SEHC-Net: Network for Image Segmentation Based on Information Compensation and Perceptual Enhancement

doi:10.19678/j.issn.1000-3428.0070004

Abstract

Abstract: Using U-Net as the backbone, a novel medical image segmentation network called SEHC-Net is proposed for medical image segmentation of melanoma. A new structure named Sense and Edge Boost Module (SEBM) is designed specifically to address the challenges in segmenting melanoma images having irregular shapes, diverse sizes, and blurry boundaries. SEBM can expand the receptive fields of features, which enhances the model's ability to extract the target edge information and further capture the connections between pixels. Additionally, a hierarchical compensation module is proposed to solve the problem of information redundancy caused by long connections during information concatenation. This can compensate for the defect that mainstream segmentation networks cannot fully balance spatial contextual information and high-level semantic information in the feature extraction stage. GoogleNet's Inception is used to reduce the parameter increase by reducing the kernel size and increasing the model depth. The segmentation algorithm is verified on the ISIC2018 melanoma dataset. Experimental results show that the Intersection over Union (IoU), sensitivity, precision, Dice coefficient, and accuracy are 79.54%, 86.29%, 90.92%, 84.39%, and 94.83%, respectively. Therefore, the proposed algorithm can effectively improve the melanoma segmentation performance.

Key words: melanoma, Sense and Edge Boost Module (SEBM), Hierarchical Compensation Module (HCM), semantic information, information compensation

摘要： 针对黑色素瘤的医学图像分割,以U-Net为骨干提出一种新的医学图像分割网络SEHC-Net。设计一个感知及边缘增强模块(SEBM)的新结构来处理分割形状不规则、大小多样和边界模糊的黑色素瘤图像。SEBM可以扩大特征的感受野,增强模型提取目标边缘信息和进一步捕捉像素之间联系的能力。此外,提出层级补偿模块(HCM)来解决信息拼接过程中长连接导致的信息冗余问题,以弥补主流分割网络在特征提取阶段不能在空间上下文信息和高级语义信息之间充分平衡的缺陷。同时,为了缓解由于引入以上两种结构导致的参数量增加问题,引入GoogleNet中的Inception思想,减小网络原始的编解码卷积模块中的卷积核尺寸,从而在降低模型参数量的同时增加模型的宽度和深度,并增强捕获像素间关联的能力,提升分割算法的性能。在ISIC2018黑色素瘤数据集上进行验证的结果表明,所提出的分割算法的交并比(IoU)、敏感度、精确率、Dice系数和准确率分别达到了79.54%、86.29%、90.92%、84.39%和94.83%,有效提升了黑色素瘤的分割性能。

关键词: 黑色素瘤, 感知及边缘增强模块, 分层补偿模块, 语义信息, 信息补偿

CLC Number:

TP391

WANG Shaojun, WANG Ting, WANG Chao, YANG Wankou, LU Keyu. SEHC-Net: Network for Image Segmentation Based on Information Compensation and Perceptual Enhancement[J]. Computer Engineering, 2026, 52(2): 177-185.

王少军, 王婷, 王超, 杨万扣, 陆柯宇. 基于信息补偿和感知增强的图像分割网络SEHC-Net[J]. 计算机工程, 2026, 52(2): 177-185.

