A Review of Quantum Machine Learning for Image Classification

In the machine learning domain of image classification, computational capability stands as a significant constraining factor. However, harnessing the robust computational power of quantum computers presents a viable solution. This paper reviews the application of QML in image classification, with quantum convolutional neural network (QCNN) and quantum K-nearest neighbor algorithm (QKNN) as the core content. Primarily, a compendium of data embedding methodologies, commonly enlisted within the domain of quantum machine learning, is elucidated. Subsequently, an intricate exploration ensues into the present state of QKNN and QCNN deployment in image classification, thereby delineating the latent and meritorious attributes underpinning QML's efficacy within the purview of image classification. Ultimately, a prospective trajectory for the future advancement of QML within the precincts of image classification is delineated.