Abstract

Object detection from Synthetic Aperture Radar (SAR) imagery is picking up steam with SAR's all-weather, day and night imaging capabilities. Object detection within SAR images is difficult due to speckle noise, lack of texture, and significant domain shift from optical images for which deep learning models are pretrained. The study proposes to mitigate this with a large dataset and a Multi-Stage Filter Augmentation (MSFA) framework with improved detection performance with diverse backbones and anchor-based assignment methods as suggested by Y. Li et al. The contribution of this work extends this by keeping the MSFA-based pretraining with the highest performing ConvNeXt backbone while adding a change in the anchor box assignment method. Specifically, by using Adaptive Training Sample Selection (ATSS), an anchor-free, statistics-based sample selection method with an existing MSFA-based approach, replacing heuristic-based systems like Faster R-CNN. Experiments show that adding ATSS significantly enhances the detection model's generalizability, particularly in noisy or low-contrast SAR environments. This paper compares the baseline MSFA-based systems with the proposed pipeline using ATSS and demonstrates that ATSS outperforms in detecting small and cluttered objects.

References

• Frolind, P.O., Gustavsson, A., Lundberg, M., Ulander, L.M.: Circular-aperture vhf-band synthetic aperture radar for detection of vehicles in forest concealment. IEEE Transactions on Geoscience and Remote Sensing (2011)
• Braun, A.: Radar satellite imagery for humanitarian response. Bridging the gap between technology and application. Ph.D. thesis, Universität Tübingen (2019)
• Moreira, A., Prats-Iraola, P., Younis, M., Krieger, G., Hajnsek, I., Papathanassiou, K.P.: A tutorial on synthetic aperture radar. IEEE Geoscience and remote sensing magazine (2013)
• Lin, T.Y., Maire, M., Belongie, S.J., Hays, J., Perona, P., Ramanan, D., Dollár, P., Zitnick, C.L.: Microsoft COCO: Common objects in context. In: ECCV (2014)
• Lin, X., Zhang, B., Wu, F., Wang, C., Yang, Y., Chen, H.: Sived: A sar image dataset for vehicle detection based on rotatable bounding box. Remote Sensing (2023)
• Shifeng Zhang, Cheng Chi, Yongqiang Yao, Zhen Lei, and Stan Z. Li. “Bridging the Gap Between Anchor-Based and Anchor-Free Detection via Adaptive Training Sample Selection.” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020, pp. 9759–9768
• Y. Li, X. Li, W. Li, Q. Hou, L. Liu, M.-M. Cheng, and J. Yang, “SARDet-100K: Towards open-source benchmark and toolkit for large- scale SAR object detection,” in Proc. Adv. Neural Inf. Process. Syst. (NeurIPS), 2024
• Lienhart, R., Maydt, J.: An extended set of haar-like features for rapid object detection. In: International Conference on Image Processing (2002)
• Dalal, N., Triggs, B.: Histograms of oriented gradients for human detection. In: CVPR (2005)
• Canny, J.: A computational approach to edge detection. TPAMI (1986)
• Lin, Y.N., Hsieh, T.Y., Huang, J.J., Yang, C.Y., Shen, V.R., Bui, H.H.: Fast iris localisation using haar-like features and adaboost algorithm. Multimedia Tools and Applications (2020)
• Jin, Y., Duan, Y.: Wavelet scattering network-based machine learning for ground penetrating radar imaging: Application in pipeline identification. Remote Sensing (2020)
• Dai, J., Qi, H., Xiong, Y., Li, Y., Zhang, G., Hu, H., Wei,Y.: Deformable convolutional networks. In: ICCV. pp. 764– 773 (2017)
• Besnassi, M., Neggaz, N., Benyettou, A.: Face detection based on evolutionary haar filter. Pattern Analysis and Applications (2020)
• He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 770–778 (2016)
• Jiao, J., Zhang, Y., Sun, H., Yang, X., Gao, X., Hong, W., Fu, K., Sun, X.: A densely connected end-to-end neural network for multiscale and multiscene sar ship detection. IEEE Access (2018)
• Cai, Z., Vasconcelos, N.: Cascade R-CNN: High quality object detection and instance segmentation. TPAMI (2019)
• Lu, X., Li, B., Yue, Y., Li, Q., Yan, J.: Grid r-cnn. In: CVPR. pp. 7363–7372 (2019)
• Chen, Y., Yuan, X., Wu, R., Wang, J., Hou, Q., Cheng, M.M.: YOLO-MS: Rethinking multiscale representation learning for real-time object detection. arXiv (2023)
• Chen, K., Wang, J., Pang, J., Cao, Y., Xiong, Y., Li, X., Sun, S., Feng, W., Liu, Z., Xu, J., et al.: Mmdetection: Open mmlab detection toolbox and benchmark. arXiv (2019)
• Chen, L., Luo, R., Xing, J., Li, Z., Yuan, Z., Cai, X.: Geospatial transformers are what you need for aircraft detection in sar imagery. TGRS (2022)
• Liu, Z., Mao, H., Wu, C.Y., Feichtenhofer, C., Darrell, T., Xie, S.: A convnet for the 2020s. In: CVPR (2022)
• Guo, H., Yang, X., Wang, N., Gao, X.: A centernet++ model for ship detection in sar images. Pattern Recognition (2021)
• Deng, J., Dong, W., Socher, R., Li, L.J., Li, K., Fei-Fei, L.: ImageNet: A large-scale hierarchical image database. In: CVPR (2009)
• Xian, S., Zhirui, W., Yuanrui, S., Wenhui, D., Yue, Z., Kun, F.: Air-sarship-1.0: High-resolution sar ship detection dataset. J. Radars (2019)
• Wei, S., Zeng , X., Qu, Q., Wang, M., Su, H., Shi, J.: Hrsid: A high-resolution sar images dataset for ship detection and instance segmentation. IEEE Access (2020)
• Zhang, T., Zhang, X., Li, J., Xu, X., Wang, B., Zhan, X., Xu, Y., Ke, X., Zeng, T., Su, H., et al.: Sar ship detection dataset (ssdd): Official release and comprehensive data analysis. Remote Sensing (2021)
• Wang, Y., Wang, C., Zhang, H., Dong, Y., Wei, S.: A sar dataset of ship detection for deep learning under complex backgrounds. remote sensing 11(7), 765 (2019)
• Liu, Z., Mao, H., Wu, C.-Y., Feichtenhofer, C., Darrell, T., & Xie, S.: A ConvNet for the 2020s (2022).
• Xia, G.S., Bai, X., Ding, J., Zhu, Z., Belongie, S., Luo, J., Datcu, M., Pelillo, M., Zhang, L.:DOTA: A large-scale dataset for object detection in aerial images. In: CVPR (2018)