DEEPCRACK: LEARNING HIERARCHICAL CONVOLUTIONAL FEATURES FOR CRACK DETECTION

Abstract

Cracks are typical line structures that are of interest in many computer-vision applications. In practice, many cracks, e.g., pavement cracks, show poor continuity and low contrast, which brings great challenges to image-based crack detection by using low-level features. In this paper, we propose DeepCrack - an end-to-end trainable deep convolutional neural network for automatic crack detection by learning high-level features for crack representation. In this method, multi-scale deep convolutional features learned at hierarchical convolutional stages are fused together to capture the line structures. More detailed representations are made in larger-scale feature maps and more holistic representations are made in smaller-scale feature maps. We build DeepCrack net on the encoder-decoder architecture of SegNet, and pairwisely fuse the convolutional features generated in the encoder network and in the decoder network at the same scale. We train DeepCrack net on one crack dataset and evaluate it on three others. The experimental results demonstrate that DeepCrack achieves F-Measure over 0.87 on the three challenging datasets in average and outperforms the current state-of-the-art methods.

Existing System

Seed-growing methods, Canny edge detector, edge detector.

Proposed System

The contributions of this work lie in three-fold: Our main contribution is the design of a new neural network architecture for crack detection. This new network takes full use of the information of the encoder and decoder network, and builds a trainable end-to-end network for crack detection. In the proposed network, a convolutional layer of the encoder network and a convolutional layer of the decoder network at one same scale are fused to compute the training loss at the corresponding scale. The fusion of hierarchical convolutional features is found to be very effective for inferring the cracks out from the image background. Four datasets are constructed for performance evaluation, where one dataset containing 260 pavement images is used for training the network, and three others are used for test. For the three test datasets, two are pavement image datasets and one is stone surface image dataset. The groundtruth cracks are manually labeled by human expert, and the datasets are shared to the community to promote the research of crack detection.

Architecture

BLOCK DIAGRAM