An improved method to distinguish between scanned and rasterized document images

摘要

Office automation(OA)has evolved with the development of computer science,improving staff efficiency.Unstructured information processing is an important aspect of OA;therefore,in this paper,we propose an efficient method for distinguishing scanned and rasterized document images which can be used in this process.To ensure the efficiency and precision of our method,two steps are included:rapid processing and classification using noise features.In the first step,color,skew,and isolated noise features are used to identify the source of the images. In the second step,noise features are extracted from the input image and a support vector machine(SVM)classifier is used for classification.Our experiments show that our method has high precision and speed for distinguishing scanned and rasterized document images.