摘要： 采用计算机视觉原理与神经网络技术的自动化检测方法是计算机检测的新发展，具有非接触性、速度快、效率高、柔性好等优点，在现代产品质量检测中有着广泛的应用前景。该文介绍了基于神经网络与计算机视觉的产品质量检测系统的一般结构，阐述了这种系统的一个实例——玻璃瓶裂纹在线检测系统的实现方法。由于神经网络的应用，使得该检测系统具有良好的自学习、自适应能力，成功地实现了对生产线上玻璃瓶裂纹的快速、精确的检测。

关键词: 神经网络, 计算机视觉, 产品质量, 图像采集, 图像处理

Abstract: The automatic detection method adopting computer vision principle and artificial neural network technology is a new development in computer detection. It has lots of advantages such as no touch with inspected objects, high speed, high efficiency and good flexibility etc. So it has extensive applied foreground in the modern detection of product quality. This paper introduces the general structure of the product quality detection system based on neural network and computer vision, expounds the realization method of an example of this kind of system, namely the on-line crack detection system of glass bottles. The system’s ability to learn itself and adapt itself is very powerful due to the application of neural network. It has realized a quick and exact detection of on-line glass bottles’ crack successfully.

Key words: neural network, computer vision, product quality, image collecting, image processing

中图分类号: 

• TP39