Adaptive whale optimization-based convolutional neural network for aircraft composite crack classification

Cracks in aircraft structures can lead to catastrophic failures. Engineers rely on precise crack classification to determine damage severity and ensure passenger safety. This study presents a novel adaptive whale optimization based convolutional neural network (AWOA-CNN) to categorize composite structure cracks. The composite material cracks are identified by the CNN model, and the model parameters are fine-tuned using the AWOA. Electronic Viewfinder (EVF) and Scanning Electron Microscope (SEM) machines capture aircraft material images that are considered for analysis. A convolutional neural network processes the images with fine-tuning parameters by the AWOA to enhance crack detection accuracy and reduce the time complexity. The proposed method achieves accuracy, precision, and recall of 98.56, 98.50 and 97.76% on the EVF dataset and 95.18, 97.18 and 96.50% on the SEM dataset, respectively.