Cut Paste Detection in Document Images Using Neural Network

To manipulate and modify digital images are very easy due to rapid advances of image processing software. So, to judge the authenticity of a given image is very difficult for a viewer. Many documents are created by Cut-And-Paste (CAP) of existing documents. In this thesis, we proposed a novel technique to detect CAP in document images using Neural Network. This can help in detecting unethical CAP in document of image collections. Recognition free and scalable is the solution to large collection of documents. The formulation is also independent of the imaging process (camera based or scanner based) and does not use any language specific information for matching across documents. After that, we model the solution as finding a mixture of homo-graphics, and design a linear programming (LP) based solution to compute the same. The proposed method is presently limited by the fact that we do not support detection of CAP in documents formed by editing of the textual content. The proposed results demonstrate that without loss of generality (i.e. without assuming the number of source documents), it can be correctly detect and match the CAP content in a questioned document image by simultaneously comparing with large number of images in the database. For the implementation of this proposed work we use the Image Processing Toolbox under MATLAB software.