A bottle finish inspect method based on fuzzy support vector machines and wavelet transform

Bottle finish chinks are regarded as the most critical defects of bottles adopted in food and beverage production. This paper presents a bottle finish inspecting methods. An algorithm based on probability and statistic is applied for determination of region of interest. For extracting feature, the circular law is used to carry on the scanning, and then the 1D wavelet transform is performed. The fuzzy support vector machines are used as classifiers, which synthesize the fuzzy theory and SVMs. At the same time an optimization algorithm based on genetic algorithm is proposed to confirm the parameters of the fuzzy support vector machines in this paper, and it ensure that the fuzzy SVMs have good classification ability. The proposed algorithms are proved to be efficient in automatic finish inspection.