Real-world binary crack dataset

该数据集包含了14,000个样本，专门用于检测柔性路面的裂缝问题。该数据集被用于微调六种先进的卷积神经网络模型，以提升裂缝检测的准确性，同时研究数据增强技术对此过程的影响。这一数据集的规模为14,000个样本，其任务是专注于裂缝检测。

This dataset contains 14,000 samples specifically designed for crack detection on flexible pavements. It is used to fine-tune six state-of-the-art Convolutional Neural Networks (CNNs) to improve the accuracy of crack detection, while investigating the impact of data augmentation techniques on this process. With 14,000 samples, this dataset focuses on the task of crack detection.