Underwater Animal Identification and Classification Using a Hybrid Classical-Quantum Algorithm

Underwater Animal Identification and Classification is gaining significant importance in recent times due to the growing demand for ecological surveillance and biodiversity monitoring. Classical Deep learning techniques have been prominently used for these tasks, but due to the live capture of animals in complex environments, a limited sea-animal image dataset, and the complex topography of the seafloor, particularly in shallow waters, sediments, reefs, submarine ridges, and ship radiation, the efficacy of identification and classification is still a bottleneck for several researchers. In this paper, three hybrid Classical-Quantum neural networks ResNet50-QCNN, ResNet18-QCNN and InceptionV3-QCNN have been proposed for underwater quantum-classical Animal Identification and Classification. It significantly lessens the complexity of classical computer processing data by using quantum devices to minimize dimension and denoise datasets. The numerical simulation results demonstrate that the quantum algorithm is capable of effective dimensionality reduction and an improvement in classification accuracy. The hybrid approach offers polynomial acceleration in dimension reduction beyond classical techniques, even when quantum data is read out classically. The three hybrid models, viz., ResNet50-QCNN, ResNet18-QCNN, and InceptionV3-QCNN, displayed classification test accuracy of 88%, 80.29%, and 70%, respectively, revealing that ResNet50-QCNN performed best in identifying and classifying underwater animals.