Abstract: Segmenting objects from cluttered backgrounds in single-channel images, such as marine radar echoes, medical images, and remote sensing images, poses significant challenges due to limited ...

Abstract: This paper introduces an enhanced U-Net network designed to extract fine crack features against complex backgrounds and balance crack pixel categories for bridge crack segmentation.

Worst case improvement: Bright/colourful images improved from 17-18 dB (baseline) to 23+ dB through targeted preprocessing augmentation — a +6 dB gain on previously failing cases.

Triton is a language and compiler for writing highly efficient custom deep-learning primitives. Not officially supported on Windows, but a fork provides pre-built wheels. 3.6.x RTX 50xx (Blackwell), ...