Principal Component Analysis Using Structural Similarity Index for Images
This work addresses the need for principled image quality assessment in machine learning, offering a novel approach that bridges image quality assessment and manifold learning, though it appears incremental as it adapts an existing method (PCA) with a known measure (SSIM).
The paper tackles the problem of image fidelity and similarity assessment by proposing Image Structural Component Analysis (ISCA), which uses the Structural Similarity Index (SSIM) instead of Euclidean distance in PCA to better capture structural features of images, resulting in a method that discriminates different types of image distortions.
Despite the advances of deep learning in specific tasks using images, the principled assessment of image fidelity and similarity is still a critical ability to develop. As it has been shown that Mean Squared Error (MSE) is insufficient for this task, other measures have been developed with one of the most effective being Structural Similarity Index (SSIM). Such measures can be used for subspace learning but existing methods in machine learning, such as Principal Component Analysis (PCA), are based on Euclidean distance or MSE and thus cannot properly capture the structural features of images. In this paper, we define an image structure subspace which discriminates different types of image distortions. We propose Image Structural Component Analysis (ISCA) and also kernel ISCA by using SSIM, rather than Euclidean distance, in the formulation of PCA. This paper provides a bridge between image quality assessment and manifold learning opening a broad new area for future research.