A capsule gastroscope trajectory tracking method and device

Abstract

The present invention provides a capsule gastroscope trajectory tracking method and device, the method comprising: acquiring a video picture sequence captured by a capsule gastroscope; inputting each frame of pictures in the video picture sequence into a trained picture feature extraction model in sequence, The picture feature points and corresponding feature descriptors of each frame of pictures are obtained; according to the picture feature points and the feature descriptors, the trajectory tracking of the capsule gastroscope is performed through the ORB SLAM3 algorithm. The present invention combines the deep learning network with the ORB SLAM3 algorithm to realize the matching and tracking of the image features through the ORB SLAM3 algorithm on the basis of using the convolutional neural network to extract the feature vector, so as to improve the robustness of the extracted features, so that the capsule The trajectory tracking of the gastroscope realizes precise positioning.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a capsule gastroscope track tracking method and device. Background technique [0002] There are currently two methods for capsule gastroscope trajectory tracking, one is to use a capsule with a magnetic marker inside, and the other is to additionally stain the stomach. [0003] For the realization of the first method, it is necessary to provide multiple magnetic field sensors. The magnetic field sensors are distributed and fixed at set positions on the body surface, and are connected with the power manager and multi-way switch of the external positioning control device to capture the three-dimensional position of the capsule in the body. motion track. For capsule trajectory tracking using capsules with magnetic markers inside, the tester needs to wear multiple magnetic field sensors on the body, which will also cause inconvenience to the tester during the test process; ...

AI Technical Summary

Problems solved by technology

For the method of using additional dye, since the examiner needs to use additional dye during detection, when the doctor detects and identifies the lesion, it is affected by the dye, and the difficulty for the doctor to identify the lesion will be greatly increased. This method plays a role. The effect is counterproductive. Even if the trajectory of the capsule can be tracked, it will eventually affect the accuracy of the picture and is not suitable for practical applications.

[0005] Based on the above-mentioned existing technology when tracking the trajectory of the capsule gastroscope, the problem of low tracking accuracy exists, and now there is an urgent need for a capsule gastroscope trajectory tracking method and device to solve the above problems

Images