| Multiresolution Using Principal Component Analysis (2007) | |||||||||||||||
Abstract | |||||||||||||||
| This paper proposes Principal Component Analysis (PCA) to find adaptive bases for multiresolution. An input image is decomposed into components (compressed images) which are uncorrelated and have maximum l 2 energy. With only minor modification, a single layer linear network using the Generalized Hebbian Algorithm (GHA) is used for multiresolution PCA. The decomposition has been successfully applied to face classification [3]. Good results with biological signals have also been reported [1]. 1. INTRODUCTION Principal Component Analysis (PCA) has many advantages for signal reconstruction. PCA is widely applied, the theory is well known, and numerous implementations are available. Other benefits include: 1. components provide optimal l 2 reconstruction, 2. components are uncorrelated, 3. the transformation is linear. PCA lends itself easily to signal compression by selecting components with high energy. However, PCA features do not guarantee optimal classification [6]. Multiresolu... | |||||||||||||||
Details der Publikation | |||||||||||||||
| |||||||||||||||