针对宫颈异常细胞检测的SER-DC YOLO
SER-DC YOLO for the Detection of Abnormal Cervical Cells
李超炜 1杨晓娜 1赵司琦 1何勇军2
作者信息
- 1. 哈尔滨理工大学 计算机科学与技术学院,哈尔滨 150080
- 2. 哈尔滨工业大学 计算机科学与技术学院,哈尔滨 150006
- 折叠
摘要
由于宫颈细胞样本的液基薄层细胞学检测(thin prep cytologic test,TCT)图像内容复杂,背景颜色丰富多样,而且不同女性的宫颈细胞具有一定程度的天然差异,这给宫颈异常细胞的检测带来了很大的困难.为解决这一难题,提出了一种名为基于特征压缩与激发和可变形卷积(SE-ResNet-deformable convolution you only look once,SER-DC YOLO)的目标检测网络.该网络在YOLOv5 的Backbone中融合注意力机制,添加了SE-ResNet模块,然后改进了SPP层的网络结构,并且使用可变形卷积来替换普通卷积,最后修改了边界框的损失计算函数,将广义交并比(generalized intersection over union,GIoU)改为α-IOU Loss.实验表明,该网络与YOLOv5 网络相比,在宫颈图片数据集上召回率提高了 19.94%,精度提高了 3.52%,平均精度均值提高了 7.19%.相关代码链接:https://github.com/sleepLion99/SER-DC_YOLO.
Abstract
Due to the complex content of Thin Prep Cytology Test(TCT)images of cervical cell samples with rich and diverse background colors and a certain degree of natural variation of cervical cells among different women,this poses a great difficulty in the detection of abnormal cervical cells.To solve this challenge,a target detection network called SE-ResNet-Deformable Convolution You Only Look Once(SER-DC YOLO)is proposed.The network incorporates the attention mechanism in YOLOv5s Backbone,adds the SE-ResNet module,then improves the network structure of the SPP layer and replaces the normal convolution with deformable convolution,and finally modifies the loss calculation function of the bounding box by replacing the Generalized Intersection over Union(GIoU)to α-IOU Loss.Experiments show that the network improves recall by19.94%,precision by3.52%,and average precision by 7.19%on the cervical image dataset compared with the YOLOv5 network.Link to related code:https://github.com/sleepLion99/SER-DC_YOLO.
关键词
SER-DC/YOLO/YOLOv5/目标检测/注意力机制/可变形卷积Key words
SER-DC YOLO/YOLOv5/target detection/attention mechanism/deformable convolution引用本文复制引用
基金项目
国家自然科学基金面上项目(61673142)
黑龙江省自然科学基金杰出青年基金(JJ2019JQ0013)
哈尔滨市杰出青年人才基金(2017RAYXJ013)
哈尔滨理工大学杰出青年人才项目(20200203)
出版年
2024
NETL
NSTL
万方数据