端到端深度图像分割网络中抑制无效率学习的目标损失函数设计

叶靖雯; 吴晓峰

Chinese

您当前的位置：

首页 >

文章列表页 >

端到端深度图像分割网络中抑制无效率学习的目标损失函数设计

更新时间：2026-09-04

- 端到端深度图像分割网络中抑制无效率学习的目标损失函数设计
- Loss function for ineffective learning reduction in End-to-End deep image segmentation network
- 微电子学与计算机 2019年36卷第9期页码：38-43
- 作者机构：
  
  复旦大学信息科学与工程学院图像与智能信息处理实验室,上海,200433
- 作者简介：
- 基金信息：
  
  上海市卫生计生委智慧医疗专项研究项目 2018ZHYL0204
- DOI：
  中图分类号：
- 纸质出版：2019
- 稿件说明：
移动端阅览
叶靖雯, 吴晓峰. 端到端深度图像分割网络中抑制无效率学习的目标损失函数设计[J]. 微电子学与计算机, 2019, 36(9): 38-43.

YE Jing-wen, WU Xiao-feng. Loss function for ineffective learning reduction in End-to-End deep image segmentation network[J]. 2019, 36(9): 38-43.
叶靖雯, 吴晓峰. 端到端深度图像分割网络中抑制无效率学习的目标损失函数设计[J]. 微电子学与计算机, 2019, 36(9): 38-43. DOI：

YE Jing-wen, WU Xiao-feng. Loss function for ineffective learning reduction in End-to-End deep image segmentation network[J]. 2019, 36(9): 38-43. DOI：

摘要

在端到端深度图像分割网络训练时

常出现前景和背景区域相差巨大的情况

造成目标特征学习不足而背景特征学习过度.对此提出一种基于代价敏感学习的目标函数构造方法:借鉴难例挖掘思想

使用表征难易程度的Focal因子对样本训练误差加权处理

有效抑制无效率学习; 仿人类视觉系统引入感受野因子

兼顾上下文信息.在医学影像数据集上对方法的有效性和可扩展性进行了测试.结果表明

新方法有助于提升网络对于小目标的检出能力

同时分割结果更贴合目标轮廓.

Abstract

In image segmentation tasks based on deep learning methods

it is common that foreground pixels occur significantly more frequently than background pixels

and consequently bias the trained network towards them. In this paper

based on cost-sensitive learning

a design method of loss function for end-to-end image segmentation network is proposed

where two improvements are provided as follows: 1) Inspired by the conception of "hard examples mining"

focal loss is introduced and extended to work for ineffective learning reduction. 2) Inspired by human visual systems

adaptive weights of receptive field are added to further consider the context information. In order to verify the validity and expansibility

the proposed method has been evaluated on severalmedicalimage datasets. The results show that the proposed method can improve the detection performance of the network for small objects

and obtain segmentation results that are more suitable for object contour.

关键词

Keywords

references

RONNEBERGER O, FISCHER P, BROX T. U-Net: convolutional networks for biomedical image segmentation[C]// International Conference on MICCAI. Munich, Germany: Springer, 2015: 234-241.

ENROTH-CUGELL C, ROBSON J G. The contrast sensitivity of retinal ganglion cells of the cat[J]. The Journal of physiology, 1966, 187(3): 517-552. DOI: 517

CHAURASIA A, CULURCIELLO E. Linknet: Exploiting encoder representations for efficient semantic segmentation[C]//Visual Communications and Image Processing 2017. St. Petersburg, FL, USA: IEEE, 2017: 1-4.

CAELLES S, MANINIS K K, PONT-TUSET J, et al. One-shot video object segmentation[C]//Proceedings of the IEEE conference on CVPR. Honolulu, HI, USA: IEEE, 2017: 221-230.

LONG J, SHELHAMER E, DARRELL T. Fully convolutional networks for semantic segmentation[C]//Proceedings of the IEEE conference on CVPR.Boston, MA, USA: IEEE, 2015: 3431-3440.

WU Z, SHEN C, HENGEL A. High-performance semantic segmentation using very deep fully convolutional networks[J]. arXiv preprint arXiv: 1604.04339, 2016.

MILLETARI F, NAVAB N, AHMADI S A. V-net: Fully convolutional neural networks for volumetric medical image segmentation[C]//2016 Fourth International Conference on 3D Vision. Stanford, California, USA: IEEE, 2016: 565-571.

BADRINARAYANAN V, KENDALL A, CIPOLLA R. Segnet: A deep convolutional encoder-decoder architecture for image segmentation[J]. IEEE transactions on pattern analysis and machine intelligence, 2017, 39(12): 2481-2495. DOI: 2481

LIN T Y, GOYAL P, GIRSHICK R, et al. Focal loss for dense object detection[C]//Proceedings of the IEEE international conference on computer vision.Venice, Italy: IEEE, 2017: 2980-2988.

马海荣, 程新文.一种处理非平衡数据集的优化随机森林分类方法[J].微电子学与计算机, 2018, 35(11):28-32. DOI: 28

RODIECK R W, STONE J. Analysis of receptive fields of cat retinal ganglion cells[J]. Journal of Neurophysiology, 1965, 28(5): 833-849. DOI: 833

芮波, 张道强.脑白质损伤分割方法比较和策略改进[J].微电子学与计算机, 2011, 28(10):84-86. DOI: 84

LITJENS G, TOTH R, VAN DE VEN W, et al. Evaluation of prostate segmentation algorithms for MRI: the PROMISE12 challenge[J].Medical image analysis, 2014, 18(2): 359-373. DOI: 359

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

AlN/GaN混合缓冲层双异质结p-GaN HEMT直流特性仿真

大模型发展综述：从BERT到Deepseek的技术演进

基于反向体偏置的高增益稳定性高温运放设计

一种用于极低压boost系统的高效率电荷泵

注意力机制与改进密集卷积引导的信息隐藏模型