Structural cracks in concrete have a significant influence on structural safety, so it is necessary to detect and monitor concrete cracks. Deep learning is a powerful tool for detecting cracks in concrete structures. However, it requires a large quantity of training samples and is costly in terms of computational time. In order to solve these difficulties, a deep learning target detection framework combining texture features with concrete crack data is proposed. Texture features and pre-processed concrete data are merged to increase the number of feature channels in order to reduce the demand of training samples for the model and improve training speed. With this framework, concrete crack detection can be realized even with a limited number of samples. To accomplish this aim, self-made steel fiber reinforced concrete crack data is used for comparison between our framework and those without texture feature mergence or pre-processed concrete data. The experimental results show that the number of parameters that need to be fitted in the model training and training time can be correspondingly reduced and the detection accuracy can also be improved.

In view of the low contrast between the edge of concrete cracks and the background, this paper proposes a R-Unet target detection framework combined with edge detection operator. Firstly, the Roberts operator is used to extract the target edge feature (3-band), which is fused with the original data to form a 6-band image as input data. Secondly, based on the Unet model, the residual module is introduced into the encoding path to reduce the network overfitting, and the attention module is added in the decoding path to enhance the learning ability of the pixels in the crack area and improve the accuracy of concrete crack extraction; a combination of the cross entropy loss function and Dice loss function is used for evaluation. The results of experiments on the self-made concrete crack datasets show that R-Unet target detection framework is significantly better than the original Unet model and other relevant models. The overall extraction accuracy reached 95.75% with the recall rate of 94.33%, and the MIou of 90.67%. The R-Unet target detection framework proposed can greatly reduce the crack edge false detection rate and improve the target detection accuracy.

Concrete crack identification methods based on machine learning can greatly improve extraction efficiency and precision. However, in many cases, model training requires a large amount of sample data, and insufficient data makes it difficult to effectively obtain model parameters. This study introduces a deep learning framework that integrates filters, principal component analysis, and attention mechanisms suitable for small sample sizes. Firstly, the histogram equalization method is used for the raw images, which can effectively enhance image contrast. Then, to acquire effective images of the crack, different methods are employed for crack detection, which are subsequently handled by principal component analysis (PCA) for optimal feature choice. Att-Unet and Att-Mask R-cnn segmentation models are used to design the detection for concrete cracks. To raise the learning ability of the segmentation models, an attention mechanism is applied to each feature layer of the decoder, and the loss function is evaluated using a combination of the Focal function and Cross Entropy. To verify the effectiveness of the proposed method, Deep Crack datasets and 76 sets of concrete crack data were collected for testing. Experimental results have shown that the method proposed can significantly reduce the model’s demand for data volume and improve training speed, which provides a new direction for small-sample crack extraction.