Query by image example: The CANDID approach
Abstract
CANDID (Comparison Algorithm for Navigating Digital Image Databases) was developed to enable content-based retrieval of digital imagery from large databases using a query-by-example methodology. A user provides an example image to the system, and images in the database that are similar to that example are retrieved. The development of CANDID was inspired by the N-gram approach to document fingerprinting, where a ``global signature`` is computed for every document in a database and these signatures are compared to one another to determine the similarity between any two documents. CANDID computes a global signature for every image in a database, where the signature is derived from various image features such as localized texture, shape, or color information. A distance between probability density functions of feature vectors is then used to compare signatures. In this paper, the authors present CANDID and highlight two results from their current research: subtracting a ``background`` signature from every signature in a database in an attempt to improve system performance when using inner-product similarity measures, and visualizing the contribution of individual pixels in the matching process. These ideas are applicable to any histogram-based comparison technique.
- Authors:
-
- Los Alamos National Lab., NM (United States). Computer Research and Applications Group
- Univ. of New Mexico, Albuquerque, NM (United States). Dept. of Electrical and Computer Engineering
- Publication Date:
- Research Org.:
- Los Alamos National Lab., NM (United States)
- Sponsoring Org.:
- USDOE, Washington, DC (United States)
- OSTI Identifier:
- 28339
- Report Number(s):
- LA-UR-95-374; CONF-950226-12
ON: DE95006284; TRN: AHC29510%%144
- DOE Contract Number:
- W-7405-ENG-36
- Resource Type:
- Conference
- Resource Relation:
- Conference: SPIE `95: SPIE conference on optics, electro-optics, and laser application in science, engineering and medicine, San Jose, CA (United States), 5-14 Feb 1995; Other Information: PBD: [1995]
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 99 MATHEMATICS, COMPUTERS, INFORMATION SCIENCE, MANAGEMENT, LAW, MISCELLANEOUS; INFORMATION RETRIEVAL; C CODES; DATA BASE MANAGEMENT; INFORMATION SYSTEMS; USES; REMOTE SENSING; MEDICINE; IMAGES; DIGITAL SYSTEMS
Citation Formats
Kelly, P M, Cannon, M, and Hush, D R. Query by image example: The CANDID approach. United States: N. p., 1995.
Web.
Kelly, P M, Cannon, M, & Hush, D R. Query by image example: The CANDID approach. United States.
Kelly, P M, Cannon, M, and Hush, D R. Wed .
"Query by image example: The CANDID approach". United States. https://www.osti.gov/servlets/purl/28339.
@article{osti_28339,
title = {Query by image example: The CANDID approach},
author = {Kelly, P M and Cannon, M and Hush, D R},
abstractNote = {CANDID (Comparison Algorithm for Navigating Digital Image Databases) was developed to enable content-based retrieval of digital imagery from large databases using a query-by-example methodology. A user provides an example image to the system, and images in the database that are similar to that example are retrieved. The development of CANDID was inspired by the N-gram approach to document fingerprinting, where a ``global signature`` is computed for every document in a database and these signatures are compared to one another to determine the similarity between any two documents. CANDID computes a global signature for every image in a database, where the signature is derived from various image features such as localized texture, shape, or color information. A distance between probability density functions of feature vectors is then used to compare signatures. In this paper, the authors present CANDID and highlight two results from their current research: subtracting a ``background`` signature from every signature in a database in an attempt to improve system performance when using inner-product similarity measures, and visualizing the contribution of individual pixels in the matching process. These ideas are applicable to any histogram-based comparison technique.},
doi = {},
url = {https://www.osti.gov/biblio/28339},
journal = {},
number = ,
volume = ,
place = {United States},
year = {1995},
month = {2}
}