计算机科学 ›› 2023, Vol. 50 ›› Issue (4): 96-102.doi: 10.11896/jsjkx.220300054
白雪飞1, 靳智超1, 王文剑1,2, 马亚楠1
BAI Xuefei1, JIN Zhichao1, WANG Wenjian1,2, MA Yanan1
摘要： 皮肤病变形状、颜色、大小类型多样,给皮肤病变的准确分割带来了巨大挑战。针对这一问题,提出了一种融合边缘增强与多尺度注意力的皮肤病变分割网络(BEMA U-Net)。该网络包含一个用于提取全局特征的空间多尺度注意力模块和一个用于增强病变区域边缘特征的边缘增强模块,将两种模块添加到以编码解码结构为主干的网络(U-Net)中,能够有效抑制病变图像中背景噪声的干扰并强化病灶的边缘细节。此外,设计了混合损失函数,结合Dice Loss和Boundary Loss,并在训练过程中实现混合损失函数的动态权重调整,使网络对病变图像整体特征和边缘细节特征的提取进行多重监督,缓解了皮肤病变图像分割中毛发干扰和边缘模糊的问题。在ISIC2017和ISIC2018两个公开数据集上的实验结果表明,所提算法对皮肤病变部位的分割图像边缘连续、轮廓清晰,具有更好的分割效果。
中图分类号:
| [1]JENKINS R W,FISHER D E.Treatment of advance melanoma in 2020 and beyond treatment in review[J].Journal of Investigative Dermatology,2020,141(1):23-31. [2]MEMON A,BANNISTER P,ROGERS I,et al.Changing epidemiology and age-specific incidence of cutaneous malignant melanoma in england:an analysis of the national cancer registration data by age,gender and anatomical site,1981－2018[J].The Lancet Regional Health-Europe,2021,2:10024-10036. [3]VIALE P H.The American cancer society’s facts & figures:2020 edition[J].Journal of the Advanced Practitioner in Oncology,2020,11(2):135-136. [4]PERUCH F,BOGO F,BONAZZA M,et al.Simpler,faster,more accurate melanocytic lesion segmentation through meds[J].IEEE Transactions on Biomedical Engineering,2014,61(2):557-565. [5]CELEBI M,IVATOMI H,SCHAEFER G,et al.Lesion border detection in dermoscopy images[J].Computerized Medical Imaging and Graphics,2009,33(2):148-153. [6]WONG A,SCHARCANSKI J,FIEGUTH P.Automatic skin lesion segmentation via iterative stochastic region merging[J].IEEE Transactions on Information Technology in Biomedicine,2011,15(6):929-936. [7]LONG J,SHELHAMER E,DARRELL T.et al.Fully convolutional networks for semantic segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,39(4):640-651. [8]YUAN Y D,CHAO M,LO Y C.Automatic skin lesion segmen-tation using deep fully convolutional networks with jaccard distance[J].IEEE Transactions on Medical Imaging,2017,36(9):1876-1886. [9]JIANG Y,CAO S M,TAO S X,et al.Skin lesion segmentation based on multi-scale attention convolutional neural network[J].IEEE Access,2020,8:122811-122825. [10]RONNEBERGER O,FISCHER P,BROX T.U-net:convolu-tional networks for biomedical image segmentation[C]//Medical Image Computing and Computer-Assisted Intervention(MICCAI 2015).Cham:Springer,2015:234-241. [11]SARKER M,RASHWAN H,AKRAM F,et al.SLSDeep:skin lesion segmentation based on dilated residual and pyramid pooling networks[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention(MICCIA 2018).Cham:Springer,2018:21-29. [12]AZAD R,ASADI-AGHBOLAGHI M,FATHY M,et al.Bi-directional convlstm u-net with densley connected convolutions[C]//2019 IEEE/CVF International Conference on Computer Vision Workshop(ICCVW).Piscataway:IEEE,2019:406-415. [13]ASADI-AGHBOLAGHI M,AZAD R,FATHY M,et al.Multi-level context gating of embedded collective knowledge for medical image segmentation[J].arXiv:2003.05056,2020. [14]TONG X,WEI J,SUN B,et al.ASCU-Net:attention