Abstract: A method for detecting the integrity of aluminum foil seals with high detection accuracy and high detection efficiency is proposed.First of all,in order to obtain the infrared image of the sealing part aluminum foil,the method has adopted the infrared imaging technology.Then，the infrared image is preprocessed and the feature is extracted by circular symmetric Gabor wavelet.Lastly,KNN is used to identify and classify the acquired imaging.Experiments show that the circular symmetric Gabor wavelet used by this sealing detection method has higher recognition rate and faster algorithm than the traditional Gabor transform algorithm, and provides technical assurance for the automatic screening and rejection of defective sealing products in the subsequent aluminum foil sealing production line.

Key words: Aluminum foil sealing detection, Gabor wavelet, Circular symmetric Gabor wavelet, K-nearest neighbor

摘要： 提出一种检测精度和检测效率高的铝箔封口密封完整性检测方法。首先，采用红外成像技术，获取铝箔封口部位的红外图像。然后，对所获的红外图像进行预处理，并以环形对称Gabor小波对其进行特征提取。最后，用最近邻分类器对所获图像进行识别与分类处理。实验证明，此密封性检测方法所运用的环形对称Gabor小波较传统的 Gabor变换的算法识别率高、速度快，为后续的铝箔封口生产线中的密封不良产品的自动筛选、剔除提供了技术保证。

关键词: 铝箔密封性检测, Gabor小波, 环形对称Gabor小波, 最近邻分类器