Download | - View accepted manuscript: Objects recognition using SIFT and fuzzy similarity measure (PDF, 446 KB)
|
---|
Author | Search for: Karddouchi, M.; Search for: Belacel, Nabil |
---|
Format | Text, Article |
---|
Proceedings title | The 20th Annual IS&T/SPIE Symposium on Electronic Imaging [Proceedings] |
---|
Conference | The 20th Annual IS&T/SPIE Symposium on Electronic Imaging, January 27-31, 2008, San Jose, California, USA |
---|
Subject | SIFT descriptors; descripteurs SIFT; fuzzy similarity; similarité floue; object recognition; reconnaissance des objets; fuzzy triangular number; nombre triangulaire flou |
---|
Abstract | Multimedia database has been an extremely active area of research over the last 20 years. This research aims to develop techniques for searching and recognizing multimedia documents based on their content. For objects recognition, make works proposed different techniques to extract the visual contents such as color, shape, texture, etc. By comparing these visual contents, we can determine whether or not the image data contains some specific object. Since the Euclidean distance has always been employed to compare visual contents so far, using other approaches for the comparison is an interesting research way that still needs to be explored. In this paper, we propose fuzzy similarity measures as alternatives for the Euclidean distance. Visual contents to be compared are based on the Scale Invariant Feature Transform (SIFT). This approach has been applied to coil databases. Our experimentation shows that this method is more realistic than objects recognition method obtained by classical distances. |
---|
Publication date | 2008 |
---|
Language | English |
---|
NRC number | NRCC 49914 |
---|
NPARC number | 8914130 |
---|
Export citation | Export as RIS |
---|
Report a correction | Report a correction |
---|
Record identifier | d79cef04-eba1-4f1a-886a-a8be3c7e6c01 | Record created | 2009-04-22 |
---|
Record modified | 2020-08-12 |
---|