ASTA detects and analyzes astronomical trails in images using deep learning and image processing techniques. It applies a pre-trained segmentation model to astronomical FITS images to identify trail pixels, then converts detected trail positions from pixel to sky coordinates (right ascension and declination). The software can save results as CSV tables summarizing detected trails and as mask images (binary and predicted probability masks) for visualization and further analysis. Command-line options allow users to adjust the segmentation threshold, control which outputs are written, choose output directories, and report processing times for each step.