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

Semantic segmentation method, device and storage device by suppressing non-interesting information

A semantic segmentation, non-interesting technology, applied in neural learning methods, image analysis, image enhancement, etc., can solve problems such as multiple computing resources and unfavorable model deployment

Active Publication Date: 2022-06-24
北京创时空科技发展有限公司
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In the process of building a neural network, the stacking of simple neural network layers cannot suppress non-interesting information; to achieve better generalization effects, it is necessary to expand the receptive field of the model, but it requires a lot of computing resources. Not conducive to the deployment of the model on mobile devices

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 segmentation method, device and storage device by suppressing non-interesting information
  • Semantic segmentation method, device and storage device by suppressing non-interesting information
  • Semantic segmentation method, device and storage device by suppressing non-interesting information

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0028] In order to have a clearer understanding of the technical features, objects and effects of the present invention, the specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0029] refer to figure 1 , figure 1 is a flowchart of a semantic segmentation method by suppressing non-interesting information. The semantic segmentation method by suppressing non-interesting information of the present invention comprises the following steps:

[0030] Step 1), optimize the neural network based on the deep learning library, and use tensorflow to build a basic image semantic segmentation model Unet, including two stages of encoding and decoding: the encoding stage performs 4 downsampling in series, and each downsampling passes through 2 layers of space A 3×3 convolutional layer and a pooling layer with gradually increasing dimensions and the same feature map size. The convolutional layer adopts the conv function i...

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 present invention proposes a semantic segmentation method, device and storage device by suppressing non-interesting information. The present invention optimizes the neural network based on the deep learning library and improves the accuracy of the semantic segmentation results. It mainly includes the following steps: 1) Constructing the basic Unet model ; 2) Add an attention mechanism; 3) Multiply the gate feature map with the current layer result; 4) Add new output results and multiple loss functions; 5) Perform image semantic segmentation on the image to be semantically segmented. This method can improve the accuracy of semantic segmentation neural network and effectively suppress non-interesting information.

Description

technical field [0001] The present invention relates to the field of semantic segmentation, and more particularly, to a method, device and storage device for semantic segmentation by suppressing non-interesting information. Background technique [0002] In the process of using neural network for semantic segmentation of images, it is necessary to build a neural network first. In the process of building a neural network, the simple stacking of neural network layers cannot suppress non-interesting information; in order to achieve better generalization effect, it is necessary to expand the receptive field of the model, but it requires a lot of computing resources. Not conducive to the deployment of models on mobile devices. The present invention proposes a solution based on the above-mentioned problems. SUMMARY OF THE INVENTION [0003] The purpose of the present invention is to propose a high-precision semantic segmentation method for suppressing non-interesting informatio...

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 Patents(China)
IPC IPC(8): G06T7/11G06V10/44G06K9/62G06N3/04G06N3/08G06V10/80
CPCG06T7/11G06N3/082G06T2207/10004G06V10/44G06N3/045G06F18/25
Inventor 刘恒郭明强黄颖吴亮谢忠关庆锋韩成德宋振振
Owner 北京创时空科技发展有限公司