AI SearchAn algorithm for the retrieval of image spatial relationship from a pictorial database
Call Number: AIT Thesis no. CS-94-34 Material type:
TextSeries: Asian Institute of Technology. Thesis ; no. CS-94-34Publication details: Bangkok : Asian Institute of Technology, 1994Description: 64 leavesSubject(s): Online resources: Dissertation note: Thesis (M.Sc.) - Asian Institute of Technology, 1994 Summary: In this thesis, we suggest a method of retrieving pictures in pictorial database based upon the spatial relationship among the objects in the picture. When there exist a large volume of images in the image database (IDB), the processing time for image retrieval is tremendous. It is necessary to develop efficient access methods for retrieval. An ARTl is alternative method that can be used to search the desired image from IDB. Before the searching can be done by this method, some techniques for clustering, minimizing and encoding must be carried out. This thesis has designed some techniques to apply in ARTl algorithm. The ARTl algorithm was modified to apply for the different types of matching based on the threshold parameter. This threshold parameter is calculated from the input patterns, and this will identify wheter or not the retrieval is exact or similar retrieval. The proposed method is compared with the existing method of 2-D strings. The experimental results reveal that the ARTl method has advantages in response time when there exist a large amount of images in image database.
| Cover image | Item type | Current library | Home library | Collection | Shelving location | Call number | Materials specified | Vol info | URL | Copy number | Status | Notes | Date due | Barcode | Item holds | Item hold queue priority | Course reserves | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
20-AIT Publication
|
Asian Institute of Technology Library AIT Publications | AIT Thesis no. CS-94-34 (Browse shelf(Opens below)) | 1 | Available | 30050003102273 | |||||||||||||
|
20-AIT Publication
|
Asian Institute of Technology Library AIT Publications | AIT Thesis no. CS-94-34 (Browse shelf(Opens below)) | 2 | Available | 30050003102265 | |||||||||||||
40-Archives
|
Asian Institute of Technology Library Archives | AIT Thesis no. CS-94-34 (Browse shelf(Opens below)) | Available | 30050120356513 |
A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science, School of Engineering and Technology
Thesis (M.Sc.) - Asian Institute of Technology, 1994
In this thesis, we suggest a method of retrieving pictures in pictorial database based upon the spatial relationship among the objects in the picture. When there exist a large volume of images in the image database (IDB), the processing time for image retrieval is tremendous. It is necessary to develop efficient access methods for retrieval. An ARTl is alternative method that can be used to search the desired image from IDB. Before the searching can be done by this method, some techniques for clustering, minimizing and encoding must be carried out. This thesis has designed some techniques to apply in ARTl algorithm. The ARTl algorithm was modified to apply for the different types of matching based on the threshold parameter. This threshold parameter is calculated from the input patterns, and this will identify wheter or not the retrieval is exact or similar retrieval. The proposed method is compared with the existing method of 2-D strings. The experimental results reveal that the ARTl method has advantages in response time when there exist a large amount of images in image database.
There are no comments on this title.