计算机科学 ›› 2022, Vol. 49 ›› Issue (11A): 210900070-6.doi: 10.11896/jsjkx.210900070
司绍峰1,2, 张赛强1,2, 李庆2, 陈本瑶3
SI Shao-feng1,2, ZHANG Sai-qiang1,2, LI Qing2, CHEN Ben-yao3
摘要: 近年来,由于鱼眼相机被广泛应用于智能监控领域,不少学者提出了针对鱼眼图像的行人检测算法。然而,鱼眼图像场景复杂且存在畸变,其导致的数据集样本分布和算法监督分配的不均衡问题会降低检测器性能。针对上述问题,首先提出了一种针对鱼眼图像行人检测任务的数据增强方法,该方法由模式采样增强和角度直方图增强两部分组成。其中模式采样增强专注于鱼眼图像难例样本挖掘,生成的新样本丰富了鱼眼图像中心附近的行人模式;角度直方图增强基于直方图均衡的思想,对鱼眼图像样本角度分布做平滑处理,缓解了检测器对单一场景的过拟合问题。此外,基于鱼眼图像Anchor-Free行人检测器,提出将定位质量预测与监督信息融合,将Focal Loss推广到连续域用以优化检测器定位分支的监督分配。实验结果表明,所提数据增强算法能够有效缓解鱼眼图像数据集的分布不均衡,在Anchor-Based和Anchor-Free检测网络上均展现了较好的效果;连续Focal Loss结合定位质量优化监督,在不增加Anchor-Free检测器推理计算复杂度的前提下,将整体性能提升了3.8%。
中图分类号:
[1]KUBO Y,KITAGUCHI T,YAMAGUCHI J.Human tracking using fisheye images[C]//SICE Annual Conference.2007:2013-2017. [2]MEINEL L,FINDEISEN M,HES M,et al.Automated real-time surveillance for ambient assisted living using an omnidirectional camera[C]//IEEE International Conference on Consumer Electronics.IEEE,2014. [3]DEMIRKUS M,WANG L,ESCHEY M,et al.People detection in fish-eye top-views[C]//VISIGRAPP.2017:141-148. [4]SOLICHIN A,HARJOKO A.Deteksi Pejalan Kaki pada Video dengan Metode Fastest Pedestrian Detector in the West(FPDW)[J]Journal of TICOM,2013,2(1):21-37. [5]KRAMS O,KIRYATI N.People detection in top-view fisheye imaging[C]//2017 14th IEEE International Conference on Advanced Video and Signal Based Surveillance(AVSS).IEEE,2017. [6]QIAN Y,MING Y,WANG C,et al.Self-adapting part-based pedestrian detection using a fish-eye camera[C]//2017 IEEE Intelligent Vehicles Symposium(IV).IEEE,2017:779-788. [7]DUAN Z,TEZCAN M O,NAKAMURA H,et al.RAPiD:Rotation-Aware People Detection in Overhead Fisheye Images[J].arXiv:2005.11623,2020. [8]REDMON J,FARHADI A.YOLOv3:An Incremental Improvement[J].arXiv:1804.02767,2018. [9]LIN T Y,GOYAL P,GIRSHICK R,et al.Focal Loss for Dense Object Detection[C]//IEEE Transactions on Pattern Analysis &Machine Intelligence.IEEE,2017:2999-3007. [10]REN S,HE K,GIRSHICK R,et al.Faster R-CNN:Towards Real-Time Object Detection with Region Proposal Networks[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2017,39(6):1137-1149. [11]ZHANG N,DONAHUE J,GIRSHICK R,et al.Part-basedR-CNNs for Fine-grained Category Detection[C]//European Conference on Computer Vision.2014:2700-2709. [12]GIRSHICK R.Fast r-cnn[J].arXiv:1504,08083,2015. [13]HE K,GKIOXARI G,DOLLÁR P,et al.Mask R-CNN[C]//IEEE Transactions on Pattern Analysis&Machine Intelligence.IEEE,2017. [14]REDMON J,DIVVALA S,GIRSHICK R,et al.You Only Look Once:Unified,Real-Time Object Detection[C]//Computer Vision & Pattern Recognition.IEEE,2016. [15]LIU W,ANGUELOV D,ERHAN D,et al.SSD:Single Shot MultiBox Detector[C]//European Conference on Computer Vision.Cham:Springer,2016. [16]LAW H,DENG J.CornerNet:Detecting Objects as Paired Keypoints[J].International Journal of Computer Vision,2020,128(3):642-656. [17]ZHOU X,WANG D,KRHENBÜHL P.Objects as Points[J].arXiv:1904.97850,2019. [18]TIAN Z,SHEN C,CHEN H,et al.FCOS:Fully Convolutional One-Stage Object Detection[C]//2019 IEEE/CVF International Conference on Computer Vision(ICCV).IEEE,2020:516-520. [19]LIN T Y,DOLLAR P,GIRSHICK R,et al.Feature Pyramid Networks for Object Detection[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).2017. [20]LIU Z,ZHENG T,XU G,et al.Training-Time-Friendly Network for Real-Time Object Detection[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2020:11685-11692. [21]SIMONYAN K,ZISSERMAN A.Very Deep Convolutional Networks for Large-Scale Image Recognition[J].arXiv:1409.1556,2014. [22]KRIZHEVSKY A,SUTSKEVER I,HINTON G E.Imagenet classification with deep convolutional neural