This assignment is on image classification problem using deep convolutional neural networks. The dataset consists of images obtained from concrete bridge decks, pavements, and walls. The images can contain either cracks or no cracks. Your objective in this challenge is to develop a binary image classification model that can classify from a given image if it contains any cracks or not.
Acknowledgements
This dataset is adapted from SDNET 2018 dataset.
Dorafshan, S., Thomas, R. J., & Maguire, M. (2018). SDNET2018: An annotated image dataset for non-contact concrete crack detection using deep convolutional neural networks. Data in Brief, 21, 1664–1668. https://doi.org/10.1016/j.dib.2018.11.015