Aluminum Foil Packaging Sealing Testing Method Based on Gabor Wavelet and ELM Neural Network
Yibang Zhou, Qimeng Liu, Yueteng Zhao, Weijun Li · 2018
In this paper, we study aluminum foil sealing temperature field distribution characteristics.we propose a secondary feature extraction and reduction algorithm based on RGB. The color features are extracted from the three primary colors and the thermal image samples are trained and classified by BP neural network. Aiming at the shortcomings of this method, this paper proposed a aluminum foil sealing test method by using Gabor wavelet to extract image texture features combined with ELM limit learning machine to training, classify and recognize the thermal image. Compare the training time and accuracy of the two training samples and test samples, and verify the superiority of the algorithm. Experimental results show that the algorithm based on Gabor wavelet and ELM neural network has the advantages of fast response, high precision and strong generalization, which can meet the requirements of aluminum foil sealing test.