Method and device for detecting mark points in medical perspective image and electronic equipment

Abstract

The invention relates to the field of medical image processing, in particular to a method for detecting mark points in a medical perspective image, the number of the mark points in the medical perspective image is multiple, the mark points are obtained by imaging a plurality of mark points arranged on a calibration ruler, and a determined geometrical relationship exists among the mark points. The method comprises the following steps: identifying the medical perspective image to obtain an identification point; obtaining a prediction point group based on the identification points and the geometrical relationship; for each prediction point group, based on the distance between the prediction point and the identification point, obtaining a prediction matching point corresponding to the prediction point, and calculating the score value of the prediction point group; and finding an optimal prediction point group based on the prediction point group score value, and matching the mark point based on the optimal prediction point group.

Description

technical field [0001] The present application relates to the field of medical image processing, in particular to a method, device and electronic equipment for detecting marker points in medical perspective images. Background technique [0002] Medical fluoroscopy images, such as X-ray image processing, involve the positioning of surgical robots and marking points, which is one of the core technologies of medical image processing. With the rapid development of the interdisciplinary application of robotics and medical science, the research of various medical robots has also become a hot spot, and has been more and more widely used in the medical field. At present, surgical robots have been widely used in neurosurgery, artificial joint replacement, urology, and gallbladder removal. During the navigation and positioning process of the surgical robot, the C-arm is generally used to take the X-ray fluoroscopic image of the patient through the calibration scale with the marking p...

AI Technical Summary

Problems solved by technology

For example, in patent CN104123540B, it adopts the traditional image detection of marked points and the matching method of preset feature points and angles to complete, the detection accuracy is limited to about 92%, and there are the following deficiencies: 1. The connection between feature points and each point The line angle dependence is high, so that the matching cannot be completed when the angle between the photographic plane and the scale plane exceeds 5 degrees; 2. Due to the low robustness of traditional image detection in the case of relatively strong interference, in the interference environment such as occlusion and fogging 3. The front and back sides of the ruler are preset, and if the front and back are reversed during the actual operation, the accuracy of the algorithm will be significantly reduced

Images