近年來，深度學習技術在影像處理和分析領域中取得了顯著的進展，尤其在物件偵測方面，取得了顯著的成果。然而，經過訓練的物件偵測神經網路的性能在很大程度上取決於影像品質，如何提高偵測的精度仍然是一個具有挑戰性的問題。影像品質可能受到多種因素的影響，例如圖像模糊、噪音、低對比度、光照不足或過度曝光等。這些問題可能導致深度學習模型難以準確地識別和定位物件。為了克服這些挑戰，本研究提出了一種方法，使用深度強化學習 DDPG 以及 YOLOv7 來實現增強物件偵測的效果。在這種方法中，我們使用前處理技術對影像進行飽和度、亮度、對比度和銳利度處理，以提高影像品質。這些前處理步驟由 DDPG 算法負責執行，以進一步增強影像。接著，我們將經過前處理的影像作為 YOLOv7 的輸入，以達到更好的物件偵測效果。我們的實驗結果表明，相較於僅使用單獨的 YOLOv7，這種新的物件偵測方法取得更好的偵測精度。

In recent years, deep learning techniques have made significant progress in the field of image processing and analysis, especially in the area of object detection. However, the performance of trained object detection neural networks depends largely on the image quality, and it is still a challenging problem to improve the detection accuracy. Image quality can be affected by a variety of factors, such as blurred images, noise, low contrast, insufficient lighting, or overexposure. These problems may make it difficult for deep learning models to accurately identify and localize objects. To overcome these challenges, this study proposes a method to achieve enhanced object detection using deep reinforcement learning DDPG as well as YOLOv7. In this approach, we use preprocessing techniques for image saturation, brightness, contrast, and sharpness to improve image quality. These preprocessing steps are performed by the DDPG algorithm to further enhance the image. Then, we use the pre-processed images as input to YOLOv7 to achieve better object detection. Our experimental results show that this new object detection method achieves better detection accuracy than using only YOLOv7 alone.

[1] Morgane Ayle et al. “Bar—a reinforcement learning agent for bounding-box automated refinement”. Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 34. 03. 2020, pp. 2561–2568.
[2] Míriam Bellver Bueno et al. “Hierarchical object detection with deep reinforcement learning”. Deep Learning for Image Processing Applications 31.164 (2017), p. 3.
[3] Juan C Caicedo and Svetlana Lazebnik. “Active object localization with deep reinforcement learning”. Proceedings of the IEEE international conference on computer vision. 2015, pp. 2488–2496.
[4] Navneet Dalal and Bill Triggs. “Histograms of oriented gradients for human detection”. 2005 IEEE computer society conference on computer vision and pattern recognition (CVPR’05). Vol. 1. Ieee. 2005, pp. 886–893.
[5] Pedro F Felzenszwalb et al. “Object detection with discriminatively trained part based models”. IEEE transactions on pattern analysis and machine intelligence 32.9 (2009), pp. 1627–1645.
[6] Ross Girshick. “Fast r-cnn”. Proceedings of the IEEE international conference on computer vision. 2015, pp. 1440–1448.
[7] Ross Girshick et al. “Rich feature hierarchies for accurate object detection and semantic segmentation”. Proceedings of the IEEE conference on computer vision and pattern recognition. 2014, pp. 580–587.
[8] Zequn Jie et al. “Tree-structured reinforcement learning for sequential object localization”. Advances in neural information processing systems 29 (2016).
[9] Timothy P Lillicrap et al. “Continuous control with deep reinforcement learning”. arXiv preprint arXiv:1509.02971 (2015).
[10] Tsung-Yi Lin et al. “Microsoft coco: Common objects in context”. Computer Vision--ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part V 13. Springer. 2014, pp. 740–755.
[11] Siddharth Nayak and Balaraman Ravindran. “Reinforcement learning for improving object detection”. Computer Vision–ECCV 2020 Workshops: Glasgow, UK, August 23–28, 2020, Proceedings, Part V 16. Springer. 2020, pp. 149–161.
[12] Joseph Redmon et al. “You only look once: Unified, real-time object detection”. Proceedings of the IEEE conference on computer vision and pattern recognition. 2016, pp. 779–788.
[13] Shaoqing Ren et al. “Faster r-cnn: Towards real-time object detection with region proposal networks”. Advances in neural information processing systems 28 (2015).
[14] Burak Uzkent, Christopher Yeh, and Stefano Ermon. “Efficient object detection in large images using deep reinforcement learning”. Proceedings of the IEEE/CVF winter conference on applications of computer vision. 2020, pp. 1824–1833.
[15] Chien-Yao Wang, Alexey Bochkovskiy, and Hong-Yuan Mark Liao. “YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors”. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2023, pp. 7464–7475.
[16] Yan Wang et al. “Multitask learning for object localization with deep reinforcement learning”. IEEE Transactions on Cognitive and Developmental Systems 11.4 (2018), pp. 573–580.