A histogram-based approach for object-based query-by-shape-and-color in image and video databases

Abstract

Cataloged from PDF version of article.Considering the fact that querying by low-level object features is essential in image and video data, an efficient approach for querying and retrieval by shape and color is proposed. The approach employs three specialized histograms, (i.e. distance, angle, and color histograms) to store feature-based information that is extracted from objects. The objects can be extracted from images or video frames. The proposed histogram-based approach is used as a component in the query-by-feature subsystem of a video database management system. The color and shape information is handled together to enrich the querying capabilities for content-based retrieval. The evaluation of the retrieval effectiveness and the robustness of the proposed approach is presented via performance experiments. (C) 2005 Elsevier Ltd All rights reserved

Histogram similarity measure using variable bin size distance

“…The advantage of histogram is well reflected in this field especially for large database. The main types of image retrieval contain the texture image retrieval and the color image retrieval, on which many studies focus [15][16][17]. Here, we test the VBSD measure in these two retrieval tasks.…”

Silhouette-Based Method for Object Classification and Human Action Recognition in Video

“…In order to calculate the center of mass point, C m = (x Cm , y Cm ), of an object O, we use the following equation [18]:…”

“…There are numerous methods in the literature for comparing shapes [20,5,18,2,13]. The reader is especially referred to the surveys presented in [25,16] for good discussions on different techniques.…”

Automatic detection of salient objects and spatial relations in videos for a video database system

“…Some examples of prototype multimedia database systems include QBIC [2], VisualSeek [3], and VideoQ [4]. If the media is video, the queried features are spatial, spatio-temporal, semantic, motion (e.g., object trajectories) and object features (e.g., color, shape, texture) [5]. Before retrieval, the videos need to be processed to extract the features that can be queried.…”