Digestive tract lesion image recognition system and recognition method

Abstract

The invention discloses an image recognition system for digestive tract lesions, which includes a memory, an image preprocessing module, an image feature extraction module, a machine learning module, and an image recognition module, wherein the storage data communication end of the memory is connected to the image preprocessing module The data input end of the image preprocessing module, the data output end of the image preprocessing module is connected to the data input end of the image feature extraction module, the first data output end of the image feature extraction module is connected to the data input end of the machine learning module, and the second data of the image feature extraction module The output end is connected to the first data input end of the image recognition module, and the data output end of the machine learning module is connected to the second data input end of the image recognition module. The invention improves the efficiency and accuracy of image recognition of digestive tract lesions.

Description

technical field [0001] The present invention relates to the technical fields of image recognition and image processing, in particular to an image recognition system and method for digestive tract lesions. Background technique [0002] The use of capsule endoscope for gastric detection can free people from the pain and discomfort caused by traditional gastroscopy, which is a new direction for the development of gastroscopy. In the process of gastric inspection with capsule endoscopy, thousands of pictures will be generated in one inspection. If the small intestine inspection is added, the number of inspection pictures will exceed 50,000. In the future, with the increase of frame rate of capsule image transmission and power consumption A further decrease of will generate more image data. The increase of image data will increase the time and difficulty of manual reading. [0003] Patent CN103984957A of Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, i...

AI Technical Summary

Problems solved by technology

[0004] 1. The lesion identification and processing of gastric images are not involved, so the influence of bubbles, impurities and other features in the digestive tract images on lesion identification is not considered;

[0005] 2. The rotation of the capsule endoscopic image is not considered, so the rotation invariant features of the endoscopic image are not extracted for lesion identification;

[0006] 3. There is no identification method for non-flat lesions such as tumors and polyps;

[0007] 4. No redundant algorithm is used to reduce the number of capsule endoscopic images;

[0008] 5. There is no proposed algorithm for classifying the organs of the entire digestive tract, so it is impossible to effectively generate auxiliary diagnostic results for the stomach or esophagus

Images