基于多尺度边缘感知和增强的息肉图像分割

doi:10.3778/j.issn.1002-8331.2403-0436

摘要/Abstract

摘要： 针对结直肠图像中息肉尺度差异大、边界不清晰以及内窥镜图片反光等问题，提出了一种基于边缘感知和边界增强的网络模型。以金字塔结构提取多层特征，使用集中引导的边缘感知聚合策略，利用中低层和高层的互补信息来获取第一个轮廓掩码。使用了分层多尺度模块对骨干网络后三层进行特征提取，以适应不同大小息肉特征。提出正逆向综合关注单元，通过局部特征保留和轮廓掩码合并，挖掘出更多轮廓掩码的边缘信息。分别在Kvasir、CVC-ClinicalDB、ETIS、CVC-ColonDB和CVC-300五个流行的息肉分割数据集上进行实验，与目前主流的几种息肉分割方法比较三个指标，其中平均Dice和平均IoU都有所提高，MAE有所降低，性能效果明显优于其他方法。特别是，在ETIS数据集上平均Dice系数达到了0.729 7，比之前最先进的方法提升了0.042 9。

关键词: 息肉分割, 轮廓掩码, 多尺度, 注意力

Abstract: In order to solve the problems of large polyp scale differences, unclear boundaries and reflection of endoscopic images in colorectal images, this paper proposes a network model based on edge perception and boundary enhancement.Firstly, the pyramid structure is used to extract multi-layer features, and the first active contour mask is obtained by using the centrally guided edge-aware aggregation strategy and the complementary information of the middle and low layers and the upper layers. Secondly, the hierarchical multi-scale module is used to extract the features of the last three layers of the backbone network. Finally, the forward and reverse integrated attention unit is proposed, and more edge information of the contour mask is mined through local feature preservation and contour mask merging. Experiments are carried out on five popular polyp segmentation datasets, namely Kvasir, CVC-ClinicalDB, ETIS, CVC-ColonDB and CVC-300, and the three indexes are compared with several mainstream polyp segmentation methods, among which the average Dice coefficient and average intersection union ratio are improved, the average absolute error is reduced, and the performance effect is significantly better than that of other methods. In particular, the average Dice coefficient on the ETIS dataset reaches 0.729 7, an improvement of 0.042 9 over the previous state-of-the-art method.

Key words: polyp segmentation, contour mask, multi-scale, attention

杨瑞君, 陈丽叶, 程燕. 基于多尺度边缘感知和增强的息肉图像分割[J]. 计算机工程与应用, 2025, 61(1): 272-281.

YANG Ruijun, CHEN Liye, CHENG Yan. Polyp Image Segmentation Based on Multi-Scale Edge Perception and Enhancement[J]. Computer Engineering and Applications, 2025, 61(1): 272-281.

