Vascular Image Enhancement Method

Abstract

The present invention provides a blood vessel image enhancement method, the blood vessel image enhancement method comprising: Step 1, collecting blood vessel images; Step 2, determining the Harr-type line filter scale σ; Step 3, using Harr-type line filters in multiple directions Filtering the blood vessel image; and step 4, fusing the filtering results of the Harr-type line filters in each direction into one image. This vascular image enhancement method solves the problems of weak vascular image signals, the inability of gray-scale transformation algorithms such as contrast enhancement to directly and effectively extract lines, and the high complexity of existing calculation methods. Accurate and clear vascular texture extraction can be performed on video images, and real-time processing can be achieved.

Description

technical field [0001] The invention relates to the field of medical application and image enhancement, in particular to a blood vessel image enhancement method. Background technique [0002] The purpose of enhanced vascular texture extraction is to quickly and effectively extract the vascular texture in poor-quality, low-resolution or weak-contrast vascular distribution images, so that the observer can obtain real-time correlation of the vascular texture from the low-quality original image. Information, including grain direction, grain shape and its exact position, etc. At present, the vascular pattern enhancement method is mainly used in the preprocessing of directly assisting medical diagnosis or identification. The former is used, for example, in venipuncture activities. When the target is children, women, obese people, or patients with dark skin, because the blood vessels are thin or the cortex is thick, the blood vessels are not easy to see. The blood vessel enhanceme...

AI Technical Summary

Problems solved by technology

However, these methods often have high complexity, especially in the process of extracting the direction of blood vessel lines, it is necessary to use a certain size of neighborhood pixels for calculation, the larger the neighborhood range, the higher the extraction accuracy, but the corresponding algorithm time complexity O(n 2 )increase

Images