Research of Printed Matter Flaws Inspection Based on Improved K-Means and PCA

This paper presents a new approach for inspection of printed matter flaws based on K-mean clustering (KM) and principal component analysis (PCA). PCA is a method that can transform the original data that contains more vectors and some different correlative relationships between these vectors into a new one that contains fewer vectors and disrelated relationships between these vectors, while keeping the most information of the original data. First the PCA is used to obtain the best description features over the entire image which can reduce the dimension of the image and numeration, and then the reduced image data is identified by K-means clustering. The algorithm have a strong global searching capacity and avoids the local minimum problems. At the same time, it´s no longer a large degree dependent on the initialization values. The results show that the method is more effective.