一种X线医学图像的增强及角点检测算法的研究

An Algorithm Study used for Enhancement and Corner-Detection of the X-ray Medical Image

目的:利用MSR算法提高图像质量及角点检测的精度,然后再利用提取的角点来提高图像的配准与拼接速度及精度。方法:首先采用基于人眼感知亮度和色度的视觉模型改进的MSR算法对图像进行增强处理,然后采用改进的harris算法对增强后的图像进行角点提取。结果:采用本文算法对来自COX网站的一幅X线医学图像进行增强、角点检测,处理前后图像的信息熵分别为4.3706、7.3817,增强后的图像质量有明显提高,有效地避免了角点簇拥现象并使角点定位更加精确。结论:实验结果表明,本文采用的算法不仅能有效改善图像质量,而且能更有效、更精确地进行角点检测与提取。

Objective：To improve the image quality and the precision of the corner using the MSR algorithm,then to improve the efficiency and precision of the image registration and stitching using the corner extracted.Methods：Firstly,the modified MSR algorithm,which was proposed in this paper,and is based on the model of human visual perception to brightness and color,was employed to enhance and filter the image,then the corners were extracted from the image processed using the modified harris algorithm.Results：An X-ray image downloaded from the COX web site was processed by our algorithm.The entropy of the untreated and treated image was 4.3706 and 7.3817,respectively,the quality of the enhanced image is improved obviously,and the phenomenon of corner crowding is avoided effectively,and it makes corner location more precisely.Conclusions：Our experimental results show that the quality of the image processed by our algoritnm can be greatly improved and the corners in the image can be extracted effectively and precisely.