An Image Retrieval Method Based on Manifold Learning with Scale-Invariant Feature Control

Abstract: Aiming at the problem of the traditional dimensionality reduction methods cannot recover the inherent structure, and scale invariant feature transform (SIFT) achieving low precision when reinstatingimages, an Image Retrieval Method Based on Manifold Learning with Scale-Invariant Feature is proposed. It aims to find low-dimensional compact representations of high-dimensional observation data and explores the inherent low and intrinsic dimension of data. The feature extraction method-SIFT and the adaptive ISOMAP method are combined and conducted experiments on the ORL face image dataset. This paper analyzes and discusses the problem of effects of the neighborhood parameter and the intrinsic dimension size on the face image recognition.
Keywords: image retrieval; manifold learning; dimensionality reduction; intrinsic dimension
Author: Haifeng Guo
Journal Code: jptkomputergg160063

Artikel Terkait :