/ Recommend / Download Citations

URL: https://www.ecice06.com/EN/10.19678/j.issn.1000-3428.0070004

https://www.ecice06.com/EN/Y2026/V52/I2/177

References

[1] NAMBISAN A K, MAURYA A, LAMA N, et al. Improving automatic melanoma diagnosis using deep learning-based segmentation of irregular networks[J]. Cancers(Basel), 2023, 15(4): 1259.
[2] MIRIKHARAJI Z, ABHISHEK K, BISSOTO A, et al. A survey on deep learning for skin lesion segmentation[J]. Medical Image Analysis, 2023, 88: 102863.
[3] LI H, HE X, ZHOU F, et al. Dense deconvolutional network for skin lesion segmentation[J]. IEEE Journal of Biomedical and Health Informatics, 2019, 23(2): 527-537.
[4] 郝宏达, 罗健旭. 基于多尺度区域特征融合的多器官语义分割模型[J]. 计算机工程, 2025, 51(8): 270-280. HAO H D, LUO J X. Multi-organ semantic segmentation model based on multi-scale region feature fusion[J]. Computer Engineering, 2025, 51(8): 270-280. (in Chinese)
[5] 顾群, 随思懿, 王瑞, 等. 基于改进YOLOv8的皮肤黑色素瘤图像分割算法[J/OL]. 计算机工程: 1-11[2024-11-05]. https://doi.org 10.19678/j.issn.1000-3428.0069910. GU Q, SUI S Y, WANG R, et al. Skin melanoma image segmentation algorithm based on improved YOLOv8[J/OL]. Computer Engineering: 1-11[2024-11-05]. https://doi.org 10.19678/j.issn.1000-3428.0069910. (in Chinese)
[6] DING X H, GUO Y C, DING G G, et al. ACNet: strengthening the kernel skeletons for powerful CNN via asymmetric convolution blocks[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. Washington D. C., USA: IEEE Press, 2019: 1911-1920.
[7] RONNEBERGER O, FISCHER P, BROX T. U-Net: convolutional networks for biomedical image segmentation[C]//Proceedings of the 18th International Conference on Medical Image Computing and Computer. Berlin, Germany: Springer, 2015: 234-241.
[8] CHEN L C, PAPANDREOU G, KOKKINOS I, et al. DeepLab: semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 40(4): 834-848.
[9] HU J, SHEN L, SUN G. Squeeze-and-excitation networks[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2018: 7132-7141.
[10] WALTER B. Analysis of convolutional neural network image classifiers in a hierarchical max-pooling model with additional local pooling[J]. Journal of Statistical Planning and Inference, 2023, 224: 109-126.
[11] LIN T Y, DOLLAR P, GIRSHICK R, et al. Feature pyramid networks for object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, USA: IEEE Press, 2017: 2117-2125.
[12] FAN Y, LI J B, ASLAM BHATTI U, et al. A multi-watermarking algorithm for medical images using InceptionV3 and DCT[J]. Computers, Materials & Continua, 2023, 74(1): 1279-1302.
[13] FAIZAL S, RAJPUT C A, TRIPATHI R, et al. Automated cataract disease detection on anterior segment eye images using adaptive thresholding and fine tuned InceptionV3 model[J]. Biomedical Signal Processing and Control, 2023, 82: 104550.
[14] PILLAI R, SHARMA A, SHARMA N, et al. Brain tumor classification using VGG 16, ResNet50, and InceptionV3 transfer learning models[C]//Proceedings of the 2nd IEEE/CVF Conference for Innovation in Technology. Washington D. C., USA: IEEE Press, 2024: 1885-197.
[15] JANDHYALA S S, JALLEDA R R, RAVURI D M. Forest fire classification and detection in aerial images using InceptionV3 and SSD models[C]//Proceedings of the International Conference on Intelligent Data Communication Technologies and Internet of Things. Bengaluru, India:[s. n.], 2023: 320-325.
[16] 贵向泉, 张馨月, 李立. 高分辨率皮肤黑色素瘤图像的两阶段式分割算法[J]. 计算机工程, 2023, 49(11): 267-274. GUI X Q, ZHANG X Y, LI L. Two-stage segmentation algorithm of high resolution skin melanoma image[J]. Computer Engineering, 2023, 49(11): 267-274. (in Chinese)
[17] 赵宏, 王枭. 基于Swin-Transformer的黑色素瘤图像病灶分割研究[J]. 计算机工程, 2024, 50(8): 249-258. ZHAO H, WANG X. Study on lesion segmentation of melanoma images based on Swin-Transformer[J]. Computer Engineering, 2024, 50(8): 249-258. (in Chinese)
[18] 胡帅, 李华玲, 郝德琛. 改进U-Net的多级边缘增强医学图像分割网络[J]. 计算机工程, 2024, 50(4): 286-293. HU S, LI H L, HAO D C. Improved multistage edge-enhanced medical image segmentation network of U-Net[J]. Computer Engineering, 2024, 50(4): 286-293. (in Chinese)
[19] 徐蓬泉, 梁宇翔, 李英. 融合多尺度语义和剩余瓶颈注意力的医学图像分割[J]. 计算机工程, 2023, 49(10): 162-170. XU P Q, LIANG Y X, LI Y. Medical image segmentation fusing multi-scale semantic and residual bottleneck attention[J]. Computer Engineering, 2023, 49(10): 162-170. (in Chinese)
[20] 李萍, 张雪英, 王夙喆, 等. 基于半监督多尺度一致性学习的医学影像分割[J]. 计算机工程, 2025, 51(10): 295-307. LI P, ZHANG X Y, WANG S Z, et al. Medical image segmentation based on semi-supervised multi-scale consistency learning[J]. Computer Engineering, 2025, 51(10): 295-307. (in Chinese)
[21] 郭敏, 张熙涵, 李阳. 融合注意力的教师互一致性半监督医学图像分割[J]. 计算机工程, 2024, 50(9): 313-323. GUO M, ZHANG X H, LI Y. Integrated attentional teacher mutual consistency semi-supervised medical image segmentation[J]. Computer Engineering, 2024, 50(9): 313-323. (in Chinese)
[22] LI X T, ZHAO H L, HAN L, et al. Gated fully fusion for semantic segmentation[J]. Artificial Intelligence, 2020, 34(7): 11418-11425.
[23] ZHOU Z W, RAHMAN SIDDIQUEE M M, TAJBAKHSH N, et al. UNet++: a nested U-Net architecture for medical image segmentation[C]//Proceedings of the 4th IEEE International Workshop on Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support. Berlin, Germany: Springer, 2018: 3-11.
[24] SUN Y, BI F K, GAO Y T, et al. A multi-attention U-Net for semantic segmentation in remote sensing images[J]. Symmetry, 2022, 14(5): 906.
[25] CAO Y, VASSANTACHART A, YE J C, et al. Automatic detection and segmentation of multiple brain metastases on MR images using simultaneous optimized Double U-Net architecture[J]. International Journal of Radiation Oncology, Biology, Physics, 2020, 108(3): S130-S131.
[26] YAQUB M, FENG J C, AHMED S, et al. DeepLabV3, IBCO-based ALCResNet: a fully automated classification, and grading system for brain tumor[J]. Alexandria Engineering Journal, 2023, 76: 609-627.
[27] THOMAS E, PAWAN S J, KUMAR S, et al. Multi-res-Attention U-Net: a CNN model for the segmentation of focal cortical dysplasia lesions from magnetic resonance images[J]. IEEE Journal of Biomedical and Health Informatics, 2021, 25(5): 1724-1734.

Please choose a citation manager

Content to export