Ying Liu
ABSTRACT
Tire pattern image has always been an important clue in the handling of traffic accident cases and criminal investigation cases. By analyzing the tire pattern information left at the scene of the accident, the investigators can quickly narrow the scope of case detection, and even lock the vehicle and personnel information. Compared with the existing deep learning networks, EfficientNet uses the method of composite scaling to effectively scale the model to obtain the retrieval accuracy under the optimal structure. In this paper, based on the EfficientNet model, we propose a tire pattern image retrieval algorithm based on optimized EfficientNet. The basic network model EfficientNet-B0 is selected to extract the tire pattern image features, and the cross-entropy loss function is used in the model training, In the model optimization, we use AdamW + Cosine annealing, a variant of Adam Gradient Descent method, to further improve the network performance and the retrieval accuracy.
- Xu Zhang, Wei, and Pan, P. Xiao. 2018.In-Vivo Skin Capacitive Image Classification Using AlexNet Convolution Neural Network. In 2018 IEEE 3rd International Conference on Image, Vision and Computing (ICIVC), June 27-29, 2018, Chongqing, China, 5 pages. https://doi.org/10.1109/ICIVC.2018.8492860.Google Scholar
Cross Ref
- Jie Hu, Li Shen and Guang Sun. 2018.Squeeze-and-Excitation Networks. In IEEE Conference on Computer Vision and Pattern Recognition(CVPR), June 18 -23, 2018, Salt Lake City, UT, USA, 10 pages. https://doi.org/10.1109/CVPR.2018.00745Google ScholarCross Ref
- Yanping Huang, Youlong Cheng, Ankur Bapna, Orhan Firat, Dehao Chen, Mia Xu Chen, HyoukJoong Lee, Jiquan Ngiam, Quoc V. Le, and Yonghui Wu. 2019. GPipe: Efficient Training of Giant Neural Networks Using Pipeline Parallelism. In IEEE Conference on Computer Vision and Pattern Recognition, June 27-30, 2016, Las Vegas, NV, USA,9 pages.Google Scholar
- Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun. 2016.Deep Residual Learning for Image Recognition. In IEEE Conference on Computer Vision and Pattern Recognition, June 27-30, 2016, Las Vegas, NV, USA,9 pages. https://doi.org/10.1109/CVPR.2016.90.Google ScholarCross Ref
- Zagoruyko, S., Nikos K..2016.Wide Residual Networks. In British Machine Vision Conference 2016.York,UK, https://doi.org/10.5244/C.30.87Google Scholar
- Howard, Andrew G., Menglong Zhu, Bo Chen, Dmitry K., Weijun Wang, Tobias W., Marco A., and Hartwig A.. 2017. MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications. ArXiv Preprint ArXiv:1704.04861.Google Scholar
- Mingxing Tan, and Quoc V. Le. 2019. EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. In International Conference on Machine Learning, July 26-28, 2019, Rome, Italy , 10 pages.Google Scholar
- Krizhevsky A., Sutskever I., Hinton G. E. 2012. Imagenet Classification with Deep Convolutional Neural Networks. Advances in neural information processing systems, 25, 1097-1105, 2019.Google Scholar
- Ozcan, S., and Mustacoglu, A.. 2018. Transfer Learning Effects on Image Steganalysis with Pre-Trained Deep Residual Neural Network Model. In 2018 IEEE International Conference on Big Data (Big Data),Dec 10-13,2018, Seattle, WA, USA , 8 pages. https://doi.org/10.1109/BigData.2018.8622437.Google Scholar
- Yaqi, L., Jing Jiang, Kun Zhang, Yilun Hua, and Miao Cheng. 2019. Factorizing and Reconstituting Large-Kernel MBConv for Lightweight Face Recognition. In 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW), Oct 27-28,2019, Seoul, Korea (South),9 pages. https://doi.org/10.1109/ICCVW.2019.00329.Google Scholar
- Tingyun, H., Yungchang Chang, Hsinhung Chou, and Chingte, C.. 2019. Filter-Based Deep-Compression with Global Average Pooling for Convolutional Networks. Journal of Systems Architecture 95(2019 February), 9–18. https://doi.org/10.1016/j.sysarc.2019.02.008.Google ScholarDigital Library
- Mannor S., Peleg D., Rubinstein R.. 2005. The cross entropy method for classification. In Proceedings of the 22nd international conference on Machine learning. 561-568, 2005, Bonn, Germany. https://doi.org/10.1145/1102351.1102422.Google ScholarDigital Library
- Ying Xue. 2019 An overview of overfitting and its solutions. In Journal of Physics: Conference Series. IOP Publishing, 1168(2): 022022. 2019. https://doi.org/10.1088/1742-6596/1168/2/022022.Google Scholar
- Kangrui Wang, and Xiaobing Yu.2021. MobileNet and EfficientNet Demonstration on Google Landmark Recognition Dataset. International Core Journal of Engineering(3), https://doi.org/10.6919/ICJE.202103_7(3).0043.Google Scholar
- Henderson P., Ferrari V.. 2016. End-to-end training of object class detectors for mean average precision. In Asian Conference on Computer Vision. 198-213, 2016, Springer, Cham. https://doi.org/10.1007/978-3-319-54193-8_13.Google Scholar
- Guanghai Liu, Jingyu Yang, and Zouyong Li.2015. Content-based image retrieval using computational visual attention model. pattern recognition, 48,8,(February 2015), 2554-2566. https://doi.org/10.1016/j.patcog.2015.02.005Google ScholarDigital Library
- J.Z. Wang, Jia Li, and Wiederhold G., 2001.SIMPLIcity: Semantics-sensitive integrated matching for picture libraries. IEEE Transactions on pattern analysis and machine intelligence, 2001, 23,9(2001 September), 947-963. https://doi.org/10.1109/34.955109Google ScholarDigital Library
Index Terms
- Tire Pattern Image Retrieval Algorithm Based on Optimized Efficientnet
Recommendations
Method of tire pattern image retrieval based on wavelet transform and Siamese network
ICASIT 2020: Proceedings of the 2020 International Conference on Aviation Safety and Information TechnologyImage retrieval of tire tread pattern plays a vital role in traffic accidents and criminal investigation. We propose a tire pattern retrieval method based on wavelet transform and Siamese network, which combine traditional techniques with deep learning. ...
Sketch Based Image Retrieval with Adversarial Network
ICVIP '19: Proceedings of the 3rd International Conference on Video and Image ProcessingSketch retrieval is a specific cross-domain retrieval task. The core of sketch retrieval is to learn a common feature subspace, where the features of sketches and natural images can be both discriminative and domain-invariant. However, similarity ...
Image Retrieval Using Fused Deep Convolutional Features
This paper proposes an image retrieval using fused deep convolutional features to solve the semantic gap between low-level features and high-level semantic features of traditional contend-based image retrieval method. Firstly, the improved network ...
Comments