03-12-2012, 03:03 PM
Content Based Image Retrieval Using Boundary Based Shape Descriptors
1Image Retrieval.ppt (Size: 1.74 MB / Downloads: 44)
Abstract
More and more images have been generated in digital form around the world. There is a growing interest in finding images in large collections. From the large number of images it is very important for users to retrieve required images via an efficient and effective mechanism. Image has various low level features which specifies its content such as color, texture, shape. Exploration of the image content representation, similarity matching and ranking and retrieving techniques based on low level features. This project focuses the boundary based shape features such as chain codes, shape signature, Fourier descriptors and the study of those features in image retrieval. The input image features value will be compared with the feature value in the database (MPEG-7 database). The proposed shape features and benchmarking database will improve the performance metrics.
Keyword Based Image Retrieval
Retrieving images by manually annotated keywords.
In the database images and the keyword is present .
When we query by a keyword ,the system matches the query to the keywords present in the database.
Finally ,the images are retrieved
Eg: google, yahoo , lycos search engines
Content Based Image Retrieval
Content-based means that the search will analyze the actual contents of the image.
Images are retrieved based on visual/low level features.
Searches the images similar to an query image.
similarity measures
Similarity measurement is the process of finding the similarity/difference between the database images and the query image using their features
Different methods are
Euclidean distance
Earth movers distance
Cross-Bin Dissimilarity Measures
CONCLUSION
Thus the literature survey has studied, images has collected and the feature expression were collected and finished