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

A human contour extraction method based on depth learning

A human body contour and deep learning technology, which is applied to instruments, character and pattern recognition, and computer components, etc., can solve problems such as poor human body contour extraction and slow model training speed

Inactive Publication Date: 2018-12-18
XIAN UNIV OF TECH
View PDF9 Cites 18 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a human body contour extraction method based on deep learning, which solves the problems in the prior art that the human body contour extraction effect in static images is poor and the model training speed is slow

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
  • A human contour extraction method based on depth learning
  • A human contour extraction method based on depth learning
  • A human contour extraction method based on depth learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0163] The source of the data set is Baidu's human body image segmentation database. The data in this database are images containing human bodies taken from various angles. There are 5387 training images and labeled samples in the database. The present invention selects 1,000 images as the training set, and selects 500 images as the test set in the remaining part. In the experiment, the network input image size is fixed at 224×224. In order to accurately and objectively evaluate the effect of the method in this paper, and to facilitate comparison with existing methods, the overlap rate is used to measure the performance of the improved human contour extraction model. The overlap rate is defined as follows:

[0164]

[0165] Among them, S is the degree of overlap, A P Extract the body area predicted by the network for the body contour, A GT Is the actual human body area. The higher the S, the higher the degree of overlap and the better the effect of human contour extraction.

...

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 human contour extraction method based on depth learning, which is implemented according to the following steps: step 1, extracting Gabor texture features of an original image; 2, extracting Canny edge features of that original image; 3, building a convolution neural network structure suitable for human body contour extraction; 4, transmitting the original image, the Gabortexture feature map extracted in the step 1 and the Canny edge feature map extracted in the step 2 into the convolution neural network construct in the step 3 for training to generate a CNN charactermodel; 5, testing that structure of the trained CNN character model to obtain a human body contour image; Step 6, recording the overlap rate and time consumption of the human body contour images through the testing process of step 5, and evaluating the human body contour images. The method of the invention achieves high accuracy, improves detection rate and shortens test time.

Description

Technical field [0001] The invention belongs to the technical field of machine vision, and specifically relates to a method for extracting a human body contour based on deep learning. Background technique [0002] Human body contour extraction occupies an important position in the field of computer vision, and is the core technology of human body detection and human behavior recognition. Human contour extraction technology is currently widely used in intelligent monitoring, medical treatment and other fields. The virtual reconstruction of the human body model is a key technology in the modern medical visualization system. Only accurate human contour information collection can guarantee a reasonable medical analysis of the patient's symptoms. On the other hand, with the strengthening of the requirements for personal and public property safety in modern society, the utilization rate of intelligent monitoring systems is gradually increasing. The primary goal of intelligent video s...

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): G06K9/00G06K9/46G06K9/62
CPCG06V40/10G06V10/449G06F18/2413
Inventor 王林董楠
Owner XIAN UNIV OF TECH