Media.Player

Automatic and Semi-automatic Image Classification, Annotation and Tagging through the Use of Image Acquisition Parameters and Metadata

Summary

This is a powerful search engine for images. A method for characterizing image contents automatically or semi-automatically using image acquisition parameters and metadata is presented. The method establishes probabilistic and deterministic relationships between different types of metadata and the semantic attributes and contents of images. It furnishes a mechanism that enables the automatic and semi-automatic classification, annotation, tagging, indexing, searching, identification or retrieval of images based on their contents, semantic properties and metadata characteristics. The method utilizes image capture metadata such as focal length, exposure time, relative aperture, flash information, ISO setting, angle of view, subject distance, timestamp, GPS information as well as other forms of metadata, including but not limited to, captions, keywords, headings, tags, comments, remarks, titles which may be automatically, semi-automatically, or manually generated.

Technology Benefits

High consistency, precision and reliability and time-saving

Technology Application

Image databases, web searching, personal search, community search, broad-based or vertical search engines for internet, intranet, extranet or other usages

Detailed Technology Description

The number of electronic images being created is increasing at a rapid rate, and searching them semantically presents a significant challenge. Many raw images are made available with few meaningful direct annotations of semantic content, limiting their search and discovery. While some image repositories or Web sites encourage tags or keywords to be included manually, such is far from universal. Manual characterization of semantic image contents is often subjective, labor intensive, and inconsistent. The relative time-consuming process of manual characterization or annotation is also unlikely to be able to keep up with the rate of creation of images through digital and non-digital means. This invention introduces GPS data, time information or image-recognition techniques such as face detection, face recognition, annotations and characterization of the content of images. Color feature extraction, shape feature extraction, and texture feature extraction are further applied to images to create a semantic database of images. The state-of-the-art technology is improved annotation and classification of semantic image contents.

Supplementary Information

Patent Number: US8520909B2
Application Number: US13255241A
Inventor: Leung, Clement Ho Cheung | Wong, Chun Fan | Liu, Jiming
Priority Date: 11 Mar 2009
Priority Number: US8520909B2
Application Date: 8 Sep 2011
Publication Date: 27 Aug 2013
IPC Current: G06K000900
US Class: 382118
Assignee Applicant: Hong Kong Baptist University
Title: Automatic and semi-automatic image classification, annotation and tagging through the use of image acquisition parameters and metadata
Usefulness: Automatic and semi-automatic image classification, annotation and tagging through the use of image acquisition parameters and metadata
Summary: Method for semantically characterizing semantic attributes and contents of digital or non-digital images such as Web images and non-Web images, located in public and private image repositories use during web searching, personal search, image databases, community search, broad-based or vertical search engines for internet, intranet and extranet (all claimed).
Novelty: Semantic characterizing method for semantic attributes and contents of e.g. web image, located in public image repositories use during web searching, involves using semantic image characterization to index image contents in storage medium

Industry

Electronics

Sub Category

Computer System

Application Date

2010.03.11

Application No.

US 13/255,241

Others

Granted

ID No.

G06/LC/Auto/US01

Country/Region

Hong Kong