Design of a deep-learning-based holographic diagnosis and treatment electromechanical vision system for traditional Chinese medicine

Abstract

The improvement of the efficiency of traditional Chinese medicine (TCM) diagnosis and treatment through motor driven camera arm and data-driven means, as well as the assistance in TCM diagnosis and treatment, which is an important purpose direction for the future. The motor driven camera existing research has mainly focused on the intelligent control, which the paper has introduced how to make intelligent diagnostic tools more flexible and efficient in assisting TCM diagnosis and treatment on the recommendation of TCM knowledge databases. This paper has been introduced the datadriven means design of a TCM holographic diagnosis and treatment visual system based on web, which has token the open source framework of everyone as the template for secondary development, and completed the implementation of humancomputer interaction interface. The holographic visual diagnosis and treatment part of TCM was encapsulated into an interface by the python flask framework to complete the connection between the visual diagnosis and treatment part and the web interaction module of TCM diagnosis and treatment. The SVM and CNN convolutional neural network were used to classify tongue diagnosis respectively, and the advantages and disadvantages of the two algorithms were compared. CNN convolutional neural network and opencv's haar classifier were used to realize simple facial diagnosis and treatment. Finally, web RTC technology was used to achieve real-time acquisition of camera information for diagnosis. It lays a theoretical and methodological foundation for the analysis and design of traditional Chinese medicine holographic diagnosis and other intelligent auxiliary diagnosis tools, and has important guiding significance for the realization and development of traditional Chinese medicine holographic diagnosis system.