簡易檢索 / 詳目顯示

回結果列表
研究生： 黃瀞儀
Huang, Ching-Yi
論文名稱： 基於顏色獨立特徵提取的拜耳原始影像無影像信號處理物件偵測
Color-Independent Feature Extraction for ISP-less Object Detection on Bayer Raw Images
指導教授： 蔡家齊
Tsai, Chia-Chi
學位類別： 碩士
Master
系所名稱： 電機資訊學院 - 電機工程學系
Department of Electrical Engineering
論文出版年： 2025
畢業學年度： 113
語文別： 英文
論文頁數： 96
中文關鍵詞： 拜耳原始圖像 、物件偵測 、圖像增強 、機器學習 、電腦視覺 、影像訊號處理器
外文關鍵詞： Bayer raw image, object detection, image enhancement, machine learning, computer vision, ISP
相關次數： 點閱：45  下載：0 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報
    • 摘要 i ABSTRACT iii 致謝 v Content vi List of Tables ix List of Figures x Chapter 1 Introduction 1 1.1 Motivation 1 1.2 Thesis Contribution 2 1.3 Thesis Organization and Structure 3 Chapter 2 Background and Related Work 5 2.1 Object Detection 5 2.2 Bayer Raw Image 7 2.2.1 Introduction to Bayer Raw Image 7 2.2.2 Image Signal Processing (ISP) 9 2.2.3 Inverse ISP 14 2.3 Image Enhancement 15 2.4 Object Detection Using Raw Image Data 17 Chapter 3 Problem Analysis and Design Methodology 23 3.1 Simulated Raw Dataset 23 3.2 Problem Analysis of Bayer Raw Images 24 3.3 Proposed Method 31 3.3.1 System Architecture 31 3.3.2 Low-light Enhancement Module 32 3.3.3 Independent Color Processing Module 39 3.3.4 System Components Co-optimization 43 3.4 Loss Function 43 Chapter 4 Experimental Evaluation and Results 45 4.1 Experiment Setup 45 4.1.1 Dataset 45 4.1.2 Data Augmentation 46 4.1.3 Detectors 48 4.1.4 Evaluation Metrics 50 4.2 Experimental Results 54 4.2.1 Experimental results of different preprocessing methods for Bayer raw images 54 4.2.2 Experimental results of our method based on YOLOv3-custom 57 4.2.3 Comparison of increased FLOPs and our method on YOLOv3-custom 59 4.2.4 Experimental results of our method based on YOLOv4-tiny 62 4.2.5 Comparison of increased FLOPs and our method on YOLOv4-tiny 63 4.2.6 Experimental results of our method based on YOLOX 66 4.2.7 Comparison of increased FLOPs and our method on YOLOX 66 4.3 Ablation Study 69 4.3.1 Ablation study of SCI learning module and ICP module 69 4.3.2 Ablation study of space-to-depth and demosaicing 70 4.4 Comparison with Other Papers 71 Chapter 5 Conclusion and Future Work 75 5.1 Conclusion 75 5.2 Future Work 76 References 78

