Intelligent excrement analysis method based on computer vision and neural network

Abstract

The invention discloses an intelligent excrement analysis method based on computer vision and a neural network, and belongs to the technical field of digital image processing and machine learning. Themethod comprises the following steps: detecting the color of excrement; acquiring a 3D image of the excrement, and obtaining the excrement character through automatic analysis; acquiring a microscopic examination image according to the excrement color and the excrement character, and analyzing the microscopic examination image to obtain an excrement visible component; and integrating and classifying the analysis results of the excrement color, the excrement character and the excrement visible component. According to the method, intelligent and automatic analysis of the excrement can be realized, the color and character of the excrement can be accurately recognized, visible components are selected for detection in a targeted manner, results are classified, the method is an effective supplement to a traditional excrement recognition technology, the efficiency is higher than that of a traditional recognition method, and visual discomfort of detection personnel can be reduced.

Description

technical field [0001] The invention relates to the fields of digital image processing technology and machine learning technology, in particular to an intelligent stool analysis method based on computer vision and neural network. Background technique [0002] At present, the commonly used stool inspection methods include manual method and instrument method. The manual method needs to identify the color and shape of the stool with the naked eye, and pick the abnormal parts of the stool for detection; although the application of the automatic stool analyzer makes routine stool inspection convenient, Standard, but for the identification of stool color and properties, the instrument method still uses visual judgment after taking pictures, and the inspection of stool under a microscope is to add diluent to all the stool in the stool box, mix well, and then draw the suspension for inspection. The process of visual judgment requires visual recognition of each stool. The suspension ...

AI Technical Summary

Problems solved by technology

[0002] At present, the commonly used stool inspection methods include manual method and instrument method. The manual method needs to identify the color and shape of the stool with the naked eye, and pick the abnormal parts of the stool for detection; although the application of the automatic stool analyzer makes routine stool inspection convenient, Standard, but for the identification of stool color and properties, the instrument method still uses visual judgment after taking pictures, and microscopic examination of stool is to add diluent to all the stool in the stool box, mix well, and then draw the suspension for inspection

The process of visual judgment requires visual recognition of each stool. The suspension drawn after adding diluent and mixing may be diluted due to the small amount of abnormal substances, so that it cannot be found under the microscope, resulting in a faster recognition of stool properties. Not fast enough, and the selection of abnormal stool parts is not accurate enough

Images