Automatic inspection of wave soldered joints using neural networks

The problem of solder joint inspection is viewed here as a two-step process of pattern recognition and classification. A modified Intelligent Histogram Regrading (IHR) technique is used that divides the gray-level histogram of the captured image from a joint into different modes. Each distinct mode is identified and the corresponding range of gray levels is separated and regraded by employing neural networks. The output pattern of the networks is presented to a second stage of the neural networks to select and interpret a histogram's features. The back-propagation algorithm is employed to train the neural networks in the second stage. After training, the neural network is employed to successfully identify and classify the defective solder joints. The identification and classification capabilities based on an adaptive learning algorithm are the major new features that are not present in existing systems.