期刊名称：International Journal of Innovative Research in Science, Engineering and Technology
印刷版ISSN：2347-6710
电子版ISSN：2319-8753
出版年度：2017
卷号：6
期号：4
页码：5666
DOI：10.15680/IJIRSET.2017.0604125
出版社：S&S Publications
摘要：Main objective of this paper is the currency recognition using ANN (Artificial Neural Network)withfeatures of paper currency sample extracted using Canny Edge Histogram and GLCM. Canny edge detection histogramdetects even weak edges and it has low error rate.A1x5edge orientation histogram is computed. As sobel mask isapplied to find the max sobel gradient. GLCM (Gray Level Co-Occurrence Matrix) is used for analyzing the texture. Itis a statistical method that considers the spatial relationship of the pixels and characterizes the texture of the image.The extracted features are used in feed forward back propagation network.The project is implemented in MATLAB environment. Neural network uses 75% of extracted features fortraining and remaining 25% features for testing. The network performance is evaluated with the percentage of correctclassification.