gate,spatial and channel attention u-net for skin lesion segmentation[J].Diagnostics,2021,11(3):501. [15]JHA D,RIEGLER M A,JOHANSEN D,et al.DoubleU-Net:a deep convolutional neural network for medical image segmentation[C]//2020 IEEE 33rd International Symposium on Compu-ter-Based Medical Systems(CBMS).Piscataway:IEEE,2020:558-564. [16]ALI A,LI J,O’SHEA S,et al.A deep learning based approach to skin lesion border extraction with a novel edge detector in dermoscopy images[C]//The 2019 International Joint Confe-rence on Neural Networks(IJCNN).Piscataway:IEEE,2019:1-7. [17]CHAKKARAVARTHY A,CHANDRASEKAR A.Anatomicalregion segmentation method from dermoscopy images of pigmented skin lesions[J].International Journal of Imaging Systems and Technology,2020,30(3):636-652. [18]IOFFE S,SZEGEDY C.Batch normalization:accelerating deep network training by reducing internal covariate shift[C]//The 32nd International Conference on machine Learning.SAN DIEGO:JMLR,2015:448-456. [19]QAMAR S,AHMAD P,SHEN L.Dense encoder-decoder-based architecture for skin lesion segmentation[J].Cognitive Computation,2021,13(2):583-594. [20]QIAO Z,CUI Z,NIU X,et al.Image segmentation with pyramid dilated convolution based on resnet and u-net[C]//24th International Conference on Neural Information Processing(ICO-NIP).Cham:Springer,2017:364-372. [21]ZHANG L,GAO H J,ZHANG J,et al.Optimization of the convolutional neural networks for automatic detection of skin can-cer[J].Open Medicine,2020,15(1):27-37. [22]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(3DV).Piscataway:IEEE,2016:565-571. [23]BOKHOVKIN A,BURNAEV E.Boundary loss for remotesensing imagery semantic segmentation[C]//International Symposium on Neural Networks(ISNN).Cham:Springer,2019:388-401. [24]SIMONYAN K,ZISSERMAN A.Very deep convolutional networks for large-scale image recognition[C]//International Conference on Learning Representations(ICLR 2015).ICLR,2015:1-14. [25]JIA D,WEI D,SOCHER R,et al.ImageNet:a large-scale hierarchical image database[C]//2009 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).Piscataway:IEEE,2009:248-255. [26]CODELLA N,GUTMAN D,CELEBI M E,et al.Skin lesion analysis toward melanoma detection: a challenge at the 2017 international symposium on biomedical imaging (ISBI),hosted by the international skin imaging collaboration (ISIC)[C]//15th IEEE International Symposium on Biomedical Imaging (ISBI).Piscataway:IEEE,2018:168-172. [27]CODELLA N,ROTEMBERG V,TSCHANDL P,et al.Skin lesion analysis toward melanoma detection 2018:a challenge hosted by the international skin imaging collaboration (ISIC)[J].arXiv:1902.03368,2019. [28]ABADI M,BARHAM P,CHEN J,et al.TensorFlow:a system for large -scale machine learning[C]//12th USENIX Sympo-sium on Operating Systems Design and Implementation(OSDI).San Diego:USENIX ASSOC,2016:265-283. [29]TAQI A M,AWAD A,AL-AZZO F.The impact of multi-optimizers and data augmentation on tensorflow convolutional neural network performance[C]//IEEE Conference on Multimedia Information Processing & Retrieval(MIPR).Piscataway:IEEE,2018:140-145. |
