摘要
提出了一种基于多方向、多尺度Top-hat变换的图像边缘检测方法,应用于阻塞性睡眠呼吸暂停低通气综合症(Obstructive Sleep Apnea Hypopnea Syndrome,OSAHS)早期病理图像的边缘检测及诊断;通过对OSAHS早期病理图像的观察分析,针对待处理图像形状各异,并且含有噪声的特点,构造了不同方向、不同尺度的Top-hat算子增强图像的对比度,利用形态学梯度进行边缘检测,然后把各个算子检测到的图像边缘按照一定的权重进行组合,得到理想的边缘,以便准确地获得病理图像的相关参数,进而实现医学电子诊断;对OSAHS早期病理图像:口腔图像、咽喉声带处图像、鼻道内部图像处理、分析的结果表明,与传统的边缘算子相比较,该方法能使图像的边缘信息更完整、更准确,图像的边缘闭合度可达到97.67%,为实现医学电子诊断打下坚实的基础。
Proposes an image edge detection method based on multi-directional, multi-scale Top-hat operators, and apply the method to the edge detection and diagnosis of OSAHS （Obstructive Sleep Apnea Hypopnea Syndrome, OSAHS） early pathological images. Through the observation and analysis of the OSAHS early pathological images, considering the image Shapesand noise, construct multi-directional, multi-scale Top-hat operators, and they are used to enhance the contrast of images. Use morphological gradient to detect the edge. Then the ideal image edge is obtained by combining the edges of the image detected by each operator according to a certain weight, so that I can obtain relevant parameters of pathology images accurately, and then achieve electronic medical diagnosis. The processing results of OSAHS early pathological images including oral image, vocal cords oral image and inside nose image show that the operator proposed in this paper can make the edge information of the image more complete and accurate, compared with conventional edge operator, and edge closure of image is 97.67%, in the future can provide a solid foundation for electronic medical diagnosis.
出处
《计算机测量与控制》
2016年第2期133-136,共4页
Computer Measurement &Control
基金
黑龙江省教育厅科学技术研究项目(12541165)
