A learning-based image processing method and device for geometric key points of fundus blood vessels
An image processing and fundus image technology, applied in the field of image processing, can solve the problems of finding the lesion area, insufficient geometric key points of blood vessels, etc., and achieve the effects of wide distribution, improved stability and reliability, and high repeatability
Active Publication Date: 2018-06-19
SHENZHEN REETOO BIOTECHNOLOGY CO LTD
View PDF4 Cites 1 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] In view of the deficiencies in the prior art above, the purpose of the present invention is to provide a method and device for image processing of fundus vascular geometric key points based on learning, aiming at solving the problem of lack of vascular geometric key points and difficulties in existing fundus image processing and registration. Problems such as finding the lesion area and precise registration in the fundus image with serious lesions
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0029] The present invention provides a learning-based image processing method and device for geometric key points of fundus blood vessels. In order to make the purpose, technical solution and effect of the present invention clearer and clearer, the present invention will be further described in detail below. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.
[0030] Such as figure 1 Shown is a specific embodiment of a learning-based image processing method for geometric key points of fundus blood vessels in the present invention. The method comprises the steps of:
[0031] S100. Obtain a fundus image, obtain a dendrogram of the fundus image according to a deep learning algorithm, and process the dendrogram with a graph connectivity algorithm to obtain a retinal vascular tree image.
[0032] Deep learning (deep learning), also known as feature learning, is a new field in m...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a learning-based eyeground blood vessel geometric key point image processing method and an apparatus. The learning-based eyeground blood vessel geometric key point image processing method comprises the steps of obtaining eyeground image dendrogram based on depth learning algorithm, and conducting image communication algorithm processing on the dengrogram and obtaining a retina vascular tree image; calculating curvature entropy dispersion to obtain and store a plurality of geometric key point positions in the retina vascular tree image. The depth learning method is adopted to divide the retina vascular tree image, improving the reliability of finding the appropriate feature points. After the dividing of the blood vessel images, the blood vessel mark point measurement and image processing method based on geometric significance is adopted, and feature points which are stable in existence, wide in distribution and high in repetition in the eyeground red free light / angiography images can be found out. The learning-based eyeground blood vessel geometric key point image processing method can provide more reliable technology support for ophthalmologist in eyeground disease image processing and eyeground laser operation images.
Description
technical field [0001] The present invention relates to the technical field of image processing, in particular to a method and device for image processing of geometric key points of fundus blood vessels based on learning. Background technique [0002] Retinal fundus images are an important basis for the diagnosis of diabetes, glaucoma, hypertension, coronary heart disease and other diseases. These diseases usually lead to changes in the shape of retinal blood vessels, and the characteristics of retinopathy are closely related to the characteristics of various stages of many diseases. Therefore, research Retinal fundus images can provide important basis for medical workers to diagnose diseases. [0003] Fundus images are very important diagnostic data in fundus examinations. At present, the analysis of fundus images is mainly carried out through the images acquired by fundus cameras. The analysis software realizes the analysis and processing of the image, so as to finally ob...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06T7/00G06T7/10
Inventor 李乔亮顾其威邓永春梁平谢林培苏柔
Owner SHENZHEN REETOO BIOTECHNOLOGY CO LTD