The Discrimination between Digital Photos by Using Canonical Discriminate Function

Al - Rawi, Omar

doi:10.33899/tanra.2007.161684

	The Discrimination between Digital Photos by Using Canonical Discriminate Function
TANMIYAT AL-RAFIDAIN
Article 12, Volume 29, Issue 88, 2007, Pages 221-233 PDF (743.69 K)
Document Type: Research Paper
DOI: 10.33899/tanra.2007.161684
Author
Omar Al - Rawi
Abstract
Abstract In current research, one of scientific methods to discriminate between populations will be treated. This can be done by using (Matlab) program to transfer digital photo to a digital form, after that photo has been treated as an independent population, then we analyzed data and compare the results. It is important to imitate them by using canonical discriminate function to classify digital photos that will give us a new way of facilitation search. However, after we transferred it to the digital form, we depend on a higher eigenvector who is corresponding higher eigenvalue. This one according to the number of population we dial with it. In this research, we considered four populations, and every population is digital photo we transferred it to data by using (Matlab) program, symbolic every population depending on the canonical discriminate. We depended on many program (SPSS, Minitab, S-Plus, Matlab) to get the result of three higher eigenvalue which corresponding three higher eigenvector for (A) Matrix. The final steps assimilation has been demonstrated to get three canonical discriminate functions to discriminate between digital photos which facilitate the eigenvector for (A) matrix, and the resulted as the flowing, discriminate between digital photos in the first canonical discriminate function which is the best one. The second rank is the second canonical discriminate function; in the last stage is the third canonical discriminate function. .
Keywords
Digital Photos; Canonical Discriminate Function

Statistics Article View: 315 PDF Download: 366