Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Infrared target detection method

A technology of infrared target and detection method, which is applied in the field of multi-scale infrared target detection, can solve the problems of slow speed and low detection accuracy, achieve effective results and reduce computing time

Active Publication Date: 2020-02-21
XIDIAN UNIV
View PDF4 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to overcome the above defects in the prior art, and propose a multi-scale infrared target detection algorithm based on iterative quantization-local sensitive hashing, which is used to solve the problem of accurate detection in the existing multi-scale infrared target detection algorithm. Low rate and slow technical issues

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
  • Infrared target detection method
  • Infrared target detection method
  • Infrared target detection method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0047] Refer to attached figure 1 , a multi-scale infrared target detection method based on iterative quantization-locality sensitive hashing, including the following steps:

[0048] Step 1. Compress the input infrared image, and use windows of different sizes to perform window sliding on the compressed image to obtain multiple candidate windows;

[0049] Step 2, generate a data matrix X according to the candidate window and the target template;

[0050] Specifically, when detecting and recognizing an infrared target, under the condition of ensuring correct detection and recognition of the target, the number of candidate windows can be reduced by compressing the image, thereby shortening the detection time. Since the objects to be detected have different sizes and shapes in the infrared image, windows of different sizes should be used to slide the infrared image to obtain candidate windows.

[0051] Set windows of different sizes, and then slide the multiple windows sequenti...

Embodiment 2

[0066] On the basis of the first embodiment above, in order to reduce the quantization error that occurs during the conversion of the data matrix X into the binary coding matrix B and improve the target detection hit rate, this embodiment adopts an iterative quantization method to obtain the optimal rotation matrix R, and correspondingly calculate the optimal binary coding matrix B, the process is as follows:

[0067] On the basis of the binary coding matrix B in embodiment one, set transition matrix C=XWB ', carry out singular value decomposition to C'C and CC ', make C'C=U 1 ΣU' 1 , CC'=U 2 ΣU' 2 , let the rotation matrix R=U 1 u 2 , where B' is the transposition of the binary coding matrix B, and C' is the transposition of the transition matrix C; by transposing the binary coding matrix B and bringing it into the transition matrix C, a new rotation matrix R is obtained, that is, from the step The initial rotation matrix R in step 312 obtains a new rotation matrix after...

Embodiment 3

[0070] In the second embodiment, in order to speed up the speed of target detection, the input infrared image is compressed, causing the image to lose some information, so more accurate fine screening is required. This embodiment is further improved on the basis of embodiment two, see figure 2 Shown specifically as follows:

[0071] Calculate the Hamming distance between each candidate window and the target template according to the binary coding matrix B, and select the 10 candidate windows closest to each target template as the first round of candidate windows.

[0072] The Hamming distance refers to the number of different codes in each bit of the same position in two binary codes of the same length, which can measure the similarity of the two binary codes. The smaller the distance, the more similar the candidate window and the target template are. Possibly where the target is located. The binary coding matrix B contains the binary coding of the candidate window and the ...

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

The invention provides an infrared target detection method, which comprises the following steps: compressing an input infrared image, and roughly screening an image sliding window by using windows with different sizes; quantizing the continuous data into binary codes by using a local sensitive hash and iterative quantization method; outputting an optimal rotation matrix and a binary coding matrix;calculating a Hamming distance between the candidate window and the target template to obtain a first round of candidate window; mapping the first round of candidate window into the original infraredimage, and carrying out fine screening; wherein the window with the Hamming distance smaller than the threshold value is the position of the target. According to the technical scheme, the technical problems of low detection probability, low speed and poor complex scene adaptability in an existing infrared target detection algorithm are solved.

Description

technical field [0001] The invention belongs to the field of infrared target detection and relates to an infrared target detection method, in particular to a multi-scale infrared target detection method based on iterative quantization-local sensitive hashing. Background technique [0002] Infrared target detection technology plays an important role in the search and tracking of targets in complex scenes. In a large field of view infrared scene, the infrared image has a complex background and contains various interference factors, which are prone to false alarms and missed alarms. When the background contains multiple objects of different sizes and orientations, it is a challenging task to accurately detect all objects from a complex scene. [0003] In the existing target detection algorithm, the method based on the candidate region is used in R-CNN (convolutional neural network), which reduces the number of candidate windows, but it needs to merge regions through continuous...

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
IPC IPC(8): G06K9/32
CPCG06V10/245G06V10/255G06V2201/07
Inventor 吴鑫谢建张建奇黄曦刘德连
Owner XIDIAN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com