Recognition method, system and equipment for medical endoscope image and endoscope image system

Abstract

The invention discloses a recognition method, system and equipment for a medical endoscope image and an endoscope image system. The method comprises the steps of obtaining an original endoscope imageaccording to a medical endoscope video stream; filtering the obtained original endoscope image through a neural network to generate a target endoscope image; predicting and recognizing organ information corresponding to the target endoscope image through classification of a neural network; according to the corresponding organ information, identifying an image type applicable to the target endoscope image through a classification network; and in the shooting mode corresponding to the image type, positioning the focus area and the focus type in the target endoscope image according to the part indicated by the organ information. The method has the advantages that the shooting switching and shaking of the endoscope in the alimentary canal are eliminated through filtering, a large amount of low-quality noise images and robustness are enhanced under the conditions of various liquids and foreign matters, the classification prediction is realized for the whole shooting process of the endoscope, and the systematic and complete image recognition is realized.

Description

technical field [0001] The invention relates to the technical field of computer applications, in particular to a medical endoscope image recognition method, system, machine equipment and an endoscope image system. Background technique [0002] The recognition of various categories based on deep learning has always been an important tool for solving large amounts of data classification in various application scenarios. For example, in application scenarios such as image and natural language processing, large-scale classification and recognition of large amounts of data can be implemented to quickly and accurately obtain relevant classification prediction results and accelerate the function realization of the application scenario. [0003] In the classification prediction of images, depending on the deployed application scenarios, for example, AI+medical scenarios, the specific images used to achieve classification prediction and the realization of classification prediction ar...

AI Technical Summary

Problems solved by technology

[0008] In order to solve the technical problem that the classification and prediction of medical images in the related art cannot be applied to the whole process of endoscopic shooting of medical endoscopic images and has poor robustness, the present invention provides a recognition method, system and machine for medical endoscopic images Recognition of medical endoscopic images of equipment and endoscopic imaging systems

Images