Summary of Cycle-yolo: a Efficient and Robust Framework For Pavement Damage Detection, by Zhengji Li et al.
Cycle-YOLO: A Efficient and Robust Framework for Pavement Damage Detection
by Zhengji Li, Xi Xiao, Jiacheng Xie, Yuxiao Fan, Wentao Wang, Gang Chen, Liqiang Zhang, Tianyang Wang
First submitted to arxiv on: 28 May 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper presents an enhanced method for detecting pavement damage using CycleGAN and improved YOLOv5 algorithm. The authors first augment their initial dataset of 7644 pavement damage images using CycleGAN to improve the diversity of the data. They then propose a data enhancement method combining an improved Scharr filter, CycleGAN, and Laplacian pyramid to better handle complex backgrounds and multiscale targets. To further enhance the target recognition effect, they introduce the convolutional block attention module attention mechanism and atrous spatial pyramid pooling with squeeze-and-excitation structure. The optimized loss function of YOLOv5 replaces CIoU with EIoU for improved performance. Experimental results show a precision of 0.872, recall of 0.854, and mean average precision@0.5 of 0.882 in detecting three main types of pavement damage. The method achieves real-time detection on GPU with 68 frames per second, exceeding the performance of YOLOv7. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper is about a new way to quickly and accurately detect problems with roads (like cracks or potholes). This matters because as more cars drive, roads get worn out faster. The researchers used two special tools: CycleGAN and an improved version of YOLOv5. They made their dataset bigger by using CycleGAN, which makes the images look more like real road pictures. Then they added a new filter to make the computer better at recognizing objects on complex backgrounds. They also changed how the algorithm loses to make it work better. The results show that this method is really good, with high accuracy and speed. It can even do better than some other advanced methods. |
Keywords
» Artificial intelligence » Attention » Loss function » Mean average precision » Precision » Recall
