Ground target detection method based on feature matching

Abstract

The invention discloses a ground target detection method based on feature matching, and the method comprises the following steps: 1, obtaining an image: obtaining the image or video data of a ground target region through a sensor; 2, image enhancement: preprocessing the image, such as de-noising, contrast enhancement and the like, so as to improve the feature extraction precision; according to the method, target images under different illumination and view angles can be processed based on a feature matching method, so that the robustness of target recognition is enhanced, interference factors can be overcome to a certain extent by extracting stable feature points and optimizing a matching algorithm, the success rate of target recognition is improved, and the target recognition efficiency is improved. The method based on feature matching can achieve automatic processing, and target detection and recognition tasks can be completed without manual intervention.

Description

Technical Field

[0001] The present invention relates to the technical field of ground target detection, and more specifically to a ground target detection method based on feature matching. Background Art

[0002] Ground object detection is a widely used technology in computer vision and image processing, which is mainly used to identify, locate and classify various target objects on the ground. These target objects can be vehicles, pedestrians, animals, buildings or any other entities of interest. Ground object detection aims to automatically identify and mark target objects on the ground by processing and analyzing image or video data.

[0003] Ground object detection In complex ground environments, objects may be difficult to identify due to factors such as changes in viewing angles, differences in lighting conditions, or occlusion. Therefore, a new technical solution is needed to solve this problem. Summary of the invention

[0004] The purpose of the present invention is to provide a ground target detection method based on feature matching, which solves the problem that in a complex ground environment, the target may be difficult to identify due to factors such as viewing angle changes, different lighting conditions or occlusion.

[0005] To achieve the above object, the present invention provides the following technical solution: a ground target detection method based on feature matching, the ground target detection method based on feature matching comprises the following steps:

[0006] Step 1: Image acquisition: obtaining image or video data of the ground target area through sensors;

[0007] Step 2: Image enhancement, preprocessing the image, such as denoising, contrast enhancement, etc., to improve the accuracy of feature extraction;

[0008] Step 3: Select a suitable feature detection algorithm based on the application scenario and image characteristics, and use the selected feature detection algorithm to extract feature points with unique properties from the image;

[0009] Step 4: Describe and encode the extracted feature points to generate feature descriptors;

[0010] Step 5: Select a suitable matching algorithm according to the type and number of feature descriptors, and use the selected matching algorithm to match the feature descriptors in the image to be detected with the feature descriptors in the reference image to find feature points with high similarity;

[0011] Step 6: Screen the matching results based on their quality and quantity to remove mismatched and unreliable matching point pairs;

[0012] Step 7: Based on the calculated matching feature point pairs, locate the position and range of the target in the image to be detected, draw a rectangular box, polygon, etc. on the image to mark the target area, and further analyze and identify the located target.

[0013] As a preferred implementation manner of the present invention, the feature extraction in step 3 includes shape, texture, and color.

[0014] As a preferred implementation of the present invention, the feature detection algorithm in step 3 includes a scale-invariant feature transformation algorithm, an accelerated robust feature algorithm and a directional gradient histogram algorithm.

[0015] As a preferred embodiment of the present invention, the directional gradient histogram algorithm is used in step 3 to calculate the horizontal direction G of each pixel in the image. x and vertical direction G y The gradient of , the gradient calculation formula is as follows;

[0016] Gx(x,y)=I(x+1,y)-I(x-1,y) Gy(x,y)=I(x,y+1)-I(x,y-1)

[0017] The gradient direction calculation formula is as follows:

[0018]

[0019] The gradient size calculation formula is as follows:

[0020]

[0021] G x (x,y) and G y (x, y) represents the gradient value in the horizontal and vertical directions of the point respectively;

[0022] I usually represents the pixel value of an image.

[0023] As a preferred implementation of the present invention, the matching algorithm in step 5 includes a brute force matching algorithm and a fast nearest neighbor search algorithm.

[0024] Compared with the prior art, the present invention has the following beneficial effects:

[0025] The present invention improves the accuracy of target detection by selecting stable feature points that can maintain good stability under different illumination, viewing angles and scales. The efficient matching algorithm can accurately find feature point pairs with high similarity, while removing false matches and unreliable matches, further improving the accuracy of target detection. The method based on feature matching can process target images under different illumination and viewing angles, thereby enhancing the robustness of target recognition. By extracting stable feature points and optimizing the matching algorithm, interference factors can be overcome to a certain extent and the success rate of target recognition can be improved. The method based on feature matching can realize automated processing and complete target detection and recognition tasks without human intervention. DETAILED DESCRIPTION

[0026] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] The present invention provides a technical solution: a ground target detection method based on feature matching, the ground target detection method based on feature matching comprises the following steps:

[0028] Step 1: Image acquisition: obtaining image or video data of the ground target area through sensors;

[0029] Step 2: Image enhancement, preprocessing the image, such as denoising, contrast enhancement, etc., to improve the accuracy of feature extraction;

[0030] Step 3: Select a suitable feature detection algorithm based on the application scenario and image characteristics, and use the selected feature detection algorithm to extract feature points with unique properties from the image;

[0031] Step 4: Describe and encode the extracted feature points to generate feature descriptors;

[0032] Step 5: Select a suitable matching algorithm according to the type and number of feature descriptors, and use the selected matching algorithm to match the feature descriptors in the image to be detected with the feature descriptors in the reference image to find feature points with high similarity;

[0033] Step 6: Screen the matching results based on their quality and quantity to remove mismatched and unreliable matching point pairs;

[0034] Step 7: Based on the calculated matching feature point pairs, locate the position and range of the target in the image to be detected, draw a rectangular box, polygon, etc. on the image to mark the target area, and further analyze and identify the located target.

