Tiny-Crack-Net: A multiscale feature fusion network with attention mechanisms for segmentation of tiny cracks

Convolutional neural networks (CNNs) have gained growing interest in recent years for their advantages in detecting cracks on concrete bridge components. Class imbalance is a fundamental problem in crack segmentation, resulting in unsatisfactory segmentation for tiny cracks. Besides, limited by the local receptive field, CNNs often cannot integrate local features with global dependencies, thus significantly affecting the detection accuracy of tiny cracks across the entire image. To solve those problems in segmenting tiny cracks, a multiscale feature fusion network with attention mechanisms named Tiny-Crack-Net (TCN) is proposed. The modified residual network was used to capture the local features of tiny cracks. The dual attention module was then incorporated into the architecture to better separate the tiny cracks from the background. Also, a multiscale fusion operation was implemented to preserve the edge details of tiny cracks. Finally, a joint learning loss of the cross-entropy and similarity was proposed to alleviate the poor convergence induced by the severe class imbalance of the pixels representing tiny cracks. The capability of the network in segmenting tiny cracks was remarkably enhanced by the aforementioned arrangements, and the Tiny-Crack-Net achieved a Dice similarity coefficient of 87.96% on an open-source data set, which was at least 5.84% higher than those of the six cutting-edge networks. The effectiveness and robustness of the Tiny-Crack-Net were validated with field test results, which showed that the intersection over union (IOU) for cracks with a width of 0.05 mm or wider reaches 91.44%.

Automated summary

This study proposes a multiscale feature fusion network named Tiny-Crack-Net (TCN) to address the class imbalance and limited local receptive field issues in crack segmentation. By modifying the residual network, incorporating dual attention modules, and implementing multiscale fusion operations, the network's ability to segment tiny cracks is significantly enhanced. Evaluation on an open-source dataset and field test results validate the effectiveness and robustness of Tiny-Crack-Net.