As the most common road distress, cracks have a substantial influence on the integrity of pavement structures. Accurate identification of crack existence and quantification of crack geometry are thus critical for the decision-making of maintenance measures. This paper proposes a novel neural network for the detection and evaluation of road cracks at pixel level, which combines the advantage of encoding–decoding network and attention mechanism and can thus extract the crack pixels more accurately and efficiently. The proposed network achieves an excellent detection performance with IOU = 92.85%, precision = 96.90%, recall = 95.36%, F1 = 95.53%. Compared with the other advanced networks, the accuracy of the proposed method is substantially enhanced. The quantitative estimation of key geometrical features of cracks including length, width, and area is successfully realized with the development of a prototype of an intelligent mobile system. Compared with the ground truth, the maximum crack width shows the lowest relative error rate, which ranges −31.75%∼28.57%.

中文翻译：

---

一种具有注意机制的新型 U 形编码器-解码器网络，用于像素级道路裂缝的检测和评估

作为最常见的道路问题，裂缝对路面结构的完整性有着重要的影响。因此，准确识别裂缝存在和量化裂缝几何形状对于维护措施的决策至关重要。本文提出了一种新的用于像素级道路裂缝检测和评估的神经网络，它结合了编码-解码网络和注意力机制的优势，从而可以更准确、更有效地提取裂缝像素。所提出的网络实现了出色的检测性能，IOU = 92.85%，精度 = 96.90%，召回率 = 95.36%，F1 = 95.53%。与其他先进网络相比，该方法的准确性大大提高。裂纹关键几何特征的定量估计，包括长度、宽度、并成功实现了智能移动系统原型的开发。与ground truth相比，最大裂缝宽度显示出最低的相对误差率，范围为-31.75%∼28.57%。