Content-Based Image Retrieval: A Comparison between Query by Example and Image Browsing Map Approaches

Abstract

Efficient and effective retrieval techniques of images are desired because of the explosive growth of digital images. Content-based image retrieval is a promising approach because of its automatic indexing and retrieval based on their semantic features and visual appearance. The similarity of images depends on the feature representation and feature dissimilarity function. However, users have difficulties in representing their information needs in queries to content-based image retrieval systems. In this paper, we investigate two approaches, query by example and image browsing map. Activities to support the information seeking behavior are analyzed. The performance of these approaches is measured by a user evaluation. It is found that the image browsing map provides more functionalities and capabilities to support the features of information seeking behavior and produces better performance in searching images.

Representación conceptual de imágenes médicas digitales: Integración de Contexto y Contenido Visual

“…Fuente: Propia Existen varios tipos de características que pueden ser extraídas de las imágenes para la representación de su contenido, estas características son usadas según sea su desempeño en la representación del contenido de la imagen. Por ejemplo, las características cromáticas y de textura se comportan mejor en imágenes generales, las características que representan información acerca de las formas identificadas en la imagen tienen un mejor rendimiento en imágenes que contienen gráficos construidos en computador y las características estructurales, en imágenes que contienen objetos construidos por el hombre, tales como los son las edificaciones (Yang, 2004). A continuación se presenta una clasificación de características que permiten extraer el contenido de una imagen digital (Mera, 2012):…”

Retrieval of similar chess positions

“…The search problem, however, is closely related to the problem of retrieving approximately matching entities (typically non-text) from a collection, given a query entity (typically also non-text) as an example. Examples of this class of search tasks include content-based image search [18], mathematical equations search [7], searching for the least edit distance reference sentence for example-based machine translation (EBMT) [10], etc. These search tasks differ widely from standard text search in the following ways.…”

“…The underlying scenario of this search problem is somewhat analogous to the content-based image retrieval (CBIR) problem, where the intention is to retrieve the top-most similar images, given a query image [18]. Conceptually speaking, the chess position search problem can be visualized as a CBIR problem, in which the intention is to retrieve a ranked list of snapshot images of chess positions ordered by decreasing values of similarities with respect to the query snapshot image.…”

Content based Image Retrieval System with Hybrid Feature Set and Recently Retrieved Image Library

“…CBIR systems that use low-level features for retrieval identifies the data base images that have visual similarity with the query image by comparing the low-level image features [8] like color, texture, shape and structure that are extracted from the images [9] [10]. The high-level description is an attributed graph attained by the structural representation of the image [16].…”