Image database querying using digital images
By: Roses, Jose Nelson III
Publisher: Cebu City ; CIT-U ; 2011DDC classification: T R7331 2011 Summary: Today, image-searching querying has gone from using tags and identifiers to using the actual image content. This method is called Content ? based Image Retrieval. This paper presents a practical application of this querying method. For a brief description, this system takes a digital image of an object, and will recognize in the image, the object to be searched, and by the using the results as entry, will search and return a matching object in he database. The user is allowed two methods of input. First is by loading an image. Second is by capturing an image of the object itself using a PC camera. The output is an array of four images considered to be the closest match to the input object arranged in descending order according to similarity. Practical applications include information systems where textual input is generally inefficient. This system is useful especially when images are the only means of finding information. The system is tested using of fruits, flags, flowers and brand logos. Our tests have shown that the method works and is able to distinguish certain features of perceptually similar images.| Item type | Current location | Home library | Call number | Status | Date due | Barcode | Item holds |
|---|---|---|---|---|---|---|---|
RESERVED BOOKS
|
COLLEGE LIBRARY | COLLEGE LIBRARY | T R7331 2011 (Browse shelf) | Available | T1654 |
Browsing COLLEGE LIBRARY Shelves Close shelf browser
Today, image-searching querying has gone from using tags and identifiers to using the actual image content. This method is called Content ? based Image Retrieval. This paper presents a practical application of this querying method. For a brief description, this system takes a digital image of an object, and will recognize in the image, the object to be searched, and by the using the results as entry, will search and return a matching object in he database. The user is allowed two methods of input. First is by loading an image. Second is by capturing an image of the object itself using a PC camera. The output is an array of four images considered to be the closest match to the input object arranged in descending order according to similarity. Practical applications include information systems where textual input is generally inefficient. This system is useful especially when images are the only means of finding information.
The system is tested using of fruits, flags, flowers and brand logos. Our tests have shown that the method works and is able to distinguish certain features of perceptually similar images.
000-099
There are no comments for this item.