Ultrasound Image Segmentation Algorithm of Ocular Muscle of Breeding Pig Based on Continuous Minimum Cut
Yun Liang, Yijin Chen, Hanxing Liu, Yuqing Zhang, Yingjie Kuang · 2022
In the measurement of ocular muscles in live pigs, segmentation methods based on ultrasound images are commonly used. However, due to various interferences in the ocular muscle region and ultrasound images, the segmentation accuracy is low. We proposed a segmentation algorithm for porcine ocular muscle ultrasound images based on continuous minimum cuts. Firstly, the ultrasound image of the ocular muscle is preprocessed to reduce common interference; then, the initial prediction edge is extracted, and the foreground and background templates are roughly constructed; finally, a segmentation algorithm is constructed according to the two templates to obtain the contour edge of the ocular muscle of the breeding pig, so as to achieve accurate measurement. The experimental results show that the segmentation accuracy of the proposed algorithm is significantly better than that of the mainstream ultrasound image segmentation algorithms, and the segmentation results are closer to the manual measurement results.