[0035] Further improved, the feature extraction in step 3 includes shape, texture, and color. By considering shape, texture, and color features simultaneously, the target object can be described more comprehensively, and the accuracy and robustness of feature extraction can be improved, thereby enhancing the reliability of target detection.

[0036] Further improved, the feature detection algorithm in step 3 includes a scale-invariant feature transformation algorithm, an accelerated robust feature algorithm and a directional gradient histogram algorithm. By using multiple feature detection algorithms, the most suitable algorithm can be selected according to different application scenarios and image characteristics to improve the adaptability and accuracy of feature detection.

[0037] Further improved, in step 3, the directional gradient histogram algorithm is used to calculate the horizontal direction G for each pixel in the image. x and vertical direction G y The gradient of , the gradient calculation formula is as follows;

[0038] Gx(x,y)=I(x+1,y)-I(x-1,y) Gy(x,y)=I(x,y+1)-I(x,y-1)

[0039] The gradient direction calculation formula is as follows:

[0040]

[0041] The gradient size calculation formula is as follows:

[0042]

[0043] G x (x,y) and G y (x, y) represents the gradient value in the horizontal and vertical directions of the point respectively;

[0044] I usually represents the pixel value of the image;

[0045] The directional gradient histogram algorithm can capture the edge and texture information of the image by calculating the horizontal and vertical gradients of each pixel in the image. It has good robustness to changes in illumination and perspective, thereby improving the stability of feature extraction.

[0046] Further improved, the matching algorithm in step 5 includes a brute force matching algorithm and a fast nearest neighbor search algorithm, providing a variety of matching algorithm options. The appropriate matching algorithm can be selected according to the type and number of feature descriptors and the requirements of matching efficiency to improve the matching speed and accuracy.

[0047] Working principle: obtain image or video data of the ground target area through the sensor; pre-process the acquired image, including denoising, contrast enhancement, etc., to improve the accuracy of subsequent feature extraction; select a suitable feature detection algorithm according to the application scenario and image characteristics, and extract feature points with unique properties from the image; describe and encode the extracted feature points to generate feature descriptors for subsequent feature matching; select a suitable matching algorithm according to the type and number of feature descriptors, match the feature descriptors in the image to be detected with the feature descriptors in the reference image, and find feature points with high similarity; screen according to the quality and quantity of the matching results, remove mismatches and unreliable matching point pairs, and ensure the accuracy of the matching results; locate the position and range of the target in the image to be detected based on the calculated matching feature point pairs, and draw rectangular boxes, polygons, etc. on the image to mark the target area, and further analyze and identify the located target.

[0048] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention. Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description, and it is intended that all changes that fall within the meaning and scope of the equivalent elements of the claims are included in the present invention.

[0049] Finally, a few points should be explained: First, in the description of this application, it should be noted that, unless otherwise specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, and can be mechanical or electrical connections, or internal connectivity between two components, or direct connections. "Up", "down", "left", "right", etc. are only used to indicate relative position relationships. When the absolute position of the object being described changes, the relative position relationship may change.

[0050] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A ground target detection method based on feature matching, characterized in that: The ground target detection method based on feature matching includes the following steps: Step 1: Image acquisition: obtaining image or video data of the ground target area through sensors; Step 2: Image enhancement, preprocessing the image, such as denoising, contrast enhancement, etc., to improve the accuracy of feature extraction; Step 3: Select a suitable feature detection algorithm according to the application scenario and image characteristics, and use the selected feature detection algorithm to extract feature points with unique properties from the image; Step 4: Describe and encode the extracted feature points to generate feature descriptors; Step 5: Select a suitable matching algorithm according to the type and number of feature descriptors, and use the selected matching algorithm to match the feature descriptors in the image to be detected with the feature descriptors in the reference image to find feature points with high similarity; Step 6: Screen the matching results based on their quality and quantity, and remove mismatched and unreliable matching point pairs; Step 7: Based on the calculated matching feature point pairs, locate the position and range of the target in the image to be detected, draw a rectangular box, polygon, etc. on the image to mark the target area, and further analyze and identify the located target.

2. The method for detecting ground targets based on feature matching according to claim 1, characterized in that: The feature extraction in step 3 includes shape, texture, and color.

3. The method for detecting ground targets based on feature matching according to claim 1, characterized in that: The feature detection algorithm in step 3 includes a scale-invariant feature transformation algorithm, an accelerated robust feature algorithm, and a directional gradient histogram algorithm.

4. The method for detecting ground targets based on feature matching according to claim 1, characterized in that: In step 3, the directional gradient histogram algorithm is used to calculate the horizontal direction G for each pixel in the image. x and vertical direction G y The gradient of , the gradient calculation formula is as follows; Gx(x,y)=I(x+1,y)-I(x-1,y)Gy(x,y)=I(x,y+1)-I(x,y-1)The gradient direction calculation formula is as follows: The gradient size calculation formula is as follows: G x (x,y) and G y (x, y) represents the gradient value in the horizontal and vertical directions of the point respectively; I usually represents the pixel value of an image.

5. The method for detecting ground targets based on feature matching according to claim 1, characterized in that: The matching algorithm in step 5 includes a brute force matching algorithm and a fast nearest neighbor search algorithm.

Images