BAN: A Boundary-Aware Network for Accurate Colorectal Polyp Segmentation

Colonoscopy images exhibit multi-frequency features, with polyp boundaries residing in a mid-frequency range, which are critical for accurate polyp segmentation. However, current deep learning models tend to prioritize low-frequency features, leading to reduced segmentation performance. To address this challenge, we propose a novel <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">boundary-aware network</i> (BAN) that integrates trainable Gabor filters into the polyp segmentation process through a dedicated module called <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Gabor-driven feature extraction</i> (GFE). By developing and using a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">trajectory-directed frequency learning</i> approach, Gabor filters are trained along a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">damping sinusoidal</i> path, dynamically optimizing their frequency parameters within a proper mid-frequency range. This enhances boundary feature representation and significantly improves polyp segmentation accuracy. Extensive experiments demonstrate that our BAN outperforms existing state-of-the-art methods.