Abstract <p>In the digital documentation identity era documents are most commonly shared and used for the authentication of the identity of a person. The various tools are used to create or produce fake identity documents. In the current scenario identification of forged or tampered document is the challenging task. In this paper, we proposed an identity document forgery detection technique with the help of the Haralick texture feature. The proposed system is organized into two sections document digitization and document verification. In the document digitization section first, convert the document image into binary and locate the template header and footer part. Eliminate the header and footer of the original image Then convert to gray level co-occurrence matrix (GLCM). The Haralick texture attributes are fetched from each block, find the sum and if it is greater than 0 then store it to the knowledge base. In the document verification section locate the header and footer template from the verification image after converting it to binary then convert to GLCM. The sum of Haralick texture features are calculated and predicted using probabilistic neural network (PNN). Finally based on the maximum occurrence of the blocks and the percentage of occurrence will display whether the document is original or forged. The result of the proposed system shows that it gives good accuracy as compared with the other systems.</p>