Principal Component Multi Linear Analysis for Content Based Image Retrieval
نویسنده
چکیده
In the process of content based Image retrieval (CBIR), image information is presented in descriptive features to obtain retrieval of image information. In the representation of descriptive features a large feature count is observed, which results in the overhead in processing. To reduce these descriptive features different dimensional reduction logic were used in which PCA is the most commonly used approach. However the approach of dimensional reduction is carried out by a Histogram space transformation and mapping, features when processing for retrieval exhibits multiple feature similarity among object classes. Hence considering all dataset features are not useful. In this paper, we present a Principal Component Multi-Linear Analysis (PCMLA) for dimension reduction approach to feature reduction based on feature relations for dimensional reduction approach
منابع مشابه
The Concept of a Multi-step Search-engine for the Content-based Image Retrieval Systems
For the Content-Based Image Retrieval System (CBIR) we propose how to put together vectors of features for segmented image objects and a spatial relationship of the objects by constructing a multistep search-engine, taking into account multi-set data mining and the object spatial relationship. The paper presents a combination of two aspects of image representation, namely: features of segmented...
متن کاملContent Based Image Retrieval of Corel Images Along With Face Recognition
: Image Retrieval basically deals with identification of similar images from a large database. Image retrieval based on rich content of the image is known as Content Based Image Retrieval (CBIR). The visual content of an image is analysed in terms of different features extracted from the image. The efficiency of CBIR techniques depends on the database selected. The database considered is Corel ...
متن کاملUsing Discriminant Eigenfeatures for Image Retrieval
|This paper describes the automatic selection of features from an image training set using the theories of multi-dimensional linear discriminant analysis and the associated optimal linear projection. We demonstrate the eeectiveness of these Most Discriminating Features for view-based class retrieval from a large database of widely varying real-world objects presented as \well-framed" views, and...
متن کاملContent based Image Retrieval using Gaussian Mixture Model based Subspaces Representation
Content Based Image Retrieval (CBIR) plays a significant role in case of image processing. Generally, in case of large scale dataset the two problems which are common viz. lower memory cost and higher retrieval accuracy. To solve the problem of the large scale retrieval the mixture of subspaces image representation is used. In this approach the group of the local descriptors of every individual...
متن کاملMammoSysLesion: a Content-Based Image Retrieval System for Mammographies
In this paper, we present a content-based image retrieval system designed to retrieve mammographies from large medical image databases. The system is developed based on breast density and lesion, according to the categories defined by the American College of Radiology, and is integrated to the database of the Image Retrieval in Medical Applications (IRMA) project, that provides images with clas...
متن کامل