Detection Method of Conductor Strand Defects Based on Multi-modal Data Fusion

Combined with the advantages of infrared image and visible image, this paper proposes a multi-modal data fusion method for detecting the wire strand defects.The segmentation performance of UNet model in images with poor illumination conditions is improved; it solves the problem that the contrast between the wire and the background in the infrared image is low and it is difficult to distinguish, and reduces the false alarm rate of the loose detection.The experimental results show that this method can be used in UAV patrol images under different lighting conditions, which is more conducive to embedded in UAV for all-weather intelligent patrol, and has the advantages of high recall rate and low false detection rate, compared with the reference method using only infrared or visible light images.