Dual-branch crack segmentation network with multi-shape kernel based on convolutional neural network and Mamba

Jianming Zhang, Dianwen Li, Zhigao Zeng, Rui Zhang, Jin Wang

Abstract: Highlights•Propose a hybrid CNN-Mamba dual-branch crack segmentation network (DBCNet).•Construct a Cross-aware Mamba Module (CMM) to form a hybrid CNN-Mamba decoder.•Propose a Feature Fusion Module integrating spatial details and contexts dynamically.•Conduct extensive experiments on three public datasets, and achieve good performances.