, classifying the corner points by using an SVM (support vector machine), and displaying the finally determined corner points on the corner point image.
As a preferred embodiment of the present invention, the step S1 specifically includes: and converting the collected corner point image into a gray image.
As a preferred embodiment of the present invention, the step S3 specifically includes:
s31, selecting a pixel point p of a predicted corner position, simultaneously obtaining the positions of a plurality of pixel points k adjacent to the point p, and storing the position coordinates of the plurality of pixel points k adjacent to the point p in a vector;
s32, performing corner detection operation, and marking a pixel point p as 1 when k is less than p-t; when k is greater than p + t, marking the pixel point p as 2; wherein k-p is [ -255,255], and a lookup table with the size of 512 is constructed;
specifically, the lookup table refers to a table with rows 1 to 512, and is used for recording the position of the pixel point p.
And S33, performing corner filtering operation, measuring the strength of the corner by using a response function, classifying the corner by using an SVM (support vector machine), checking the condition of the corner after SVM classification, and displaying the finally determined corner on a corner image.
As a preferred embodiment of the present invention, the response function is a sum of absolute differences of adjacent arcs and intermediate pixel points.
Claims (4)
1. An image corner detection method based on a machine learning SVM is characterized by comprising the following steps:
s1, preprocessing the collected corner image;
s2, setting a threshold t, acquiring pixel points and gray level change intensity of an angular point image, and predicting the angular point position;
and S3, measuring the strength of the predicted corner points, classifying the corner points by using an SVM (support vector machine), and displaying the finally determined corner points on the corner point image.
2. The method for detecting corner points of images based on machine learning SVM as claimed in claim 1, wherein said step S1 specifically includes: and converting the collected corner point image into a gray image.
3. The method for detecting corner points of images based on machine learning SVM as claimed in claim 1, wherein said step S3 specifically includes:
s31, selecting a pixel point p of a predicted corner position, simultaneously obtaining the positions of a plurality of pixel points k adjacent to the point p, and storing the position coordinates of the plurality of pixel points k adjacent to the point p in a vector;
s32, performing corner detection operation, and marking a pixel point p as 1 when k is less than p-t; when k is greater than p + t, marking the pixel point p as 2; wherein k-p is [ -255,255], and a lookup table with the size of 512 is constructed;
and S33, performing corner filtering operation, measuring the strength of the corner by using a response function, classifying the corner by using an SVM (support vector machine), and displaying the finally determined corner on the corner image.
4. The method of claim 3, wherein the response function is the sum of absolute differences between adjacent arcs and intermediate pixels.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111148810.8A CN113837206A (en) | 2021-09-29 | 2021-09-29 | Image corner detection method based on machine learning SVM |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111148810.8A CN113837206A (en) | 2021-09-29 | 2021-09-29 | Image corner detection method based on machine learning SVM |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113837206A true CN113837206A (en) | 2021-12-24 |
Family
ID=78967519
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111148810.8A Pending CN113837206A (en) | 2021-09-29 | 2021-09-29 | Image corner detection method based on machine learning SVM |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113837206A (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107240112A (en) * | 2017-06-28 | 2017-10-10 | 北京航空航天大学 | Individual X Angular Point Extracting Methods under a kind of complex scene |
CN109146861A (en) * | 2018-08-04 | 2019-01-04 | 福州大学 | A kind of improved ORB feature matching method |
CN111640157A (en) * | 2020-05-28 | 2020-09-08 | 华中科技大学 | Checkerboard corner detection method based on neural network and application thereof |
-
2021
- 2021-09-29 CN CN202111148810.8A patent/CN113837206A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107240112A (en) * | 2017-06-28 | 2017-10-10 | 北京航空航天大学 | Individual X Angular Point Extracting Methods under a kind of complex scene |
CN109146861A (en) * | 2018-08-04 | 2019-01-04 | 福州大学 | A kind of improved ORB feature matching method |
CN111640157A (en) * | 2020-05-28 | 2020-09-08 | 华中科技大学 | Checkerboard corner detection method based on neural network and application thereof |
Non-Patent Citations (1)
Title |
---|
马大哈先生: "图像特征提取(四)——FAST算法解析", pages 2 - 3, Retrieved from the Internet <URL:《https://blog.csdn.net/qq_37764129/article/details/80970032》> * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110210409B (en) | Method and system for detecting form frame lines in form document | |
CN107705288B (en) | Infrared video detection method for dangerous gas leakage under strong interference of pseudo-target motion | |
CN107977639B (en) | Face definition judgment method | |
CN109426814B (en) | Method, system and equipment for positioning and identifying specific plate of invoice picture | |
CN111340027A (en) | Steel pile identification method and system, electronic equipment and medium | |
CN103852034B (en) | A kind of method for measuring perendicular | |
WO2021000948A1 (en) | Counterweight weight detection method and system, and acquisition method and system, and crane | |
CN110751619A (en) | Insulator defect detection method | |
CN105678737A (en) | Digital image corner point detection method based on Radon transform | |
CN113780484B (en) | Industrial product defect detection method and device | |
CN114882010A (en) | Surface defect detection method based on picture recognition | |
CN109086643B (en) | Color box label detection method and system based on machine vision | |
CN111626104B (en) | Cable hidden trouble point detection method and device based on unmanned aerial vehicle infrared thermal image | |
CN112419225B (en) | SOP type chip detection method and system based on pin segmentation | |
CN116091503B (en) | Method, device, equipment and medium for discriminating panel foreign matter defects | |
CN113837206A (en) | Image corner detection method based on machine learning SVM | |
CN111935480B (en) | Detection method for image acquisition device and related device | |
CN114663365A (en) | Defect detection method, device and computer storage medium | |
CN114627463A (en) | Non-contact power distribution data identification method based on machine identification | |
CN112801963A (en) | Video image occlusion detection method and system | |
CN112927197B (en) | Method, device, equipment and storage medium for detecting rust of air conditioner outdoor unit bracket | |
CN108805855A (en) | A kind of concrete cracks method for quickly identifying | |
CN113344858B (en) | Feature detection method, device and computer storage medium | |
CN114119754B (en) | Pixel-level fire smoke root node detection method based on connected particles | |
CN117952963A (en) | Photovoltaic connector fault detection method based on contour matching |
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 |