Texture recognition by using artificial neural network

• Main Author: Foong, Lee Sai
• Format: Undergraduates Project Papers
• Language: English
• Published: 2013
• Subjects: QA Mathematics
• Online Access: http://umpir.ump.edu.my/id/eprint/7243/1/Texture_recognition_by_using_artificial_neural_network.pdf; http://umpir.ump.edu.my/id/eprint/7243/

This thesis describes the texture recognition by using the Artificial Neural Network (ANN). There are hard to understand on how to perform the texture recognition on any new set of image data. Therefore, to ease up the process on texture recognition, ANN has been chosen as the classifier to enhance the process of the texture recognition. There are thirteen types of Brodatz textures are considered as the dataset for this research and five sets for each type texture with different level of histogram equalized, noise for the training dataset. Backpropagation algorithm is one of the methods for the ANN. After the feature is obtained from the dataset, the feature will be trained and classifier by using theBack-propagation algorithm. All in all, this project will tell us how the Back-propagation classifier help in texture recognition and how to increases the success rate in texture recognition.