Abstract: Quality control is one of the bottleneck problems hindering the development of food production. In this paper, a neural network pattern recognizer was used to recognize the typical patterns in statistical quality control(SQC). Experiments and computer simulations were conducted to evaluate the performance of the pattern recognizer. This research demonstrates the potential and procedure of applying computer vision and neural network pattern recognition techniques to food quality control. Performance evaluation indicates that back propagation neural network algorithm is effective for pattern recognition and can be applied to extrusion process successfully.