Automatic Crack Classification Model based on Convolutional Neural Network and Metaheuristic JAYA Algorithm

Abstract

The main motive of this research is to build an automatic crack classification model in order to enhance the infrastructure system in civil engineering. In order to accomplish this goal, a convolutional neural network is taken into consideration, which takes the crack and non-crack image datasets as input and classifies them. Besides that, the input images are pre-processed using the enhancement and segmentation method in order to enhance the quality and find the region of interest in the images. Furthermore, the characteristics of the input images vary widely. Therefore, the JAYA algorithm is utilised to design the adaptive enhancement and segmentation method, which, according to the objective function, works on the image characteristics. The simulation evaluation is done on the standard dataset images of the crack. Further, various parameters are evaluated based on the confusion matrix in order to validate the performance of the proposed model. Finally, the comparative analysis shows that the proposed model outperforms the existing model.