Indian Journal of Science and Technology

Open Access Open Access  Restricted Access Subscription Access

A New Method in Determining the Degree of Similarity Using Symbols and Average Color Spectrum Characteristics


Affiliations
1 Islamic Azad University, Jouybar Branch, Jouybar, Iran; 1Islamic Azad University, Sari Branch, Sari, Iran, Islamic Republic of
 

A lot of researches have been done about image processing these years. One of the most widely used fields is face recognition based on the appearance features. In this research, 250 samples of different people's faces have been taken in RGB mode as the input data and every image is changed to HSV colored mode. Then we calculate each image's symbols and the average colored of each symbol. Then we calculate the angles between applied symbols. In the next step we calculate the considered parameters based on symbols, average colored spectrum of symbols and the angles between them and they have been saved in the database. Finally, we determine the degree of similarity between images using features such as symbols, average colored spectrum of symbols and the angle between them. They were grouped as set. Based on the observed experimental results, this method's efficiency is 10 percent more than the method which is based on the symbol. Also it is more efficient 10.5 percent than the methods based on spatial and objective similarity or only based on the color.

Keywords

Image processing, the average spectrum of color, HSV color mode, symbolic images
User

  • Gonzalez R, Woods RE (2007) Digital image processing. 3rd ed, Addison Wesley Publ. Co., NY.

  • Punithaa P, Gurub DS (2008) Symbolic image indexing and retrieval by spatial similarity:An approach based on B-tree. Pattern Recognition. 41, 2068–2085.

  • Shu Ming H and Chiun Chieh H (2008) Retrieval of images by spatial and object similarities. Information Processing & Management. 44, 1214–1233.

  • Subramanyam VP and Sett SK (2007) Image retrieval system using R-tree self-organizing map. Data & Knowledge Engg. 61, 524–539.

  • Umbaugh S (1998) Computer vision & image processing. Indianapolis: Books craft.

  • Venkateswara RR and Subramanyam VP (2008) Rotation-invariant texture retrieval using waveletbased hidden Markov trees. Signal Processing. 88, 2593–2598.


Abstract Views: 381

PDF Views: 104




  • A New Method in Determining the Degree of Similarity Using Symbols and Average Color Spectrum Characteristics

Abstract Views: 381  |  PDF Views: 104

Authors

Hossein Shirgahi
Islamic Azad University, Jouybar Branch, Jouybar, Iran; 1Islamic Azad University, Sari Branch, Sari, Iran, Islamic Republic of
Najmeh Danesh
Islamic Azad University, Jouybar Branch, Jouybar, Iran; 1Islamic Azad University, Sari Branch, Sari, Iran, Islamic Republic of

Abstract


A lot of researches have been done about image processing these years. One of the most widely used fields is face recognition based on the appearance features. In this research, 250 samples of different people's faces have been taken in RGB mode as the input data and every image is changed to HSV colored mode. Then we calculate each image's symbols and the average colored of each symbol. Then we calculate the angles between applied symbols. In the next step we calculate the considered parameters based on symbols, average colored spectrum of symbols and the angles between them and they have been saved in the database. Finally, we determine the degree of similarity between images using features such as symbols, average colored spectrum of symbols and the angle between them. They were grouped as set. Based on the observed experimental results, this method's efficiency is 10 percent more than the method which is based on the symbol. Also it is more efficient 10.5 percent than the methods based on spatial and objective similarity or only based on the color.

Keywords


Image processing, the average spectrum of color, HSV color mode, symbolic images

References





DOI: https://doi.org/10.17485/ijst%2F2010%2Fv3i10%2F29838