Title :
Research on digital image recognition system based on multiple invariant moments theory and BP neural network
Author :
Jian-ning, Han ; Ming-quan, Wang
Author_Institution :
Sch. of Inf. & Commun. Eng., North Univ. of China, Taiyuan, China
Abstract :
Based on the pattern recognition theory, principle processes of a digital image are described. To achieve de-noising, sharpening, binary transformation, edge extraction and other image processing of digital image, the image recognition system was programmed in C language. The system calculates the seven invariant moments as all characters of an image, recognizes an image by BP neural network.
Keywords :
backpropagation; image recognition; neural nets; BP neural network; C language; binary transformation; denoising; digital image recognition system; edge extraction; image processing; multiple invariant moments theory; pattern recognition theory; sharpening; Character recognition; Data mining; Digital images; Feature extraction; Image recognition; Neural networks; Pattern recognition; Personal communication networks; Robotics and automation; System testing; BP neural network; digital image recognition system; multiple invariant moments theory;
Conference_Titel :
Informatics in Control, Automation and Robotics (CAR), 2010 2nd International Asia Conference on
Conference_Location :
Wuhan
Print_ISBN :
978-1-4244-5192-0
Electronic_ISBN :
1948-3414
DOI :
10.1109/CAR.2010.5456719
