Abstract:The traditional canny operator has the problem of being susceptible to noise interference in brain MR image segmentation.
Considering these problems, an improved Canny operator was proposed. Firstly, Gabor filter and Gaussian filter were combined to
filter out artifacts and noise points in brain MR images. Then the edge was determined by non-extreme value suppression. Finally,
according to the Ostu idea, the double threshold was automatically calculated to remove the false edge, and the final segmented brain
MR image was obtained. Experimental results showed that the improved Canny operator could effectively detect the edge of objects,
and was robust to common noise artifacts in brain MR images.
曲蕴慧a,弓明b,廖尹坤b,王鑫b,扬伍连b. 基于改进Canny算子的脑部MR图像分割算法的研究[J]. 中国医疗设备, 2018, 33(11): 37-39.
QU Yunhuia, GONG Mingb, LIAO Yinkunb, WANG Xinb, YANG Wulianb. Brain MR Image Segmentation Algorithm Based on Improved Canny Operator. China Medical Devices, 2018, 33(11): 37-39.