Tongue picture classification method based on multitask convolution neural network

A convolutional neural network, neural network technology, applied in neural learning methods, biological neural network models, instruments, etc., can solve problems such as inability to apply to multiple classifications and poor accuracy

A convolutional neural network, neural network technology, applied in neural learning methods, biological neural network models, instruments, etc., can solve problems such as inability to apply to multiple classifications and poor accuracy

CN107977671AActive Publication Date: 2018-05-01ZHEJIANG UNIV OF TECH
9 Cites 99 Cited by

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Tongue picture classification method based on multitask convolution neural network
  • Tongue picture classification method based on multitask convolution neural network
  • Tongue picture classification method based on multitask convolution neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0077] The present invention will be further described below in conjunction with the accompanying drawings.

[0078] refer to Figure 1 to Figure 6 , a tongue image classification method based on a multi-task convolutional neural network, including a series of preprocessing image operations on the collected tongue images; including a deep shared convolutional neural network for tongue overall feature extraction, including for The region-of-interest positioning network for tongue label detection, the multi-task deep convolutional neural network for deep learning and training and recognition, complete the tongue image color, fur color, thickness, rotten greasy, moistening dryness, etc. Label classification for multiple attributes.

[0079] The main process is as follows: when the system receives the tongue image taken by the user, it automatically triggers the preprocessing module to obtain the tongue image after color correction, tongue segmentation, and shadow area removal; 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 discloses a tongue picture classification method based on a multitask convolution neural network, and the method comprises the steps: carrying out a series of preprocessing operations ofa collected tongue picture, and enabling a deep convolution neural network for extracting the overall features of a tongue, an AOI (area of interest) positioning network for detecting a tongue surface label and a multitask deep convolution neural network for deep learning and training recognition to complete the label classification of the tongue color, coating color, coating thickness, curdy andgreasy fur and moistening dryness. The method effectively solve a problem that a conventional method cannot achieve the simultaneous recognition of the tongue color, coating color, coating thickness,curdy and greasy fur and moistening dryness.

Description

technical field [0001] The invention relates to an analysis method, in particular to the application of technologies such as TCM tongue diagnosis, mobile Internet, database management, computer vision, digital image processing, pattern recognition, deep learning and deep convolutional neural network in the field of automatic analysis of tongue images. Background technique [0002] Tongue diagnosis is one of the most direct and basic clinical diagnostic methods of traditional Chinese medicine. It has been praised by many doctors since ancient times and widely used in clinical practice. The tongue image contains rich physiological and pathological information of the human body. By observing the coating on the tongue surface of the patient and related properties of the tongue quality, including color, shape, etc., it is possible to determine the location of the disease and carry out syndrome differentiation and treatment, which is very important for TCM medicine and disease judg...

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
01 May 2018
Publication
CN107977671A
IPC
G06K9/62; G06K9/32; G06N3/08
CPC
G06N3/088; G06V10/25; G06F18/2431; G06F18/214
Inventors
王丽冉; 汤一平