一种基于Harris和图像对比度的角点检测方法

目前各种角点检测方法,对于不同的图像需要选取不同的阈值才能保证结果的精确度,而阈值的选取是比较麻烦的.此处提出了一种通用的方法,通过适当改变图像直方图增大图像对比度,再使用固定阈值的Harris方法对图像中的角点进行提取.该方法解决了阈值选取困难的问题,实验结果表明该方法有效提高了结果的精度,在很多场合比较适合.

A Corner Detection Method Based on Harris and Contrast Information of Image

In each method of corner detection,different thresholds should be chosen for different images in order to ensure that the results are acceptable.However,tt＇s waste of time and difficult to choose a threshold.Here a universal method is proposed to solve this problem.In this method the histogram of an image is changed first,and then Harris corner detection method with a constant threshold is used to detect the corners in this image.This method solved the difficulty of choosing a threshold.And the result of experiments showed that this method can detect the corners more precise,and is quite useful in some fields.