Unlock instant, AI-driven research and patent intelligence for your innovation.

Anti-interference remote sensing image target detection lightweight model and method thereof

A technology for target detection and remote sensing images, applied in neural learning methods, biological neural network models, character and pattern recognition, etc., can solve problems such as increasing the burden of detection models, achieve improved algorithm performance, low hardware requirements, and prevent false detection Effect

Active Publication Date: 2022-02-01
BEIHANG UNIV
View PDF6 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In addition, due to the influence of distance, weather, light, temperature and other factors, there are often various disturbances in the environment to be detected. These disturbances directly increase the burden on the detection model. Identification is a hot spot and difficult problem in current algorithm research

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
  • Anti-interference remote sensing image target detection lightweight model and method thereof
  • Anti-interference remote sensing image target detection lightweight model and method thereof
  • Anti-interference remote sensing image target detection lightweight model and method thereof

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0043] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0044] The embodiment of the present invention discloses an anti-interference remote sensing image target detection lightweight model and a method thereof:

[0045] 1. Lightweight multi-scale feature extraction network

[0046] In order to better embed the deep network model into the hardware environment, it is very important to greatly improve the inference speed of the network model while ensuring the detection and recognition perfo...

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 discloses an anti-interference remote sensing image target detection lightweight model and a method thereof. The model comprises a lightweight multi-scale feature extraction network, a feature pixel correction module and a multi-direction detection module which are connected in sequence. The method comprises the following steps: performing feature extraction on an input picture; respectively acquiring weights of pixel values at different positions according to the feature map, correcting the pixel values in the feature map through learned weight values, and acquiring a corrected feature map; and predicting the position and category of the target in the input picture according to the corrected feature map. The method is high in detection speed, high in precision, small in model parameter quantity, small in calculation amount, high in compatibility and capable of effectively resisting the influence of optical interference on target detection and recognition and preventing false detection.

Description

technical field [0001] The invention relates to the technical field of digital image processing, in particular to an anti-interference remote sensing image target detection lightweight model and a method thereof. Background technique [0002] In recent years, deep learning technology has developed rapidly. Thanks to the support of a large amount of data, various deep learning technologies have surpassed traditional machine learning algorithms in many aspects. The development of deep convolutional neural networks has promoted the development of target recognition, target segmentation, target detection and other technologies. Most of the current target detection algorithms are based on deep convolutional neural network models. These algorithms can be roughly divided into anchor-based detection models and anchor-free detection models. The anchor frame-based detection algorithm first needs to manually set the anchor frame size and aspect ratio of various targets, and then learn...

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 Applications(China)
IPC IPC(8): G06V20/10G06V10/80G06V10/82G06K9/62G06N3/04G06N3/08
CPCG06N3/084G06N3/047G06N3/045G06F18/253
Inventor 李波晏焕钱张鸿韦星星
Owner BEIHANG UNIV