| [1] | 白雪飞, 马亚楠, 王文剑. 基于特征融合的边缘引导乳腺超声图像分割方法 Segmentation Method of Edge-guided Breast Ultrasound Images Based on Feature Fusion 计算机科学, 2023, 50(3): 199-207. https://doi.org/10.11896/jsjkx.211200294 |
| [2] | 赵凯, 安卫超, 张晓宇, 王彬, 张杉, 相洁. 共享浅层参数多任务学习的脑出血图像分割与分类 Intracerebral Hemorrhage Image Segmentation and Classification Based on Multi-taskLearning of Shared Shallow Parameters 计算机科学, 2022, 49(4): 203-208. https://doi.org/10.11896/jsjkx.201000153 |
| [3] | 高心悦, 田汉民. 基于改进U-Net网络的液滴分割方法 Droplet Segmentation Method Based on Improved U-Net Network 计算机科学, 2022, 49(4): 227-232. https://doi.org/10.11896/jsjkx.210300193 |
| [4] | 赫晓慧, 邱芳冰, 程淅杰, 田智慧, 周广胜. 基于边缘特征融合的高分影像建筑物目标检测 High-resolution Image Building Target Detection Based on Edge Feature Fusion 计算机科学, 2021, 48(9): 140-145. https://doi.org/10.11896/jsjkx.200800002 |
| [5] | 张晓宇, 王彬, 安卫超, 阎婷, 相洁. 基于融合损失函数的3D U-Net++脑胶质瘤分割网络 Glioma Segmentation Network Based on 3D U-Net＋+ with Fusion Loss Function 计算机科学, 2021, 48(9): 187-193. https://doi.org/10.11896/jsjkx.200800099 |
| [6] | 瞿中, 谢钇. 基于全U网络的混凝土路面裂缝检测算法 Concrete Pavement Crack Detection Algorithm Based on Full U-net 计算机科学, 2021, 48(4): 187-191. https://doi.org/10.11896/jsjkx.200100113 |
| [7] | 李华基, 程江华, 刘通, 程榜, 赵康成. 基于U-net++网络的弱光图像增强方法 Low-light Image Enhancement Method Based on U-net++ Network 计算机科学, 2021, 48(11A): 278-282. https://doi.org/10.11896/jsjkx.210300111 |
| [8] | 王鑫, 张昊宇, 凌诚. 基于U-Net优化的SAR遥感图像语义分割 Semantic Segmentation of SAR Remote Sensing Image Based on U-Net Optimization 计算机科学, 2021, 48(11A): 376-381. https://doi.org/10.11896/jsjkx.210300260 |
| [9] | 彭磊, 张辉. 基于U-net的道路缺陷检测 U-net for Pavement Crack Detection 计算机科学, 2021, 48(11A): 616-619. https://doi.org/10.11896/jsjkx.201200059 |
| [10] | 杨春德, 贾竹, 李欣蔚. 基于U-Net++的心电信号识别分类研究 Study on ECG Signal Recognition and Classification Based on U-Net++ 计算机科学, 2021, 48(10): 121-126. https://doi.org/10.11896/jsjkx.200700103 |
| [11] | 康雁, 陈铁, 李浩, 杨兵, 张亚钏, 卜荣景. 融合多视图注意力机制的城市交通流量补全 Urban Traffic Flow Completion with Multi-view Attention Mechanism 计算机科学, 2021, 48(10): 177-184. https://doi.org/10.11896/jsjkx.200800077 |
| [12] | 唐一星, 刘学亮, 胡社教. 多方向分区网络结构的行人再识别 Multi-orientation Partitioned Network for Person Re-identification 计算机科学, 2021, 48(10): 204-211. https://doi.org/10.11896/jsjkx.210300128 |
| [13] | 高强, 高敬阳, 赵地. GNNI U-net：基于组归一化与最近邻插值的MRI左心室轮廓精准分割网络 GNNI U-net:Precise Segmentation Neural Network of Left Ventricular Contours for MRI Images Based on Group Normalization and Nearest Interpolation 计算机科学, 2020, 47(8): 213-220. https://doi.org/10.11896/jsjkx.190600026 |
| [14] | 胡浩,郁滨,沈刚. 像素不扩展视觉密码的边缘增强研究 Edge Enhancement on Size Invariant Visual Cryptography 计算机科学, 2015, 42(2): 103-107. https://doi.org/10.11896/j.issn.1002-137X.2015.02.022 |
| [15] | 韩彦芳，杨娜，缪艳，徐伯庆. 颜色特征加权滤波骨髓细胞分割方法 Bone Marrow Cell Segmentation Based on Color Feature Weighted Filter 计算机科学, 2011, 38(7): 283-286. |
|
||
首页