Image semantic segmentation method based on pyramid pooled coding-decoding structure

A technology of pyramid pooling and semantic segmentation, which is applied in the field of computer vision, can solve problems such as a large number of computing resources, lack of running speed, and difficulty in application, and achieve the effects of reducing model parameters, improving performance, and increasing running speed

Inactive Publication Date: 2018-01-30
UNIV OF SCI & TECH OF CHINA
View PDF5 Cites 57 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This way of pursuing high performance makes the network model of the mainstream method extremely large, requires a lot of computing resources, and is difficult to be practically applied in life.
At the same time, w

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 semantic segmentation method based on pyramid pooled coding-decoding structure
  • Image semantic segmentation method based on pyramid pooled coding-decoding structure
  • Image semantic segmentation method based on pyramid pooled coding-decoding structure

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] 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. Obviously, 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 belong to the protection scope of the present invention.

[0036] figure 1 It is a flowchart of an image semantic segmentation method based on a pyramid pooling codec structure provided by an embodiment of the present invention. Such as figure 1 As shown, it mainly includes:

[0037] Step 1. Process the input image through an encoding network including a convolutional neural network model and a pyramid pooling model, extract high-dimensional feature information of the input image, and form a deep feature...

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 image semantic segmentation method based on a pyramid pooled coding-decoding structure. The method comprises: an inputted image is processed by a coding network including aconvolutional neural network model and a pyramid pooling model, high-dimensional feature information of the inputted image is extracted, and a deep feature map is formed; the deep feature map is restored to one with the resolution ratio identical with that of the inputted image by a decoding network including an inverse convolutional neural network model; and a feature map outputted by the decoding network is classified by using a classifier including a convolutional layer with a class number of output channels and a Softmax layer, thereby realizing image semantic segmentation. With the methoddisclosed by the invention, the capability of image semantic segmentation by the network is enhanced and the network model size and the operation speed are both considered.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to an image semantic segmentation method based on a pyramid pooling codec structure. Background technique [0002] Image semantic segmentation combines traditional image segmentation and target recognition tasks. It needs to segment the image into several groups of pixel regions with specific semantic meanings and identify each type of region to obtain an image with semantic annotations. For example, for a traffic image containing a complex scene, the computer needs to automatically identify categories such as pedestrians, vehicles, buildings, etc. For the same category of objects, it needs to automatically segment its accurate shape and position, and express it with the same pixel value . [0003] At present, the algorithms of image semantic segmentation are mainly divided into two categories. The first type can be called the traditional method, which utilizes the relationship betw...

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): G06T7/11G06K9/62G06N3/04
Inventor 谭振涛刘斌俞能海
Owner UNIV OF SCI & TECH OF CHINA
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