A dual-image visual navigation method in the curved intestine of a dual-hemispherical capsule robot

Abstract

The invention discloses a dual-image visual navigation method in the curved intestinal tract of a dual-hemispherical capsule robot, which belongs to the technical field of automation engineering. This method uses the coaxial follow-up characteristics of the axis of the dual hemispherical capsule robot and the axis of the universal rotating magnetic field in the attitude adjustment mode to determine the pitch angle of the side swing of the axis, through the same pitch angle and two different side swing Two consecutive images taken by the monocular camera under the corner determine the rotation angle of the camera frame around the axis of the dual-hemispherical capsule robot, and then determine the camera attitude; then use the uniformity of the universal rotating magnetic field and combine the attitude information to determine the position in the fixed coordinates through coordinate conversion. The center of mass of the dark area of ​​the image of the curved intestine in the system is relative to the direction vector of the dual-hemispherical capsule robot to realize the navigation of the dual-hemispherical capsule robot in the curved intestine by the universal rotating magnetic field. The present invention can realize the navigation of the dual hemispherical capsule robot in the intestinal tract without adding any sensor or device and avoiding the use of complicated three-dimensional reconstruction technology.

Description

technical field [0001] The invention belongs to the technical field of automation engineering, and relates to a dual-image visual navigation method for a dual-hemispherical capsule robot driven by a space-universal rotating magnetic field in a curved intestinal tract. Background technique [0002] The human gastrointestinal tract is prone to various fatal diseases, among which colorectal cancer is the most common disease. However, if most gastrointestinal diseases are detected and diagnosed early, the cure rate will be significantly improved. Therefore, gastrointestinal examination and diagnosis are very important in the medical field. The most commonly used instrument for examining gastrointestinal diseases is the traditional endoscope, which cannot inspect the entire intestinal tract due to the limitation of the length of the catheter that can be inserted. Moreover, during the insertion of the endoscopic catheter, it is easy to cause damage to the soft tissue of the gast...

AI Technical Summary

Problems solved by technology

[0004] Although the above-mentioned capsule endoscope technology is relatively mature, there are still many problems in terms of clinical application feedback: (1) Active control of capsule robots: since most of the existing clinical capsule endoscopes do not have active walking mechanisms, they all rely on gastric Intestinal peristalsis advances passively, so its movement in the gastrointestinal tract is random, and the capsule cannot return after missing the lesion, and doctors cannot observe the area of ​​interest in detail, so the missed detection rate is high and the inspection efficiency is low

(2) Capsule robot navigation problem: Since the real-time position and attitude (pose) of the existing capsule endoscope in the gastrointestinal tract cannot be determined, it cannot be navigated, so that effective control of the capsule robot cannot be achieved

At present, the common visual navigation uses a multi-eye vision system. However, the internal space of the dual hemispherical capsule robot is small, and the method of installing multiple cameras will be limited by the space of the dual hemispherical capsule robot. There are high requirements for the installation accuracy of the dual hemispherical capsule robot, which makes the internal structure of the double hemispherical capsule robot too complex, and the complex circuit will reduce the reliability of the system

Although the 3D reconstruction technology can calculate the position of the current dual hemispherical capsule robot, the algorithm is complex and requires high image parameters

[0013] At present, no one has proposed to use the uniformity of the universal rotating magnetic field and the follow-up of the dual-hemispherical capsule robot in the magnetic field, without installing any sensors or other devices in the dual-hemispherical capsule robot. The image is used to determine the attitude information of the dual hemispherical capsule robot camera relative to the fixed coordinate system and determine the navigation direction through coordinate transformation. The significant advantage of this method is that it does not need to install other devices in the dual hemispherical capsule robot. The multi-eye vision system and the use of complex three-dimensional image reconstruction technology of the intestinal tract rely on the monocular camera vision of the dual hemispherical capsule robot combined with the uniformity of the universal rotating magnetic field and the follow-up of the dual hemispherical capsule robot in the universal rotating magnetic field. Through coordinate transformation, the final navigation information of the dual hemispherical capsule robot is obtained

Images