عنوان المقالة:تميز راحة ايد عن طريق اشراك خصائص سلسلة فريمان والخصائص النسجوية وبمساعدة المنطق المضبب Features with Fuzzy Logic Classifier Hand Palm Recognition using Combination of Freeman Chain Code and Texture
In this paper we propose a system for hand palm recognition and analysis based on features presented by hand palm lines. The proposed system based on images taken from free Indian Institute of Technology Delhi (IITD) palmprint database of hand palm. Firstly we extracted the region of interest (ROI) from hand palm image in order to find two groups of features, the first group contain 6 texture features represented by (mean, standard deviation, smoothing, third moment, uniformity, entropy) from gray ROI whereas the second group contain 10 features extracted from (white-black) ROI, 8 of them are extracted from freeman chain code depend on 8 neighbors (0, 1, 2, 3, 4, 5, 6, 7) the ninth and tenth features represent the area and entropy respectively. These two groups consist the vector features which contain 16 values for each ROI image which stored in our database prepared to be used in recognition phase by using fuzzy logic classifier with mamdani rule. The proposed system is applied on 120 images 60 samples are used for training and the remaining 60 samples are used for testing. The proposed system's accuracy is 90% for testing group with average of execution time is 13.6 second using Matlab environment. Thus, the recognition rate is higher when using the combination of freeman chain code and texture features with fuzzy logic classifier
تاريخ النشر
15/03/2015
الناشر
International Journal of Computer Science and Mobile Computing
زمن فاضل جبر | zamen fadhel jabr | 2840