    [1] R. Girshick, J. Donahue, T. Darrell, and J. Malik, "Region-based convolutional networks for accurate object detection and segmentation," IEEE transactions on pattern analysis and machine intelligence, vol. 38, no. 1, pp. 142-158, 2015.
    [2] R. Girshick, "Fast r-cnn," arXiv preprint arXiv:1504.08083, 2015.
    [3] S. Ren, K. He, R. Girshick, and J. Sun, "Faster R-CNN: Towards real-time object detection with region proposal networks," IEEE transactions on pattern analysis and machine intelligence, vol. 39, no. 6, pp. 1137-1149, 2016.
    [4] P. Sun et al., "Sparse r-cnn: End-to-end object detection with learnable proposals," in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2021, pp. 14454-14463.
    [5] J. Redmon, "Yolov3: An incremental improvement," arXiv preprint arXiv:1804.02767, 2018.
    [6] A. Bochkovskiy, C.-Y. Wang, and H.-Y. M. Liao, "Yolov4: Optimal speed and accuracy of object detection," arXiv preprint arXiv:2004.10934, 2020.
    [7] C.-Y. Wang, A. Bochkovskiy, and H.-Y. M. Liao, "YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors," in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2023, pp. 7464-7475.
    [8] W. Liu et al., "Ssd: Single shot multibox detector," in Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11–14, 2016, Proceedings, Part I 14, 2016: Springer, pp. 21-37.
    [9] T. Lin, "Focal Loss for Dense Object Detection," arXiv preprint arXiv:1708.02002, 2017.
    [10] H. Law and J. Deng, "Cornernet: Detecting objects as paired keypoints," in Proceedings of the European conference on computer vision (ECCV), 2018, pp. 734-750.
    [11] X. Zhou, D. Wang, and P. Krähenbühl, "Objects as points," arXiv preprint arXiv:1904.07850, 2019.
    [12] J. Redmon and A. Farhadi, "YOLO9000: better, faster, stronger," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2017, pp. 7263-7271.
    [13] Z. Ge, "Yolox: Exceeding yolo series in 2021," arXiv preprint arXiv:2107.08430, 2021.
    [14] G. Song, Y. Liu, and X. Wang, "Revisiting the sibling head in object detector," in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2020, pp. 11563-11572.
    [15] Y. Wu et al., "Rethinking classification and localization for object detection," in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2020, pp. 10186-10195.
    [16] Wikipedia. "Bayer filter." https://en.wikipedia.org/wiki/Bayer_filter (accessed 2024).
    [17] I. Morawski, Y.-A. Chen, Y.-S. Lin, S. Dangi, K. He, and W. H. Hsu, "Genisp: Neural isp for low-light machine cognition," in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022, pp. 630-639.
    [18] Z. Liang, J. Cai, Z. Cao, and L. Zhang, "Cameranet: A two-stage framework for effective camera isp learning," IEEE Transactions on Image Processing, vol. 30, pp. 2248-2262, 2021.
    [19] H. Qin et al., "Attention-aware learning for hyperparameter prediction in image processing pipelines," in European Conference on Computer Vision, 2022: Springer, pp. 271-287.
    [20] L. Liu, X. Jia, J. Liu, and Q. Tian, "Joint demosaicing and denoising with self guidance," in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020, pp. 2240-2249.
    [21] X. Wu, Y. Wu, J. Zhang, F. Zhang, and J. Ren, "Joint Demosaicing and Denoising with Gradient Guidance in Quad Bayer CFA," presented at the 2023 IEEE International Conference on Image Processing (ICIP), 2023.
    [22] T. Kerepecký, F. Šroubek, A. Novozámský, and J. Flusser, "NERD: Neural Field-Based Demosaicking," presented at the 2023 IEEE International Conference on Image Processing (ICIP), 2023.
    [23] D. S. Tan, W. Y. Chen, and K. L. Hua, "DeepDemosaicking: Adaptive Image Demosaicking via Multiple Deep Fully Convolutional Networks," IEEE Trans Image Process, Feb 7 2018, doi: 10.1109/TIP.2018.2803341.
    [24] H. S. Malvar, L.-w. He, and R. Cutler, "High-quality linear interpolation for demosaicing of Bayer-patterned color images," in 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing, 2004, vol. 3: IEEE, pp. iii-485.
    [25] E. S. Lubana, R. P. Dick, V. Aggarwal, and P. M. Pradhan, "Minimalistic image signal processing for deep learning applications," in 2019 IEEE International Conference on Image Processing (ICIP), 2019: IEEE, pp. 4165-4169.
    [26] C. Chen, Q. Chen, J. Xu, and V. Koltun, "Learning to see in the dark," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2018, pp. 3291-3300.
    [27] X. Dong et al., "Abandoning the Bayer-Filter to See in the Dark," presented at the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022.
    [28] Z. Zhou, Z. Shi, and W. Ren, "Linear Contrast Enhancement Network for Low-Illumination Image Enhancement," IEEE Transactions on Instrumentation and Measurement, vol. 72, pp. 1-16, 2023, doi: 10.1109/tim.2022.3232641.
    [29] J. Li, X. Feng, and Z. Hua, "Low-light image enhancement via progressive-recursive network," IEEE Transactions on Circuits and Systems for Video Technology, vol. 31, no. 11, pp. 4227-4240, 2021.
    [30] C. Guo et al., "Zero-reference deep curve estimation for low-light image enhancement," in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2020, pp. 1780-1789.
    [31] C. Li, C. Guo, and C. C. Loy, "Learning to enhance low-light image via zero-reference deep curve estimation," IEEE transactions on pattern analysis and machine intelligence, vol. 44, no. 8, pp. 4225-4238, 2021.
    [32] Z. Zhao, B. Xiong, L. Wang, Q. Ou, L. Yu, and F. Kuang, "RetinexDIP: A unified deep framework for low-light image enhancement," IEEE Transactions on Circuits and Systems for Video Technology, vol. 32, no. 3, pp. 1076-1088, 2021.
    [33] X. Gao, K. Zhao, L. Han, and J. Luo, "BézierCE: Low-Light Image Enhancement via Zero-Reference Bézier Curve Estimation," Sensors, vol. 23, no. 23, p. 9593, 2023.
    [34] L. Ma, T. Ma, R. Liu, X. Fan, and Z. Luo, "Toward fast, flexible, and robust low-light image enhancement," in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2022, pp. 5637-5646.
    [35] J. Cahill et al., "Exploring the Viability of Bypassing the Image Signal Processor for CNN-Based Object Detection in Autonomous Vehicles," IEEE Access, vol. 11, pp. 42302-42313, 2023, doi: 10.1109/access.2023.3270710.
    [36] P. H. Chan, C. Wei, A. Huggett, and V. Donzella, "Raw camera data object detectors: an optimisation for automotive processing and transmission," Authorea Preprints, 2023.
    [37] Y. Hong, K. Wei, L. Chen, and Y. Fu, "Crafting Object Detection in Very Low Light," in BMVC, 2021, vol. 1, no. 2, p. 3.
    [38] T. Brooks, B. Mildenhall, T. Xue, J. Chen, D. Sharlet, and J. T. Barron, "Unprocessing images for learned raw denoising," in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2019, pp. 11036-11045.
    [39] K. Wei, Y. Fu, J. Yang, and H. Huang, "A physics-based noise formation model for extreme low-light raw denoising," in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020, pp. 2758-2767.
    [40] W. Ljungbergh, J. Johnander, C. Petersson, and M. Felsberg, "Raw or cooked? object detection on raw images," in Scandinavian Conference on Image Analysis, 2023: Springer, pp. 374-385.
    [41] R. Xu et al., "Toward raw object detection: A new benchmark and a new model," in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 13384-13393.
    [42] S. Yang, P. Luo, C.-C. Loy, and X. Tang, "Wider face: A face detection benchmark," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 5525-5533.
    [43] A. G. Howard, "Mobilenets: Efficient convolutional neural networks for mobile vision applications," arXiv preprint arXiv:1704.04861, 2017.

    無法下載圖示 校內：2030-03-13公開
    校外：2030-03-13公開
    電子論文尚未授權公開，紙本請查館藏目錄
    QR CODE