[1]陈亦男,季荣宝,史健婷.GAM-Net:基于U-Net改进的结肠息肉图像分割算法[J].计算机技术与发展,2025,(10):62-70.[doi:10.20165/j.cnki.ISSN1673-629X.2025.0144]
　CHEN Yi-nan,JI Rong-bao,SHI Jian-ting.GAM-Net:An Improved Colon Polyp Image Segmentation Algorithm Based on U-Net[J].,2025,(10):62-70.[doi:10.20165/j.cnki.ISSN1673-629X.2025.0144]
点击复制

GAM-Net:基于U-Net改进的结肠息肉图像分割算法()

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2025年10期
页码:
62-70
栏目:
媒体计算
出版日期:
2025-10-10

文章信息/Info

Title:
GAM-Net:An Improved Colon Polyp Image Segmentation Algorithm Based on U-Net
文章编号:
1673-629X(2025)10-0062-09
作者:
陈亦男; 季荣宝; 史健婷
黑龙江科技大学计算机与信息工程学院,黑龙江哈尔滨 150022
Author(s):
CHEN Yi-nan; JI Rong-bao; SHI Jian-ting
School of Computer and Information Engineering,Heilongjiang University of Science and Technology,Harbin 150022,China
关键词:
深度学习; 息肉图像分割; U-Net; 注意力机制; 特征融合
Keywords:
deep learning; polyp image segmentation; U-Net; attention mechanism; feature fusion
分类号:
TP301
DOI:
10.20165/j.cnki.ISSN1673-629X.2025.0144
摘要:
针对结肠图像中的腺体与息肉之间边界不清晰和息肉本身尺度差异大而导致影响分割精度等问题,提出了一种基于U-Net改进的结肠息肉图像分割算法。首先,将原始图像通过全局通道空间注意力机制(Global Channel Spatial Attention,GCSA)来增强深度卷积神经网络的表征能力。其次,在该模型的下采样后嵌入Assem卷积模块,通过将卷积与Transformer相结合的方法,有效地学习息肉图像的全局和局部特征信息。最后,用多尺度双时特征融合模块(Multi-scale Bitemporal Fusion Module,MBFM)代替传统U-Net通道拼接方式,有效地将编码器部分的低层特征与相应的高层特征进行融合,以优化网络特征的提取来提升分割精度。该算法在CVC-ClinicDB、ETIS-ColonDB两个公开息肉分割数据集上进行实验。通过实验采用MIou、Dice等五项评价指标验证其有效性。以MIou指标和Dice系数为主要评价指标,在CVC-ClinicDB数据集上,MIou指标和Dice系数分别达到了83.49%和89.38%,相较于改进前分别提升了3.01百分点和2.67百分点。在ETIS-ColonDB数据集上,这两个指标分别达到了71.56%和78.48%,分别提升了5.26百分点和2.78百分点。实验结果表明,基于U-Net改进的算法在结肠息肉分割任务上取得了更好的性能。
Abstract:
A colon polyp image segmentation algorithm based on an improved U-Net is proposed to address the issues of unclear boundaries between glands and polyps in colon images and the significant scale differences of polyps that affect segmentation accuracy.Firstly,the original image is enhanced using a global channel spatial attention (GCSA) mechanism to strengthen the representation capability of the deep convolutional neural network. Secondly,an Assem convolution module is embedded after the downsampling stage of the model. This module combines convolution with Transformer methods to effectively learn both global and local feature information of the polyp images. Finally,a multi-scale bitemporal fusion module (MBFM) is used to replace the traditional U-Net channel concatenation method,effectively merging low-level features from the encoder with corresponding high-level features to optimize feature extraction and improve segmentation accuracy. The proposed algorithm is tested on two public polyp segmentation datasets,CVC-ClinicDB and ETIS-ColonDB,and its effectiveness is verified using five evaluation metrics,including MIou and Dice coefficient. Using MIou and Dice as the primary evaluation metrics,on the CVC-ClinicDB dataset,MIou and Dice achieved 83.49% and 89.38%,respectively,showing improvements of 3.01 percentage points and 2.67 percentage points compared to the original method. On the ETIS-ColonDB dataset,these two metrics reached 71.56% and 78.48%,showing improvements of 5.26 percentage points and 2.78 percentage points, respectively. The experimental results demonstrate that the improved U-Net-based algorithm achieves better performance in colon polyp segmentation tasks.

