Butterfly Species Recognition Using Artificial Neural Network (ANN)

• Main Author: Yong, Kai Xin
• Format: Monograph
• Language: English
• Published: Universiti Sains Malaysia 2017
• Subjects: T Technology
• Online Access: http://eprints.usm.my/52927/1/Butterfly%20Species%20Recognition%20Using%20Artificial%20Neural%20Network%20%28ANN%29_Yong%20Kai%20Xin_E3_2017.pdf; http://eprints.usm.my/52927/

In 2017, there are about 20,000 species of butterfly has been discovered all over the world. Butterfly is well known because of its beautiful wings pattern and its benefits to the environment. In this research, butterfly species recognition is automated using artificial intelligence. Pattern on the butterfly wings is used as a parameter to determine the species of the butterfly. The butterfly image is captured and the background of the image is removed to make the recognition process easier. Local binary pattern (LBP) descriptor is then applied to the processed image and a histogram consist of image information is computed. Artificial neural network (ANN) is used to classify the image.