networks[J].Communications of the ACM,2017,60(6):84-90. [23]DENG J,DONG W,SOCHER R,et al.Imagenet: A large-scale hierarchical image database[C]//2009 IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2009:248-255. [24]BOCHKOVSKIY A,WANG C Y,LIAO H M.YOLOv4:Optimal Speed and Accuracy of Object Detection[J].arXiv:2004.10934,2020. [25]ZHANG H,CISSE M,DAUPHIN Y N,et al.mixup:BeyondEmpirical Risk Minimization[J].arXiv:1710.09412,2017. [26]GOODFELLOW I J,POUGET-ABADIE J,MIRZA M,et al.Gnerative Adversarial Nets[C]//Advances in Neural Information Processing Systems.2014:2672-2680. |
[1] | 王建明, 陈响育, 杨自忠, 史晨阳, 张宇航, 钱正坤. 不同数据增强方法对模型识别精度的影响 Influence of Different Data Augmentation Methods on Model Recognition Accuracy 计算机科学, 2022, 49(6A): 418-423. https://doi.org/10.11896/jsjkx.210700210 |
[2] | 蔡欣雨, 冯翔, 虞慧群. 自适应权重的级联增强节点的宽度学习算法 Adaptive Weight Based Broad Learning Algorithm for Cascaded Enhanced Nodes 计算机科学, 2022, 49(6): 134-141. https://doi.org/10.11896/jsjkx.210500119 |
[3] | 许华杰, 陈育, 杨洋, 秦远卓. 基于混合样本自动数据增强技术的半监督学习方法 Semi-supervised Learning Method Based on Automated Mixed Sample Data Augmentation Techniques 计算机科学, 2022, 49(3): 288-293. https://doi.org/10.11896/jsjkx.210100156 |
[4] | 车爱博, 张辉, 李晨, 王耀南. 基于点云数据的交通环境下单阶段三维目标检测方法 Single-stage 3D Object Detector in Traffic Environment Based on Point Cloud Data 计算机科学, 2022, 49(11A): 210900079-6. https://doi.org/10.11896/jsjkx.210900079 |
[5] | 张洲, 朱俊国, 余正涛. 融合词性与声调特征的越南语语法错误检测 Incorporating Part of Speech and Tonal Features for Vietnamese Grammatical Error Detection 计算机科学, 2022, 49(11): 221-227. https://doi.org/10.11896/jsjkx.210900247 |
[6] | 线岩团, 高凡雅, 相艳, 余正涛, 王剑. 融合多策略数据增强的低资源依存句法分析方法 Improving Low-resource Dependency Parsing Using Multi-strategy Data Augmentation 计算机科学, 2022, 49(1): 73-79. https://doi.org/10.11896/jsjkx.210900036 |
[7] | 暴雨轩, 芦天亮, 杜彦辉, 石达. 基于i_ResNet34模型和数据增强的深度伪造视频检测方法 Deepfake Videos Detection Method Based on i_ResNet34 Model and Data Augmentation 计算机科学, 2021, 48(7): 77-85. https://doi.org/10.11896/jsjkx.210300258 |
[8] | 刘舒康, 唐鹏, 金炜东. 基于智能数据增强和改进YOLOv3算法的接触网吊弦及支架检测研究 Study on Catenary Dropper and Support Detection Based on Intelligent Data Augmentation and Improved YOLOv3 计算机科学, 2020, 47(11A): 178-182. https://doi.org/10.11896/jsjkx.200200053 |
[9] | 胡海根, 孔祥勇, 周乾伟, 管秋, 陈胜勇. 基于深层卷积残差网络集成的黑色素瘤分类方法 Melanoma Classification Method by Integrating Deep Convolutional Residual Network 计算机科学, 2019, 46(5): 247-253. https://doi.org/10.11896/j.issn.1002-137X.2019.05.038 |
[10] | 刘春阳, 吴泽民, 胡磊, 刘熹. 基于似物性和空时协方差特征的行人检测算法 Pedestrian Detection Based on Objectness and Sapce-Time Covariance Features 计算机科学, 2018, 45(6A): 210-214. |
[11] | 邹冲,蔡敦波,刘莹,赵娜,赵彤洲. 组合SVM分类器在行人检测中的研究 Research of Combination SVM Classifier in Pedestrian Detection 计算机科学, 2017, 44(Z6): 188-191. https://doi.org/10.11896/j.issn.1002-137X.2017.6A.043 |
[12] | 刘春阳,吴泽民,胡磊,刘熹. 基于DCT变换的多通道特征的行人检测 Pedestrian Detection Based on DCT of Multi-channel Feature 计算机科学, 2017, 44(Z11): 221-224. https://doi.org/10.11896/j.issn.1002-137X.2017.11A.046 |
[13] | 高琦煜,方虎生. 多卷积特征融合的HOG行人检测算法 HOG Pedestrian Detection Algorithm of Multiple Convolution Feature Fusion 计算机科学, 2017, 44(Z11): 199-201. https://doi.org/10.11896/j.issn.1002-137X.2017.11A.041 |
[14] | 张永良,张智勤,吴鸿韬,董灵平,周冰. 基于改进卷积神经网络的周界入侵检测方法 Perimeter Intrusion Detection Based on Improved Convolution Neural Networks 计算机科学, 2017, 44(3): 182-186. https://doi.org/10.11896/j.issn.1002-137X.2017.03.039 |
[15] | 王坚,兰天. 基于稀疏表达和机器学习的行人检测技术研究 Study on Pedestrian Detection Based on Sparse Representation and Machine Learning 计算机科学, 2016, 43(Z6): 207-209. https://doi.org/10.11896/j.issn.1002-137X.2016.6A.049 |
|