相似文献/References:

[1]陈强锐,谢世朋.基于深度学习的肺部肿瘤检测方法[J].计算机技术与发展,2018,28(04):201.[doi:10.3969/ j. issn.1673-629X.2018.04.043]
　CHEN Qiang-rui,XIE Shi-peng.Lung Cancer Detection Method Based on Deep Learning[J].,2018,28(10):201.[doi:10.3969/ j. issn.1673-629X.2018.04.043]
[2]施泽浩,赵启军.基于全卷积网络的目标检测算法[J].计算机技术与发展,2018,28(05):55.[doi:10.3969/j.issn.1673－629X.2018.05.013]
　SHI Ze-hao,ZHAO Qi-jun.Object Detection Algorithm Based on Fully Convolutional Neural Network[J].,2018,28(10):55.[doi:10.3969/j.issn.1673－629X.2018.05.013]
[3]黄法秀,张世杰,吴志红,等.数据增广下的人脸识别研究[J].计算机技术与发展,2020,30(03):67.[doi:10. 3969 / j. issn. 1673-629X. 2020. 03. 013]
　HUANG Fa-xiu,ZHANG Shi-jie,WU Zhi-hong,et al.Research on Face Recognition Based on Data Augmentation[J].,2020,30(10):67.[doi:10. 3969 / j. issn. 1673-629X. 2020. 03. 013]
[4]陈浩翔,蔡建明,刘铿然,等. 手写数字深度特征学习与识别[J].计算机技术与发展,2016,26(07):19.
　CHEN Hao-xiang,CAI Jian-ming,LIU Keng-ran,et al. Deep Learning and Recognition of Handwritten Numeral Features[J].,2016,26(10):19.
[5]高翔,陈志,岳文静,等.基于视频场景深度学习的人物语义识别模型[J].计算机技术与发展,2018,28(06):53.[doi:10.3969/ j. issn.1673-629X.2018.06.012]
　GAO Xiang,CHEN Zhi,YUE Wen-jing,et al.Human Semantic Recognition Model Based on Video Scene Deep Learning[J].,2018,28(10):53.[doi:10.3969/ j. issn.1673-629X.2018.06.012]
[6]贺飞翔,赵启军. 基于深度学习的头部姿态估计[J].计算机技术与发展,2016,26(11):1.
　HE Fei-xiang,ZHAO Qi-jun. Head Pose Estimation Based on Deep Learning[J].,2016,26(10):1.
[7]徐 融,邱晓晖.一种改进的 YOLO V3 目标检测方法[J].计算机技术与发展,2020,30(07):30.[doi:10. 3969 / j. issn. 1673-629X. 2020. 07. 007]
　XU Rong,QIU Xiao-hui.An Improved YOLO V3 Object Detection[J].,2020,30(10):30.[doi:10. 3969 / j. issn. 1673-629X. 2020. 07. 007]
[8]曾志平[] [],萧海东[],张新鹏[]. 基于DBN的金融时序数据建模与决策[J].计算机技术与发展,2017,27(04):1.
　ZENG Zhi-ping[] [],XIAO Hai-dong[],ZHANG Xin-peng[]. Modeling and Decision-making of Financial Time Series Data with DBN[J].,2017,27(10):1.
[9]李全兵,文 钊*,田艳梅*,等.基于 WGAN 的音频关键词识别研究[J].计算机技术与发展,2021,31(08):26.[doi:10. 3969 / j. issn. 1673-629X. 2021. 08. 005]
　LI Quan-bing,WEN Zhao *,TIAN Yan-mei *,et al.Research on Audio Keywords Recognition Based on WassersteinGenerative Adversarial Network[J].,2021,31(10):26.[doi:10. 3969 / j. issn. 1673-629X. 2021. 08. 005]
[10]李宏林. 分析式纹理合成技术及其在深度学习的应用[J].计算机技术与发展,2017,27(11):7.
　LI Hong-lin. Analyzed Texture-synthesis Techniques and Their Applications in Deep Learning[J].,2017,27(10):7.

更新日期/Last Update: 2025-10-10