Applied study of VFD recognition based on computer vision

VFD (vacuum fluorescent display) is commonly used in stereos, carradio and in other electronic equipment. Then exactly recognizing the characteristics of VFD is significant. A novelty computer vision system based on artificial neural network for VFD inspection is presented in this paper. The system includes hardware and software; camera and image acquisition board can acquire the on-line image of VFD. The real-time image can be pre-processed; the image features can be extracted by certain image-processing algorithm. BP (back propagation) neural network is used to recognize the characteristics on VFD board, the neural network can be trained by sample and teacher signals off-line, and then it can recognize character on VFD image. Simulation results show the validity and effectiveness.