计算机科学 ›› 2024, Vol. 51 ›› Issue (11A): 231100122-7.doi: 10.11896/jsjkx.231100122
王喆1, 赵慧俊2,3, 谭超1, 李骏1, 申冲2,3
WANG Zhe1, ZHAO Huijun2,3, TAN Chao1, LI Jun1, SHEN Chong2,3
摘要： 自动检测脑部肿瘤在磁共振成像中的位置是一个复杂、繁重的任务,需要耗费大量时间和资源。传统识别方案经常出现误解、遗漏和误导的情况,从而影响患者的治疗进度,对患者的生命安全产生影响。为了进一步提高鉴定的效果,引入了4项关键改进措施。首先,采用了高效的多尺度注意力EMA(Efficient Multi-scale Attention),这种方法既可以对全局信息进行编码,也可以对信息进行重新校准,同时通过并行的分支输出特征进行跨维度的交互,使信息进一步聚合。其次,引入了BiFPN(Bidirectional Feature Pyramid Network)模块,并对其结构进行改进,以便缩短每一次检测所需要的时间,同时提升图像识别效果。然后采用MDPIoU损失函数和Mish激活函数进行改进,进一步提高检测的准确度。最后进行仿真实验,实验结果表明,改进的YOLOv8算法在脑癌检测中的精确率、召回率、平均精度均值均有提升,其中Precision提高了4.48%,Recall提高了2.64%,mAP@0.5提高了2.6%,mAP@0.5:0.9提高了7.0%。
中图分类号:
| [1]HE C E ,XU H J,WANG Z,et al.Automatic Segmentation Algorithm for Multimodal Magnetic Resonance-Based Brain Tumor Images[J].Acta Optica Sinica,2020,40(6):0610001. [2]MOINUL I,TANZIM M R,MOHAMMED K,et al.Effectiveness of Federated Learning and CNN Ensemble Architectures for Identifying Brain Tumors Using MRI Images[J].Neural processing letters,2022,55(4):3379-3809. [3]Abd El KADER,ISSELMOU,et al.Differential deep convolutional neural network model for brain tumor classification[J]. Brain Sciences,2021,11(3):352. [4]KANG M,TING C M,TING F F,et al.RCS-YOLO:A Fast and High-Accuracy Object Detector for Brain Tumor Detection[J].arXiv:2307.16412,2023. [5]HOSSAIN T,SHISHIR F S,ASHRAF M,et al.Brain tumordetection using convolutional neural network[C]//1st international Conference on Advances in Science,Engineering and Robotics Technology(ICASERT 2019).IEEE,2019. [6]VANKDOTHU R,HAMEED M A,FATIMA H.A braintumor identification and classification using deep learning based on CNN-LSTM method[J].Computers and Electrical Engineering,2022,101:107960. [7]TIWARI P,PANT B,ELARABAWY M M,et al.Cnn basedmulticlass brain tumor detection using medical imaging[J].Computational Intelligence and Neuroscience,2022. [8]MONTAHA S,AZAM S,RAFID A K M R H,et al.Timedistributed-cnn-lstm:A hybrid approach combining cnn and lstm to classify brain tumor on 3d mri scans performing ablation study[J].IEEE Access,2022,10:60039-60059. [9]ARG S,ALMEIDA D,LYMAN K.Resnet in resnet:Generalizing residual architectures[J].arXiv:1603.08029,2016. [10]SZEGEDY C,VANHOUCKE V,IOFFE S,et al.Rethinking the inception architecture for computer vision[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2016:2818-2826. [11]ZHU Y,NEWSAM S.Densenet for dense flow[C]//2017 IEEE International Conference on Image Processing(ICIP).IEEE,2017:790-794. [12]SINHA D,EL-SHARKAWY M.Thin mobilenet:An enhancedmobilenet architecture[C]//2019 IEEE 10th Annual Ubiquitous Computing,Electronics & Mobile Communication Conference(UEMCON).IEEE,2019:0280-0285. [13]OUYANG D,HE S,ZHANG G,et al.Efficient Multi-Scale Attention Module with Cross-Spatial Learning[C]//ICASSP 2023-2023 IEEE International Conference on Acoustics,Speech and Signal Processing(ICASSP).IEEE,2023:1-5. [14]TAN M,PANG R,LE Q V.Efficientdet:Scalable and efficient object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2020:10781-10790. [15]SILIANG M,YONG X.MPDIoU:A Loss for Efficient and Accurate Bounding Box Regression[J].arXiv:2307.07662,2023. [16]MISRA D.Mish:A self regularized non-monotonic activationfunction[J].arXiv:1908.08681,2019. [17]ABOAH A,WANG B,BAGCI U,et al.Real-time multi-classhelmet violation detection using few-shot data sampling technique and yolov8[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2023:5349-5357. [18]ZHANG Y,GUO Z,WU J,et al.Real-time vehicle detectionbased on improved yolo v5[J].Sustainability,2022,14(19):12274. [19]WANG Y,WANG H,XIN Z.Efficient detection model of steel strip surface defects based on YOLO-V7[J].IEEE Access,2022,10:133936-133944. [20]XUE J,CHENG F,LI Y,et al.Detection of farmland obstacles based on an improved YOLOv5s algorithm by using CIoU and anchor box scale clustering[J].Sensors,2022,22(5):1790. |
|
||
首页