摘要： 以扫描仪获得的数字扫描图像为研究对象，提出一种基于图像内容和噪音特征的扫描仪源辨识方法，提取数字扫描图像中的颜色特征、质量特征和邻域预测特征，生成一个72维的特征向量以辨识扫描仪的来源，并借助支持向量机确定扫描仪的品牌或型号。实验结果表明，该方法具有较高的分类精度，并且在数字扫描图像被压缩或剪切的情况下均具有较好的鲁棒性。

关键词: 数字扫描图像, 扫描仪源辨识, 支持向量机, 特征提取

Abstract: This paper focuses on digital images which are acquired using scanners, and proposes a robust sources identification technology. The method is based on using color features, image quality features, and neighborhood prediction features of scanned image samples. These features are used to design source identification algorithm in the form of a feature vector. Using Support Vector Machine(SVM) to determine the brand or model of scanner used to capture each scanned image. Experimental results demonstrate that the proposed method can effectively identify the correct scanner brands/models with higher accuracy and has better robust, compared with previous methods.

Key words: digital scanning image, scanner source identification, Support Vector Machine(SVM), feature extraction

中图分类号: 

• TP301.6