参考文献

[1] LONG J, SHELHAMER E, DARRELL T. Fully convolutional networks for semantic segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Los Alamitos: IEEE Computer Society Press, 2015: 3431-3440.
[2] KARKANIS S A, IAKOVIDIS D K, MAROULIS D E, et al. Computer-aided tumor detection in endoscopic video using color wavelet features[J]. IEEE Transactions on Information Technology in Biomedicine, 2003, 7(3): 141-152.
[3] HWANG S, OH J H, TAVANAPONG W, et al. Polyp detection in colonoscopy video using elliptical shape feature[C]//Proceedings of the 2007 IEEE International Conference on Image Processing, 2007: 465-468.
[4] RONNEBERGER O, FISCHER P, BROX T. U-Net: convolutional networks for biomedical image segmentation[C] //Proceedings of the International Conference on Medical Image Computing and Computer-Assisted Intervention. Heidelberg: Springer, 2015: 234-241.
[5] ZHOU Z W, SIDDIQUEE M M R, TAJBAKHSH N, et al. UNet++: a nested U-Net architecture for medical image segmentation[C]//Proceedings of the Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support. Heidelberg: Springer, 2018: 3-11.
[6] JHA D, SMEDSRUD P H, RIEGLER M A, et al. ResUNet++: an advanced architecture for medical image segmentation[C]//Proceedings of the IEEE International Symposium on Multimedia. Los Alamitos: IEEE Computer Society Press, 2019.
[7] FAN D P, JI G P, ZHOU T, et al. PraNet: parallel reverse attention network for polyp segmentation[C]//Proceedings of International Conference on Medical Image Computing and Computer- Assisted Intervention. Heidelberg: Springer, 2020: 263-273.
[8] 时永刚, 李祎, 周治国, 等. 基于阶梯结构的 U-Net 结肠息肉分割算法[J]. 电子与信息学报, 2022, 44(1): 39-47.
SHI Y G, LI Y, ZHOU Z G, et al. Polyp segmentation using stair-structured U-Net[J]. Journal of Electronics & Information Technology, 2022, 44(1): 39-47.
[9] ZHAO H, SHI J, QI X, et al. Pyramid scene parsing network[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017: 2881-2890.
[10] YIN Z, LIANG K, MA Z, et al. Duplex contextual relation network for polyp segmentation[C]//Proceedings of the 2022 IEEE 19th International Symposium on Biomedical Imaging (ISBI), 2022: 1-5.
[11] SRIVASTAVA A, JHA D, CHANDA S, et al. MSRF-Net: a multi-scale residual fusion network for biomedical image segmentation[J]. IEEE Journal of Biomedical and Health Informatics, 2022, 26(5): 2252-2263.
[12] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Advances in Neural Information Processing Systems, 2017.
[13] ZHANG Y, ZHU H, CHENG J, et al. Improving the quality of fetal heart ultrasound imaging with multihead enhanced self-attention and contrastive learning[J]. IEEE Journal of Biomedical and Health Informatics, 2023, 27(11): 5518-5529.
[14] LI Y, WANG J, CHEN S, et al. Enhanced multihead self-attention block network for remote sensing image scene classification[J]. Journal of Applied Remote Sensing, 2023, 17(1): 016517.
[15] LIU Z, LIN Y, CAO Y, et al. Swin transformer: hierarchical vision transformer using shifted windows[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021: 10012-10022.
[16] 崔珂, 田启川, 廉露. 基于U-Net变体的医学图像分割算法综述[J]. 计算机工程与应用, 2024, 60(11): 32-49.
CUI K, TIAN Q C, LIAN L. Review of medical image segmentation algorithms based on U-Net variants[J]. Computer Engineering and Applications, 2024, 60(11): 32-49.
[17] HOU Q, ZHOU D, FENG J. Coordinate attention for efficient mobile network design[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021: 13713-13722.
[18] 郭祥振, 李思潼, 卢锐, 等. 基于多任务联合注意力的结肠息肉分割网络[J]. 计算机工程, 2024, 50(2): 327-336.
GUO X Z, LI S T, LU R, et al. Colon polyp segmentation network based on multi-task joint attention[J]. Computer Engineering, 2024, 50(2): 327-336.
[19] BUI N T, HOANG D H, NGUYEN Q T, et al. MEGANet: multi-scale edge-guided attention network for weak boundary polyp segmentation[C]//Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 2024: 7985-7994.
[20] JHA D, SMEDSRUD P H, RIEGLER M A, et al. Kvasir-SEG: a segmented polyp dataset[C]//Proceedings of the 26th International Conference on MultiMedia Modeling, Daejeon, South Korea, Jan 5-8, 2020. Cham: Springer International Publishing, 2020: 451-462.
[21] BERNAL J, SáNCHEZ F J, FERNáNDEZ-ESPARRACH G, et al. WM-DOVA maps for accurate polyp highlighting in colonoscopy: validation vs. saliency maps from physicians[J]. Computerized Medical Imaging and Graphics, 2015, 43: 99-111.
[22] TAJBAKHSH N, GURUDU S R, LIANG J. Automated polyp detection in colonoscopy videos using shape and context information[J]. IEEE Transactions on Medical Imaging, 2015, 35(2): 630-644.
[23] VáZQUEZ D, BERNAL J, SáNCHEZ F J, et al. A benchmark for endoluminal scene segmentation of colonoscopy images[J]. Journal of Healthcare Engineering, 2017, 2017(1): 4037190.
[24] SILVA J, HISTACE A, ROMAIN O, et al. Toward embedded detection of polyps in WCE images for early diagnosis of colorectal cancer[J]. International Journal of Computer Assisted Radiology and Surgery, 2014, 9: 283-293.
[25] ZHOU L. Spatially exclusive pasting: a general data augmentation for the polyp segmentation[C]//Proceedings of the 2023 International Joint Conference on Neural Networks (IJCNN), 2023: 1-7.
[26] KIM T, LEE H, KIM D. UACANet: uncertainty augmented context attention for polyp segmentation[C]//Proceedings of the 29th ACM International Conference on Multimedia, 2021: 2167-2175.
[27] NARAYAN V, MALL P K, AWASTHI S, et al. FuzzyNet: medical image classification based on GLCM texture feature[C]//Proceedings of the 2023 International Conference on Artificial Intelligence and Smart Communication (AISC), 2023: 769-773.
[28] WANG D, XU C, AN Z, et al. JudgeNet: inverse attention and multiscale fusion networks with boundary determination in polyp segmentation[J]. IEEE Access, 2023, 11: 58279-58293.