PSOU-Net: A Neural Network Based on Improved Particle Swarm Optimization for Breast Ultrasound Image Segmentation
Chaoyi Chen, Bo Xu, Ying Nian Wu · 2021 2nd International Conference on Artificial Intelligence and Computer Engineering (ICAICE) · 2021
In order to accurately segment the lesions in breast ultrasound images with high noise, low contrast and poor uniformity, this paper proposes a neural network optimized by Particle Swarm Optimization algorithm. Based on the structure of U-Net network, the dynamic edge detection error function is designed, and then an improved PSO algorithm is proposed to optimize the error function to obtain the improved U-Net algorithm (PSOU-Net). By comparing the PSOU-Net network proposed in this paper with the traditional U-Net neural network on Dataset BUSI dataset, the experimental results show that the Precision value of PSOU-Net network is improved by 16.5% and the$F_{1}$score is improved by 6.3%, which effectively improves the segmentation performance of the network.