Publications by authors named "Shaojiang Dong"
Sensors (Basel)
September 2024
Article Synopsis
- A new method for detecting cracks in underground sewage pipelines utilizes pipeline robots and an enhanced version of the YOLOv8n algorithm to effectively identify issues in both water and sludge environments.
- The implementation of the lightweight RGCSPELAN module and adjustments to the detection head improve feature extraction while reducing the model's parameters to just 1.6 million, which enhances efficiency in real-time detection.
- The results from real-world applications demonstrate that this method is effective at identifying both small and large cracks, potentially improving the safety, detection efficiency, and cost-effectiveness of urban sewage maintenance.
