Dissertation > Excellent graduate degree dissertation topics show
Research of Various Feature-Based Image Retrieval in News Photo-Database
Author: QiuYang
Tutor: ChenZuo
School: Wuhan University
Course: Applied Computer Technology
Keywords: News Photo Gallery Image Retrieval Integrated features Relevance feedback
CLC: TP391.41
Type: Master's thesis
Year: 2004
Downloads: 161
Quote: 2
Read: Download Dissertation
Abstract
|
News Photo Gallery frequently encountered picture before storage check weight and find a group of similar pictures involving operation of image similarity judgments . The image similarity judgment requires a combination of visual perception of the picture information , such as pictures of color, composition , and so on . However , the characteristics of these visual perception is difficult to describe in words , and text descriptions by subjective factors caused by processing and query different description retrieve contradiction . Keyword search for text Similar Image Retrieval of low efficiency , poor results , we have developed a retrieval system based on image features to achieve the objective in the news photo library to retrieve similar images . Currently, more mature content - based image retrieval technology is mainly concentrated in the low-level features of the image such as color , texture , and color histogram , contrast , roughness , directionality efficient feature descriptors . The system combines news photo composition characteristics of the picture is divided into different weight four regions , integrated use of HIS color model color histogram and the MPEG-7 standard recommended color layout descriptor (CLD) and edge histogram descriptor ( EHD), respectively, of the image corresponding feature extraction , and normalization process . Proposed a relevance feedback method, based on the evaluation of the users of the search results , to access photo news text description of the main objective , the retrieval results closer to the user goals . Tests show that the image library contains 1000 photo , the system has better retrieval effect .
|
Related Dissertations
- Research on Facial Feature Extraction and Matching Algorithms for Image Retrieval,TP391.41
- The Research and Implemention of Image Retrieval Based on User Interested Feature,TP391.41
- Application of Q-Learning in the Content-Based Image Retrieval Technology,TP391.41
- Research and Implementation on Content-Based Clothing Image Retrieval,TP391.41
- The Discovery of User Concept Region Based on Multiple Instance Learning,TP391.41
- Research on Transductive Support Vector Machine and Its Application in Image Retrieval,TP391.41
- Research on Query Expansion Technique of Retrieval System in Biomedical Field,TP391.3
- Research and Application of Diverse Density Learning Algorithm,TP181
- Research of Shape Based Image Retrieval in Han Dynasty Stone,TP391.41
- The Research on Content-based Medical Image Retrieval Algorithm,TP391.41
- Research on Locally Features Aggregating and Indexing Algorithm in Large-scale Image Retrieval,TP391.3
- Integrated multi- image retrieval technology,TP391.41
- Micro- blog real-time tracking and retrieval topics related research,TP393.092
- Design and implementation of a trademark search system based on image recognition,TP391.41
- Retrieval technology based on the theory of nearness gallery,TP391.41
- Research on Techniques of Medical Image Retrieval Based on Feature Spatial Distributing,TP391.41
- Theory and Method Research on Content Based Color Image Retrieval,TP391.41
- Product Image Retrieval Based on Shape,TP391.41
- Image retrieval based on combined features of the method,TP391.41
- Fusion of Color and shape features of image retrieval method,TP391.41
- Image retrieval based on relevance feedback research,TP391.41
CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Pattern Recognition and devices > Image recognition device
© 2012 www.DissertationTopic.Net Mobile
|