Corneal reflection image screening method and device, intelligent terminal and storage medium

Abstract

The invention discloses a corneal reflection image screening method and device, a terminal and a storage medium, and the method comprises the steps: obtaining a corneal reflection image, preprocessingthe corneal reflection image to acquire a test sample; inputting the test samples into the trained network model for segmentation processing, and performing preliminary screening on the test samplesbased on the segmentation processing; and performing secondary screening on the preliminarily screened test samples based on circular ring center positioning and evaluation, performing strong edge detection on the test samples subjected to secondary screening, outputting a definition index parameter of each corneal reflection image, and performing third screening on the corneal reflection images based on the definition index parameters. According to the invention, the corneal reflex images are preliminarily screened based on the network constructed by deep training, and then further screened based on strong edge detection, so that the corneal reflex images can be screened and evaluated without human intervention, and the evaluation accuracy is improved.

Description

Technical field[0001]The present invention relates to the technical field of corneal image processing, in particular to a method, device, terminal and storage medium for screening corneal reflection images.Background technique[0002]The refractive power of the cornea accounts for about 70% of the total refractive power of the human eye, so the change of the corneal shape directly affects the change of the refractive power of the entire human eye. Surgery of the sensory cornea changes the corneal refractive power by changing the surface morphology of the cornea. In the past hundred years, scientific researchers have been working hard in this field. From the early estimation of corneal reflections to the current corneal topography using computer technology, it has experienced a long development process. At present, most eye hospitals in China have introduced corneal topographers, and they have played an important role in routine eye examinations, guiding refractive surgery, and guiding...

AI Technical Summary

Problems solved by technology

It can be seen that the existing quality evaluation of corneal reflection images or image selection methods for diagnosis requires human participation, so there are subjective errors, which affect the evaluation results of image quality

Images