Method and system for automatically extracting image adaptation region of unmanned aerial vehicle based on sparse representation
A sparse representation and automatic extraction technology, applied in computer components, instruments, computing, etc., can solve the problems of automation degree and universal suppression, affecting the universality of automatic extraction methods, etc.
Active Publication Date: 2019-01-01
WUHAN UNIV
View PDF5 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] Screening strategies based on single or several specific indicators affect the universality of automatic extraction methods, while supervised clustering methods are always subject to human factors, and the degree of automation and universality are also inhibited
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0066] The specific technical solutions of the present invention will be described below according to the drawings and embodiments.
[0067] The present invention proposes a method for automatically extracting the adaptation area of an unmanned aerial vehicle image based on sparse representation, which starts from the essential characteristics of the adaptation area, on the basis of matching characteristic analysis, and combines the synthesis of the adaptation and non-adaptation features in the image Differences, integrating the inherent connection of the matching area, weakening the dependence of the traditional manual extraction process on human subjective factors, and realizing the fully automatic extraction of the image matching area. Such as figure 1 As shown, the specific implementation method provided by the embodiment includes the following steps:
[0068] Step 1, data preprocessing, setting different SLIC superpixel numbers, and performing multi-scale superpixel se...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention provides a method and a system for automatically extracting an adaptive region of an unmanned aerial vehicle image based on a sparse representation. Firstly, a plurality of SLIC segmentation methods with different pixel number parameters are used for performing superpixel segmentation on an original image to generate a group of segmented images with different superpixel region sizes.Then, according to the segmentation results of each scale, the region with high specificity and high point density is recognized as the initial adaptation region sample, and the region with low specificity and small number of feature points is extracted as the non-adaptation region sample by combining the classic SIFT feature and Hausdorff distance matching. Then the reconstructed residuals of each superpixel region are computed by the sparse representation process as the fitness factor, and the fitness region detection results at a single scale are optimized by recursive iteration, and finally the final fitness region detection results are obtained by fusion of multi-scale detection images. The invention improves the problems of strong subjectivity, high labor cost, low working efficiency and the like, and provides technical support and reference for scene matching navigation and positioning.
Description
technical field [0001] The invention belongs to the field of computer vision and image processing, and relates to an automatic extraction technology of an unmanned aerial vehicle image adaptation area based on sparse representation. Background technique [0002] UAV (Unmanned Aerial Vehicle, UAV) has the advantages of small size, light weight, high flexibility, strong concealment, low cost, and no hidden dangers to the crew's personal safety. It is used in civil and military fields, such as disaster monitoring, geological exploration , map surveying, military reconnaissance, target attack, battlefield situation monitoring and other aspects have a very wide range of applications. Advanced navigation systems play an important role in the application of UAVs, especially for long-distance and long-running working environments that are difficult to reach by manual wire control or remote control mode. Perfect, high-precision autonomous navigation technology is the key to UAVs’ sur...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/46G06K9/62
CPCG06V10/462G06F18/25
Inventor 张永军王祥谢勋伟李彦胜
Owner WUHAN UNIV