Methods for Forensic Examination of Digital Images of Documents

Based on the approach involving the use of mathematical algorithms for information processing, this article puts forward proposals aimed at supplementing the classification of methods of forensic document examination. In particular, it is proposed to consider the following bases for classifying such methods:

By the information type displayed in a document converted into digital form (digital image of a document): physical-optical methods, frequency-statistical methods, structural-geometric methods, semantic (neural network) methods, clustering methods, metadata analysis methods.

By the analysis level of the information studied, displayed in a document converted into digital form (digital image of a document): low-level methods, medium-level methods, high-level methods. For each of the proposed classification bases the conditions are specified that allow a particular method to be assigned to a specific group. These conditions include, in particular: physical properties of the sensor, photosensitivity indicators, spectrum distribution analysis, object boundary analysis, use of neural networks and other artificial intelligence tools, data clustering features, file service information, etc. The classification of methods presented herein creates the scientific prerequisites for forensic examination of documents in digital form and expands the expert capabilities for solving specific tasks that arise in the course of document examination.

1. Moiseeva T.F. Natural Science Methods of Forensic Examinations. Course of Lectures. Moscow: RGUP, 2015. 196 p. (In Russ.).

2. Technical and Criminalistic Examination of Documents. Textbook / A.A. Protkin (ed.). Moscow: Yurlitinform, 2015. 296 p. (In Russ.).

3. Rossinskaya E.R. Neural Networks in Forensic Expertise and Expert Practice: Problems and Prospects. Bulletin of the Kutafin University (MSAL). 2024. No. 3 (115). P. 21–33. (In Russ.).

4. Chesnokova E.V., Usov A.I., Omelyanyuk G.G., Nikulina M.V. Artificial Intelligence in Forensic Expertology. Theory and Practice of Forensic Science. 2023. Vol. 18. No. 3. P. 60–77. (In Russ.). https://doi.org/10.30764/1819-2785-2023-3-60-77

5. Rossinskaya E.R. The System of Forensic Activity Digitalization Theory. Theory and Practice of Forensic Science. 2024. No. 19. No. 3. P. 20–32. (In Russ.). https://doi.org/10.30764/1819-2785-2024-3-20-32

6. Gonzalez R., Woods R. Digital Image Processing. 3rd ed. Moscow: Tekhnosfera, 2012. 1104 p. (In Russ.).

7. Yane B. Digital Image Processing. Moscow: Tekhnosfera, 2007. 584 p. (In Russ.).

8. Vizilter Yu.V., Zheltov S.Yu., Bondarenko A.V., Ososkov M.V., Morzhin A.V. Machine Vision Issues in Image Processing and Analysis: Lectures and Practical Training Course. Moscow: Fizmatkniga, 2010. 672 p. (In Russ.).

9. Vizilter Yu.V., Zheltov S.Yu., Knyaz V.A., Khodarev A.N., Morzhin A.V. Processing and Analysis of Digital Images with Examples on LabVIEW IMAQ Vision. Moscow: DMK Press, 2007. 464 p. (In Russ.).

10. Raković D. Error Level Analysis (ELA). Tehnika. 2023. Vol. 78. P. 445–451. https://doi.org/10.5937/tehnika2304445R

11. Zhou P., Han X., Morariu V.I., Davis L.S. Learning Rich Features for Image Manipulation Detection. ArXiv. 2018. https://doi.org/10.48550/arXiv.1805.04953

12. Sun J., Seung-Wook K., Sang-Won L., Sung-Jea K. A Novel Contrast Enhancement Forensics Based on Convolutional Neural Networks. Signal Processing: Image Communication. 2018. No. 63. P. 149–160. https://doi.org/10.1016/j.image.2018.02.001

13. Mishchuk V.A. Using Convolutional Block Attention Module for Interpreting the Results of Artificial Neural Networks Operation in Handwritten Signature Identification. Theory and Practice of Forensic Science. 2025. Vol. 20. No. 2. P. 65–81. (In Russ.). https://doi.org/10.30764/1819-2785-2025-2-65-81

14. Kupin A.F., Shpindler Y.V. Record Identification Program (PRIZ). Registration Certificate of the Computer Program RU 2025617305, 25.03.2025. (In Russ.).

15. Kupin A.F. Program to Establish the Content of the Written (PERSONA). Registration Certificate of the Computer Program RU 2025661243, 05.05.2025. (In Russ.).