Search

In an image classification system based on deep learning, a training dataset is a set of labelled images and is often composed of a large number of images. Image labelling tool is usually used to facilitate in creating the training dataset used by the classifier during the learning phase. This paper presents a new image labelling tool called CrackLabel that can automatically label the cracks in the asphalt pavement images. A specially designed image thresholding method called the Global and Lower Quartile Average Intensity (GLQAI) method is utilised. In this study, the training dataset is developed by using real pavement images that resized to 1024×768 resolution. First, crack images are automatically segmented into 768 small patches with 32×32 resolution (pixel). Then, a threshold-based method is applied to automatically segment these patches into two classes which are crack and non-crack patches. The image thresholding method based on the average of global average intensity (GAI) and lower quartile intensity (LQI), namely GLQAI is proposed for this task. Next, the labelling process is performed by assigning patches associated with the crack and background into the crack and non-crack folder, respectively. Finally, the performance of CrackLabel is benchmarked by comparing the results with the manual label crack images by human experts, and three commonly used thresholding methods; Otsu, Kapur and Kittler-Illingworth thresholding. Experimental results show that the proposed thresholding method achieved the best classification rate among various thresholding methods with 94.50%, 93.60% 94.00% and 94.05% for recall, precision, accuracy, and F-score respectively. In conclusion, it is observed that the proposed method using the newly threshold algorithm is very effective in label images into the crack and non-crack patches to maximize the training performance.

Pavement distress refers to the condition of pavement surface in terms of its general appearance. Cracks is a type of pavement distress and commonly occur in a road infrastructure. Crack on a pavement surface shows an early sign of pavement problems and aging. Thus, it is important to detect a serious crack as soon as possible to avoid any road accident that might occur. This study shows a comparison of three popular methods of image segmentation; watershed, k-means clustering and Otsu thresholding for pavement crack detection system in terms of it overall performance. Sample of crack images from three different types of crack such as transverse, longitudinal and crocodile crack are captured manually using digital camera and from online sources. The image is then imported into MATLAB software where it will be compressed but without reducing its quality and pixels intensity. The compressed image is then converted into grayscale to make it easier for analyzing as the system only need to work with one layer instead of three layers (RGB). The contrast of the image is then stretched to increase the level of contrast between the crack and the background. Then, the image will be segmented using three different segmentation method that are mentioned above. Lastly, morphological operation is used to reduce the noise from the image segmented. The result of the segmented image will be analyzed in term of its Structural Similarity Index (SSIM) and Mean Squared Error (MSE). The performance of the system is measure using images with a high level of contrast between the crack and the surface and images with a low level of contrast between the crack and the surface.

Pavement distress particularly cracks, are the most significant type of pavement distress that has been studied for many years due to the complicated pavement crack condition. The continuous severity of crack can cause a dangerous environment that may affect the road users. Therefore, an efficient computer algorithm plays an important role in developing analysis tools for automated crack detection. In Malaysia, many of road surveyors are still employing manual inspection, which is a labour-intensive, error-prone and hazardous task. Various attempts have been made to automate this task by using image processing techniques. However the method turns out to suffer from the problem of lighting variation and complexity of the background such as low contrast on the surrounding pavement that similar to the intensity of crack. This study proposed a deep convolution neural network (CNN) as a detection system of ashpalt pavement crack that capable to detect and classify the pavement crack robustly when dealing with complexity background image. A digital camera is used to capture the image of pavement crack. Then, the captured images are divided into two (2) different grid scales, 32× 32 and 64× 64, and further fed as input to the first deep CNN. For each grid size, the network is trained independently to detect the presence of crack in the image. In the classification stage, the captured images are binarized with the similar grid scales to extract the crack pattern. The binary images containing two types of crack, transverse and longitudinal are then fed as input to the second deep CNN and trained to identify the type of crack. Experimental results show that deep CNN using 32x32 grid scale images provides higher performance for crack detection and classification compared to 64x64. The network achieved the recall, precision and accuracy of 98.0%, 99.4% and 99.2% respectively for crack and non-crack detection, while the performance for transverse and longitudinal achieved the accuracy of 98% and 97%

Asphalt cracks are one of the major road damage problems in civil field as it may potentially threaten the road and highway safety. Crack detection and classification is a challenging task because complicated pavement conditions due to the presence of shadows, oil stains and water spot will result in poor visual and low contrast between cracks and the surrounding pavement. In this paper, the network proposed a fully automated crack detection and classification using deep convolution neural network (DCNN) architecture. First, the image of pavement cracks manually prepared in RGB format with dimension of 1024x768 pixels, captured using NIKON digital camera. Next, the image will segmented into patches (32x32 pixels) as a training dataset from the original pavement cracks and trained DCNN with two different filter sizes: 3x3 and 5x5. The proposed method has successfully detected the presence of crack in the images with 98%, 99% and 99% of recall, precision and accuracy respectively. The network was also able to automatically classify the pavement cracks into no cracks, transverse, longitudinal and alligator with acceptable classification accuracy for both filter sizes. There was no significant different in classification accuracy between the two different filters. However, smaller filter size need more processing training time compared to the larger filter size. Overall, the proposed method has successfully achieved accuracy of 94.5% in classifying different types of crack.

This study presents characterization of cracking in pavement distress using image processing techniques and k-nearest neighbour (kNN) classifier. The proposed semi-automated detection system for characterization on pavement distress anticipated to minimize the human supervision from traditional surveys and reduces cost of maintenance of pavement distress. The system consists of 4 stages which are image acquisition, image processing, feature extraction and classification. Firstly, a tool for image acquisition, consisting of digital camera, camera holder and tripod, is developed to capture images of pavement distress. Secondly, image processing techniques such as image thresholding, median filter, image erosion and image filling are applied. Thirdly, two features that represent the length of pavement cracking in x and y coordinate system namely delta_x and delta_y are computed. Finally, the computed features is fed to a kNN classifier to build its committee and further used to classify the pavement cracking into two types; transverse and longitudinal cracking. The performance of kNN classifier in classifying the type of pavement cracking is also compared with a modified version of kNN called fuzzy kNN classifier. Based on the results from images analysis, the semi-automated image processing system is able to consistently characterize the crack pattern with accuracy up to 90%. The comparison of analysed data with field data shows good agreement in the pavement distress characterization. Thus the encouraging results of semi-automated image analysis system will be useful for developing a more efficient road maintenance process.