Search

Unmanned Aerial Vehicle (UAV) inspection has gradually replaced manual inspection of transmission tower, which produces large quantity of images. While it is laborious and time-consuming to manually analyze these images, there are also challenges in automatically detecting small-size defective components such as bolts in transmission tower images, due to problems including complex background, small size, and many similar objects of bolts. In this paper, we propose a deep neural network named Camp-Net (Context Information and Multi-Scale Pyramid Network) to identify bolts defect in transmission tower images. First, multi-scale feature fusion combines deep features and shallow features in convolutional networks to detect small-size bolts. Second, context information fusion puts the information around bolts into the detection network to remove the disturbance of complex background and similar objects. Experimental results on a dataset with over 30,000 field-collected images containing both defective bolts and normal bolts show the proposed model can accurately identify bolts with loose pins and bolts without pins among fittings in transmission tower. The Average Precision (AP) of defective bolts detection of this model can be 11.4% higher than that of Faster R-CNN, the commonly used high performance model.