De-Zhi Sun

Improved deep convolutional algorithm for intelligent maintenance and detection of cables in complex background environments

To enhance the accuracy, robustness, and operational efficiency of cable anomaly recognition and defect localization within complex background environments, this study introduces a multi-scale residual deep convolutional neural network. This model integrates multi-scale convolution to improve feature representation for...