Title :
A new method of the image pattern recognition based on neural networks
Author :
Fujun Ren ; Xinhua Zhang ; Long Wang
Author_Institution :
Coll. of Mech. & Eng., Harbin Univ. of Sci. & Technol., Harbin, China
Abstract :
A large number of computations are involved in images pattern recognition. In the process not only the speed is not quick, but also the accuracy is not high enough. To solve this problem, a new image pattern recognition method which the pixel points as the neurons to be trained is proposed and the Pattern recognition is constructed based on the neural network. A neural network system with pattern recognition function is designed in the identification processing. When an image is inputted to be identified, the source image will be given after the computed. The simulation test is done and the results show that the source image is recognized at high-speed and high accuracy. It can be used in the fast pattern recognition system.
Keywords :
image recognition; neural nets; identification processing; image pattern recognition function; neural network; source image; Accuracy; Biological neural networks; Educational institutions; Image recognition; Neurons; Pattern recognition; Training; image processing; image recognition; neural network; pattern recognition; pixel points;
Conference_Titel :
Electronic and Mechanical Engineering and Information Technology (EMEIT), 2011 International Conference on
Conference_Location :
Harbin, Heilongjiang
Print_ISBN :
978-1-61284-087-1
DOI :
10.1109/EMEIT.2011.6023833
