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

Image saliency detection method combining graph theory and improved hierarchical cellular automaton

A cellular automaton and detection method technology, which is applied in the field of computer vision research and can solve problems such as incomplete target detection and background redundancy.

Pending Publication Date: 2021-10-01
CHENGDU UNIVERSITY OF TECHNOLOGY
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] The present invention proposes an image saliency detection method combining graph theory and improved hierarchical cellular automata, which solves the problems of background redundancy and incomplete target detection in the prior art

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
  • Image saliency detection method combining graph theory and improved hierarchical cellular automaton
  • Image saliency detection method combining graph theory and improved hierarchical cellular automaton
  • Image saliency detection method combining graph theory and improved hierarchical cellular automaton

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0045] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts all belong to the protection scope of the present invention.

[0046] An image saliency detection method combining graph theory and improved hierarchical cellular automata disclosed by the present invention includes step A: establishing a correlation matrix between low-level convolution features and deep-level convolution features: (1) segmenting the image; (2) ) Use the FCN network to extract low-level convolutional features and deep convolutional features from the segmented image; (3) Non-linearly fuse t...

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 relates to the field of computer vision research, in particular to an image saliency detection method combining a graph theory and an improved hierarchical cellular automaton, and solves the problems of background redundancy and incomplete target detection in the prior art. The method comprises the following steps: A, establishing a correlation matrix of low-layer convolution features and deep-layer convolution features: (1) segmenting an image; (2) extracting low-layer convolution features and deep-layer convolution features of the segmented image by using an FCN network; and (3) non-linearly fusing the correlation between the two layers of hierarchical features, and constructing a relation matrix. According to the method, the background is detected through iterative updating of the optimized similar matrix, the lost foreground is complemented to a certain extent, the association between the foreground and the background is enhanced while superposition is carried out, and background nodes are activated; a background seed selection method is used for enhancing the boundary detection capability, so that dominant states of similar boundary nodes are consistent as much as possible; the significance diffusion method further inhibits the background and optimizes the detection result.

Description

technical field [0001] The invention relates to the field of computer vision research, in particular to an image saliency detection method combining graph theory and improved hierarchical cellular automata. Background technique [0002] Saliency detection is an important research field in computer vision. Its purpose is to imitate the human visual system to find the most salient regions in the scene efficiently and quickly. In the field of computer vision, saliency detection is applied to image description, understanding, object segmentation and semantic segmentation, etc. In computer graphics, saliency detection can be applied to automatic image cropping, image relocation, etc. [0003] Traditional saliency detection is mainly based on a bottom-up approach, which solves saliency detection tasks through handcrafted low-level features and heuristic prior information; in recent years, the rapid development of deep learning has greatly improved the saliency detection performanc...

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): G06K9/32G06K9/34G06N3/04
CPCG06N3/045G06N3/0464G06N3/042Y02T10/40
Inventor 吴媛媛苟佳成
Owner CHENGDU UNIVERSITY OF TECHNOLOGY