An Enhanced Spider Wasp Optimised Convolutional Neural Network for Image Recognition System in Library

Abstract

Libraries play a vital role in fostering educational, cultural, research, and recreational growth within society, their capacity to serve the diverse needs of users relies on protecting and maintaining both print and electronic information resources. In Nigeria, Incidents such as the loss of books, removal of journals, damage to reference items, and digital breaches have become common occurrence in library, these problems limit student and staff access to vital learning contents. Although the use of Convolutional Neural Networks (CNNs), show promise for automating the detection of suspicious activity using image analysis, common CNNs face obstacles such as high computational time and low accuracy challenges. To improve the CNN model, Optimisation strategies like the Spider Wasp Optimiser (SWO) have been explored, but issues with consistency and convergence remain. Therefore, this paper introduced an Enhanced Spider Wasp Optimiser (ESWO) that incorporates roulette wheel selection, designed to better fine-tune CNN hyperparameters for automated image recognition. The spider wasp Optimisation algorithm was enhanced using roulette wheel selection method which assigns selection probabilistic based on individual fitness instead of random selection method in the standard SWO. The enhanced Spider Wasp Optimisation (ESWO) which formed ESWO-CNN was then used to optimise CNN settings for feature extraction. The study gathered face images from LAUTECH students, faculty of computing and Informatics and applied a pre-processing work. The ESWO-CNN was implemented with the preprocessed facial images in MATLAB R2023a. The performance of the formulated model was measured using sensitivity, specificity, precision, accuracy, false positive rate, computation time, and compared against existing Spider Wasp Optimised-CNN (SWO-CNN) and standard CNN approaches. The results revealed that the developed ESWO-CNN model attained the sensitivity of 99.18%, specificity of 98.92%, precision of 99.07%, accuracy of 99.07%, with False Positive Rate of 1.08% and minimal computational time of 60.09 seconds. The SWO-CNN model showed notable performance enhancement with a sensitivity of 98.12%, specificity of 97.53%, precision of 98.12%, accuracy of 97.87%, and a reduced false positive rate of 2.47%, taking 71.64 seconds to execute. In contrast, the standard CNN model achieved a sensitivity of 96.71%, specificity of 95.52%, precision of 96.60%, accuracy of 96.20%, with a false positive rate of 4.48% and computational time of 96.04 seconds. The ESWO-CNN approach constitutes a highly effective, scalable, and efficient method for safeguarding both print and electronic resources in academic library.