An Algorithm of the Sperm Head Detection Based on the Improved YOLOv5 Network
WANG Tao
WANG Zixin
PENG Tao
WANG Jiacheng
ZHOU Jinhua
Abstract：In the low-magnification imaging of high-concentration sperm samples,the dense and small size of sperm make it difficult to identify their heads,and this lead to low detection accuracy.To improve the accuracy of the sperm head detection,an al-gorithm of the sperm head detection is proposed based on the improved YOLOv5 network.Based on the YOLOv5s network model,the feature fusion block in the network is replaced with BiFPN feature fusion network.The Head structure is modified,it clips the feature detection box for detecting the sperm head to reduce the network size,and to improve the detection and feature fusion ability of small targets,so as to improve the recognition accuracy of small targets.The results show that the improved network improves ac-curacy and recall than the original YOLOv5s network,and achieves good results in local sperm sample detection.
Keywords：sperm samplestarget recognitionYOLOv5scharacteristics of the fusion
Publication Date：2025-06-20
Online Publishing Date：2025-09-23(First online date of this platform, not the publication date of the document)
Pages：7( 1547-1552,1590 )