This article presents a technical report on a novel road damage detection algorithm named ALC-Net, specifically designed for processing UAV imagery. The algorithm addresses the challenges of balancing accuracy and real-time performance in detecting road damage such as cracks and potholes. ALC-Net incorporates a lightweight convolutional module that integrates ghost convolution with a squeeze-and-excitation attention mechanism, effectively reducing model parameters while enhancing detection accuracy. The report details the architecture of ALC-Net, which includes a focus module for downsampling and channel-wise concatenation of input images, and a coordinate attention mechanism that aggregates spatial information. The performance of ALC-Net is evaluated on a UAV-captured road damage dataset, demonstrating superior detection capabilities compared to existing methods. Additionally, the article discusses the contributions of key components in ALC-Net through ablation studies and highlights its robust generalization capabilities across non-UAV road damage datasets, indicating its potential for broader applications.