Research on the Application of Mechanical Parts Using Neural Network in Image Recognition
Neural network is an analog network element of human brain cells,which are non-linear dynamic systems connected to each other by a large number of neurons formed.Research on the present online parts recognition system defects on the speed and accuracy,which based on the neural network to identify the mechanical parts of the image has been described specifically.The first is to acquire an image.Then,I make the experiment about the image processing edge detection and feature extraction.Secondly,I select the appropriate torque characteristics,and make sure BP neural network model which is based on neural network theory and the experimental structure.Finally,I make experiment to test the target.The experimental results show that the machine part image recognition method based on neural network algorithm is simple high speed precision,and won a good recognition effect.
the neural network image recognition image processing feature extraction
Liu Bo Chen Ying Qi Xinghua
Key Laboratory for Photoelectric Detection Technology Harbin University of Science and Technology Harbin, China
国际会议
秦皇岛
英文
659-662
2015-09-18（万方平台首次上网日期，不代表论文的发表时间）