抄録
G-004
A Study of Automated Fetal Head Detection by Pre-processing based on Ultrasound Image Gray Feature and Iterative Randomized Hough Transform
◎Rong Xu・Jun Ohya・Bo Zhang（Waseda University）・Yoshinobu Sato（Osaka University）・Masakatsu G. Fujie（Waseda University）
This paper discusses an automated detection of fetal head by pre-processing based on ultrasound image gray feature and iterative randomized hough transform. First, fetal ultrasound images are segmented by k-means clustering method, and the fetal skull is skeletonized by distance transform based on ultrasound image gray feature. Then, the ellipse of fetal head in skeleton image will be detected automatically by iterative randomized hough transform. Experimental results demonstrated that the proposed method is effective for automatic fetal head detection in clinical ultrasound images.