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

Semantic image segmentation method based on multichannel convolutional neural network

A convolutional neural network and image segmentation technology, applied in the field of pattern recognition, can solve problems such as ignoring the shallow features of the network, achieve the effect of enriching global information, increasing the receptive field, and improving segmentation performance

Inactive Publication Date: 2018-02-02
CHINA UNIV OF MINING & TECH
View PDF3 Cites 27 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the traditional convolutional neural network often only considers the deep features and ignores the shallow features of the network.

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
  • Semantic image segmentation method based on multichannel convolutional neural network
  • Semantic image segmentation method based on multichannel convolutional neural network
  • Semantic image segmentation method based on multichannel convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0028] The present invention will be further explained below in conjunction with the accompanying drawings.

[0029] A semantic image segmentation method based on a multi-channel convolutional neural network, comprising the steps of:

[0030] Step 1, construct a multi-channel convolutional neural network model, which contains 6 channels in total, and realize the fusion of shallow and deep features by "summing" the output of each channel. Compared with the single-channel network, this structure makes the final extracted features contain more information.

[0031] Such as figure 1 As shown, the multi-channel convolutional neural network model includes an input layer, a convolutional layer, a pooling layer, an activation function layer, a bilinear interpolation amplification layer, and a loss layer. The multi-channel convolutional neural network model has a total of 6 channels, which are A channel, B channel, C channel, D channel, E channel and F channel. The structure and para...

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 a semantic image segmentation method based on a multichannel convolutional neural network. A provided network model comprises 6 channels; fusion of shallow and deep features isrealized through addition operation of outputs of the channels; and compared with a single-channel network, the structure can improve segmentation performance of semantic images. The whole network model structure improves a receptive field by combining an a'trous algorithm, so that the captured global information is allowed to be richer; and in the test phase, the segmentation result is subjectedto optimization through a full connection condition random field, so that the segmentation performance of the semantic images can be further improved.

Description

technical field [0001] The invention belongs to the field of pattern recognition, and in particular relates to a semantic image segmentation method based on a multi-channel convolutional neural network. Background technique [0002] Image segmentation is an important research direction in machine vision. It is the process of dividing a digital image into multiple areas of interest. Objects in the same area have the same or similar properties. The purpose of segmentation is to enable the image to Better to express its meaning and easy to analyze, the segmented results can be used in image retrieval, object detection, automatic driving and other fields. Traditional image segmentation methods include histogram method, region segmentation, edge segmentation and so on. A common practice is that, for a given image, take an area centered on a certain pixel in the image, extract the corresponding features from this area, and use this feature to train the classifier, and finally cla...

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): G06K9/34G06K9/62G06N3/04G06N3/08
CPCG06N3/082G06V10/267G06N3/045G06F18/253G06F18/214
Inventor 王雪松曾杰川程玉虎
Owner CHINA UNIV OF MINING & TECH
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