Automatic visual inspection and classification based on rough sets and neural network
Mengxin Li, Chengdong Wu, Yong Yue · 2004
In this paper, a novel visual inspection and classification technology based on rough sets and neural network algorithm is presented. The rough set algorithm of data classification is discussed. As a large quantity of ambiguous and redundant data can be removed effectively using rough set theory, training time of neural networks is further decreased and the classification accuracy is also improved. Combined with anti-disturbance of the neural network, the effectiveness of classification technology is performed for the defect inspection of wood veneer with its rapid classification capacity and high classification accuracy.