Detection Method of Marine Floating Garbage based on Improved Faster R-CNN
Jing Bai, Shujie Yang · Highlights in Science Engineering and Technology · 2022
Traditional marine floating garbage cleaning is implemented manually. Workers have to suffer from the poor environment and intensive tasks, which reduce their efficiency. Neural network can help automate the process, and thus free workers from the heavy labor. This paper proposes a neural network network-based method to improve the accuracy of garbage identification and classification, especially for the small-sized garbage. The method combines the improved Faster R-CNN target detection model and Resnet50. According to the characteristics of marine floating garbage, the parameters of fast r-cnn are determined through repeated experiments, which improves the accuracy of small target detection. Experimental results show that the method provides a comprehensive recognition rate of up to 79,36%.