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

Multi-feature cyclic convolution saliency target detection method based on attention mechanism

A cyclic convolution and target detection technology, applied in image data processing, image enhancement, instruments, etc., can solve problems such as insufficient feature utilization, blurred target contour, background error detection, etc., to enhance spatial resolution and enhance feature representation. effect of ability

Inactive Publication Date: 2020-01-03
中国人民解放军火箭军工程大学
View PDF11 Cites 64 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved by the present invention is: the existing salient target detection method has problems such as insufficient feature utilization benefit, false background detection and blurred target outline. Therefore, a multi-feature circular convolution based on attention mechanism is provided. Sexual Target Detection Methods

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
  • Multi-feature cyclic convolution saliency target detection method based on attention mechanism
  • Multi-feature cyclic convolution saliency target detection method based on attention mechanism
  • Multi-feature cyclic convolution saliency target detection method based on attention mechanism

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] The present invention will be further described below in conjunction with the accompanying drawings.

[0036] A multi-feature circular convolution salient target detection method based on the attention mechanism provided by the present invention, the overall structure of the method is as follows figure 1 shown. In this method, U-Net is selected as the backbone network, and an inter-pixel-inter-channel double attention module is designed to enhance the utilization efficiency of features and reduce noise and background pixel interference; a circular convolution module is designed to refine the edge contours of significant regions through cyclic iterations; Multi-stage constraints on prediction results using a bypass output strategy. The dual-attention module proposed by the present invention considers attention from two perspectives of channel and space, calibrates the response weight between channels, highlights the pixel response strength of the foreground area, enhanc...

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 a multi-feature cyclic convolution significance target detection method based on an attention mechanism. The method comprises the following steps: ; the method comprises the following steps of: 1, analyzing common characteristics of a salient target in a natural image, including spatial distribution and contrast characteristics, using an improved U-Net full convolutional neural network, performing pixel-by-pixel prediction by adopting an encoder-decoder structure, and performing multi-level and multi-scale characteristic fusion between an encoder and a decoder by adopting a cross-layer connection mode; secondly, a large number of clutters can be introduced to interfere with the generation of a final prediction graph by carrying out concentage fusion on coding end features and decoding end features, so that an attention module is introduced to calibrate full-pixel weights from two angles between channels and between pixels, the task-related pixel weights are enhanced, and the background and noise influence is weakened; and 3, a multi-feature cyclic convolution module is used as a post-processing means, the spatial resolution capability is enhanced through iteration, the edge of an image region is further refined and segmented, and a finer significant target mask is obtained.

Description

technical field [0001] The invention belongs to the field of automatic target recognition, and in particular relates to a research on a multi-feature circular convolution salient target detection method based on an attention mechanism. Background technique [0002] Salient object detection is a basic research hotspot in the field of machine vision. Using the salient object detection method to calculate the saliency of image pixels can highlight the pixels in the foreground area of ​​the image, suppress the pixels in the background area, and achieve the purpose of data dimensionality reduction and background interference reduction. The generated saliency map will help to rationally allocate limited computing resources and provide prior information for subsequent complex vision tasks. Salient object detection has a wide range of applications in image retrieval, image / video compression, image quality assessment, virtual vision and other fields. [0003] Traditional salient obj...

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): G06T7/11G06N3/04
CPCG06T7/11G06T2207/20084G06N3/045
Inventor 席建祥谢学立李传祥杨小冈王乐胡来红
Owner 中国人民解放军火箭军工程大学
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