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

Traditional Chinese medicine tongue image tooth mark automatic detection method based on convolutional neural network multi-scale feature fusion

A convolutional neural network, multi-scale feature technology, applied in neural learning methods, biological neural network models, medical images, etc., can solve the problem of lack of semantic information in the tooth mark area, improve the expression ability, meet application requirements, reduce The effect of complexity

Pending Publication Date: 2022-01-28
BEIJING UNIV OF TECH
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, unlike traditional target detection objects, the color of the tooth mark area of ​​the tongue image is similar to the color of the surrounding tongue body, and the tooth mark area also lacks clear semantic information. In order to achieve accurate detection of tooth marks, it is necessary to improve the expressive ability of depth features.

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
  • Traditional Chinese medicine tongue image tooth mark automatic detection method based on convolutional neural network multi-scale feature fusion
  • Traditional Chinese medicine tongue image tooth mark automatic detection method based on convolutional neural network multi-scale feature fusion
  • Traditional Chinese medicine tongue image tooth mark automatic detection method based on convolutional neural network multi-scale feature fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0032] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. The following is only a specific implementation process, but the scope of protection of this patent is not limited to this implementation process. The overall block diagram of the method proposed by the present invention is as follows: figure 1 The specific implementation process is as follows:

[0033] Step 1: Build a tooth mark detection dataset

[0034] The tongue images used in the present invention are collected by a self-developed SIPL TCM tongue imager, and then calibrated one by one by a TCM physician. In order to remove the influence of the background on the detection result, the present invention firstly divides the collected tongue image, extracts the tongue body area, and then marks the tooth mark area in the tongue body to construct a tooth mark detection data set.

[0035] Step 1.1: Tongue body region extraction. The original...

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 traditional Chinese medicine tongue image tooth mark automatic detection method based on convolutional neural network multi-scale feature fusion. The method comprises steps: using convolutional neural network VGG16 as a basic network to extract features; providing a multi-scale feature fusion module and a feature enhancement module which are respectively used for fusing different scale features of the convolutional neural network and enhancing the fused features to form four detection layers; and finally, respectively carrying out tooth mark detection on the four detection layers by adopting an SSD method, and fusing all detection results by adopting a non-maximum suppression method to obtain a final tooth mark detection result. The method can improve the feature expression capability of each detection layer, achieves the automatic and accurate detection of the tooth marks in the traditional Chinese medicine tongue image, has obvious advantages in the detection precision compared with a traditional method, and can meet the actual application demands. According to the invention, the complexity of a network model is greatly reduced. The method has obvious advantages in detection precision, and can meet actual application requirements.

Description

technical field [0001] The invention belongs to the field of computer vision and TCM diagnostics, and specifically relates to technologies such as computer image processing, deep learning, and TCM tongue diagnosis. Background technique [0002] The automatic analysis of TCM tongue characteristics is the core content of the objectification of tongue diagnosis, and the accuracy of the analysis results determines the reliability of the follow-up treatment and the acceptance of TCM practitioners. According to the diagnosis principle of TCM, the changes in the characteristics of the tongue reflect the functional pathological changes of the viscera of the human body. Tooth marks are one of the important tongue features, mostly caused by qi deficiency, spleen deficiency or yang deficiency. Teeth-marked tongue is caused by the inability of the spleen to transport water and dampness, resulting in a fat tongue that is compressed by the teeth margins. It has great guiding significance...

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): G06T7/00G16H30/20G06V10/80G06K9/62G06N3/04G06N3/08
CPCG06T7/0012G16H30/20G06N3/08G06N3/045G06F18/253
Inventor 卓力黄晓东孙亮亮张辉李晓光张菁
Owner BEIJING UNIV OF 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