基于优化子波-神经网络图像识别的工件形状监控

Workpiece Shape Monitoring Based on Image Recognizing of Wavelet- neural Network

介绍了一种用普通 CCD摄像机和激光对旋转工件的形态进行图像采集和状态监控的系统 ,提出了用于图像处理与识别的子波 -神经网络结构 ,并设计了相应的优化学习算法 ,可以简化网络并加速收敛。该系统可用于实现自动化精密加工过程中的工件形状监控和刀具诊断 。

An image collecting and state monitoring system for the shape of turning workpiece realized by CCD video camera and optical maser is introduced here.And a wavelet- neural network used in image analyzing and recognizing along with its optimal learning algorithm are proposed,which can simplify the structure of network and accelerate the convergent velocity.The system can be used in shape monitoring for workpiece and cutting tool diagnosis in automatic precise processing.The validity of the theoretical algorithm is demonstrated by simulation results.