Document Forgery Detection

Abstract

Nowadays, due to the easy availability of cheap electronic devices in the market, anyone can get these devices very easily with a very low cost, on the other hand people are making forged documents with the help of these devices with no problem. Printers and scanners are the two top list electronic devices which are being used to make forged documents. Most of the organizations where a lot of paper work is processed on daily bases face these type of problems very frequently and for the purpose of forgery detection these organizations need help from document forgery detection laboratories for the verification and testing of these documents. DFDS will help these organizations to detect forgery of these documents based on source printer identification. Scanned images from five different printers are given to the classifier (SVM) in DFDS to train it gainst these specified printers, then a test image is given to the classifier for testing purpose. Experimental results show that the techniques and algorithms which are used in DFDS has an average accuracy of more than 90%.