APLIKASI JARINGAN SYARAF TIRUAN DENGAN MENGGUNAKAN METODE BACKPROPAGATION DALAM KASUS PENGENALAN POLA HURUF HIJAIYAH

ANGGI RIZKY WINDRA PUTRI, NIM. 06650020 (2013) APLIKASI JARINGAN SYARAF TIRUAN DENGAN MENGGUNAKAN METODE BACKPROPAGATION DALAM KASUS PENGENALAN POLA HURUF HIJAIYAH. Skripsi thesis, UIN SUNAN KALIJAGA.

Preview	Text (APLIKASI JARINGAN SYARAF TIRUAN DENGAN MENGGUNAKAN METODE BACKPROPAGATION DALAM KASUS PENGENALAN POLA HURUF HIJAIYAH) BAB I, VII, DAFTAR PUSTAKA.pdf Download (1MB) | Preview
	Text (APLIKASI JARINGAN SYARAF TIRUAN DENGAN MENGGUNAKAN METODE BACKPROPAGATION DALAM KASUS PENGENALAN POLA HURUF HIJAIYAH) BAB II, III, IV, V, VI.pdf Restricted to Repository staff only Download (3MB)

Abstract

At this time the technology is considered very influential role in human life. The role of technology is to be one of the factors supporting progress in various fields. One is education. Learning system by utilizing the application of technology as well be one attractive option for children in enhancing creativity and learning systems. With an attractive appearance and assisted learning tools that support will increase the child's interest in learning. In consideration of the reasons the authors propose to create a system that recognizes the letter to the specifications hijaiyah as object identification by detecting the letter by letter the input image capture through the camera. Then processed to capture the results of image processing and image will be formatted using a threshold to black and white then processed binerisasi where the black parts of the image are considered as binary 1 and the white parts are considered as binary 0. Then binerisasi images used as input of neural network. Obtained from these studies that hijaiyah recognition system that utilizes image processing supported the use of neural networks as pattern recognition hijaiyah, In matlab simulation, test data are in the know at 99% and the simulation of the real success of the test data in recognizing letters received grades 85 % by using a threshold of 100. Organization of the neural network has two hidden layers and one output layer using a binary sigmoid activation function. Ijaiyah letter recognition system gives output as text and voice information submitted. Keywords: Letter Recognition System, Neural Networks, Image Processing.

Share this knowledge with your friends :

Actions (login required)

View Item