Introduction to Fast Multi-Resolution Image Querying Using Color Characteristics Seminar Topic:
The main aim of this project is to build content-based image retrieval depending color and generic shape. Another name of content-based image retrieval is also called as query-by image content (CBIR). It is originated in 1992 and it is used by Kato. For searching the images in the large data-base content-based visual information retrieval is used. Based on the size shape and color and other information are extracted from image by content-based image retrieval. There is another way to search the images by keywords, captions but this procedure is very expensive to achieve.
The CBIR process retrieves images automatically from database by syntactical analysis of image features. The tools that are used in this process are pattern recognition, signal processing and computer vision
Technical Progress:
From day-to-day the interest in CBIR was going on. With the help of current technology textual information can be easily extracted from images. But it’s a difficult process because it should be handled personally for describing every image information in data-base. There is a loss of images when different names are used in their descriptions.
Potential users of CBIR:
- Art collections
- Photograph Archives
- Retail catalogs
- Medical Records
Content-Comparison Techniques:
There are three techniques
- Color
- Texture
- Shape
Color:
This technique follows comparing the color histograms in images. This technique is most widely used.
Texture:
This technique follows by checking visual patterns and spatial in images. Actually Textures are represented using textures by identifying number of textures the images can be easily identified.
Shape:
Shape is detected using segmentation and edge detection and also particular region in the image. But in some circumstances human involvement is necessary for shape detection.
Download Fast Multi-Resolution Image Querying Using Color Characteristics Seminar Report.