计算机工程与应用2024，Vol.60Issue(19)：242-249,8.DOI:10.3778/j.issn.1002-8331.2307-0076
考虑数据分布损失的图像分割
Design of Loss Function Considering Data Distribution
摘要
Abstract
The choice of loss function in image segmentation tasks directly impacts the convergence process and final ac-curacy of the models.Cross-entropy loss(CEL)exhibits stable convergence but achieves lower accuracy when dealing with imbalanced data distributions.Dice loss(DL),which calculates based on region overlap,achieves higher accuracy in handling imbalanced data but faces difficulties when applied to multi-class datasets.To address these issues,a modified loss function called Dice cross-entropy loss(DCEL)is proposed.DCEL computes the loss value using cross-entropy for positive samples and the product of cross-entropy and intersection over union(IOU)for negative samples.This design en-ables DCEL to have gradients positively correlated with errors on multi-class datasets,facilitating convergence.Further-more,by compressing the loss value of negative samples,DCEL enhances the focus on positive samples,thereby improv-ing the accuracy of image segmentation algorithms.The performance of DCEL is validated on ADE20k,PASCAL VOC,LoveDA,and HRF datasets.关键词
损失函数/图像分割/不平衡数据/深度学习Key words
loss function/image segmentation/imbalanced data/deep learning分类
信息技术与安全科学引用本文复制引用
张震,彭景昊,田鸿朋..考虑数据分布损失的图像分割[J].计算机工程与应用,2024,60(19):242-249,8.基金项目
河南省重大公益专项(201300311200). （201300311200）
