Small target detection method and device based on frequency domain residual in infrared image

A small target detection and infrared image technology, applied in the field of small target detection, can solve the problems of small calculation, fast speed, hardware acceleration, etc.

Inactive Publication Date: 2018-12-18
阿黛拉国际航空服务(江苏)有限公司
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In view of this, the embodiment of the present invention provides a small target detection method and device based on frequency domain residuals in infrared images. On the basis of image frequency domain saliency analysis, the frequency domain residual map is used to design a significant Calculate the possibility of targets contained in different areas by calculating the probability index, determine the area where the target is located according to the probability, and complete the detection of weak and small targets. The main calculation of this method is completed in the frequency domain. There are computationally complex issues in technology

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Small target detection method and device based on frequency domain residual in infrared image
  • Small target detection method and device based on frequency domain residual in infrared image
  • Small target detection method and device based on frequency domain residual in infrared image

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0032] Such as figure 1 As shown, in the embodiment of the present invention, aiming at the characteristics of small targets in infrared images, a small target detection method based on frequency domain residuals in infrared images is provided, including:

[0033] Step S100: using the frequency domain residual method (spectral residual) to calculate the saliency map of the original infrared image containing the small target to be detected;

[0034] Please refer to figure 2 , further, in step S100, may include:

[0035] Step S101, reducing the original infrared image containing the small target to be detected to the original δ times to obtain the target infrared image;

[0036] specific, image 3 The original infrared image containing the small target to be detected is shown, and the box indicated by mark E indicates the area where the small target is located.

[0037]In this embodiment, reducing the original infrared image can improve the speed of subsequent processing an...

Embodiment 2

[0103] An embodiment of the present invention provides a small target detection device based on frequency domain residuals in infrared images, such as Figure 13 As shown, it includes: a saliency map acquisition module 210, which is used to calculate the saliency map of the original infrared image containing the small target to be detected by using the frequency domain residual method; an integral map acquisition module 220, which is used to obtain the binary integral of the saliency map graph and composite integral graph; sliding window size acquisition module 230, for determining the size range of the sliding window in the saliency graph according to the size of the original infrared image; saliency calculation module 240, for according to the binary integral graph and the synthetic integral map, using the sliding window method to calculate the significance of each window of each size of the sliding window within the size range of the sliding window, corresponding to each siz...

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

No PUM Login to view more

Abstract

Embodiments of the present invention provide a small target detection method and device based on frequency domain residuals, which are applied in the field of small target detection. The small target detection method based on the frequency domain residual includes: using the frequency domain residual method to calculate the saliency map of the original infrared image containing the small target to be detected; obtaining the binary integral map and the synthetic integral map of the saliency map; according to the original The size of the infrared image determines the size range of the sliding window in the saliency map; calculates the saliency of each window of the sliding window of each size within the size range of the sliding window, and obtains position information and saliency information containing each window The matrix Boxa*b of matrix Boxa*b; Utilize the non-maximum value suppression algorithm for each said matrix Boxa*b, obtain the matrix new_Boxa*b that represents the window with local maximum saliency, merge all said matrixes new_Boxa*b, obtain matrix ROI: determine the area where the small target is located according to the saliency information in the matrix ROI. This method can effectively detect the area where the small target is located in the infrared image.

Description

technical field [0001] The invention relates to the field of small target detection, in particular to a small target detection method and device based on frequency domain residuals in infrared images. Background technique [0002] Object detection in images refers to the process of separating the object from the image background according to the difference between the object and the background. Depending on the type of image sensor, images commonly used for target detection include visible light images, infrared images, synthetic aperture radar images, and millimeter wave images. The present invention is aimed at infrared images. Compared with other types of imaging methods, infrared images have the following advantages: 1) Infrared images have strong environmental adaptability, especially at night and in bad weather; 2) Infrared radiation transmits The ability of fog and haze is stronger than that of visible light; 3) Infrared images have a stronger ability to identify cam...

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
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/32G06K9/46
CPCG06V10/25G06V10/462
Inventor 丁萌孙泽军魏丽张天慈陈飞王洁李明慧
Owner 阿黛拉国际航空服务(江苏)有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap