Image Retrieval to Identify Anonymous Corpses Reported from ID Card Photos Using Face Recognition
Article Sidebar
Main Article Content
Abstract
The problem of missing persons and the identification of anonymous corpses remains a major challenge in the Thai justice system due to fragmented data sources and lengthy verification processes. This study aims to develop an image retrieval system from national ID card photographs using face recognition technology to enhance the efficiency of personal identification. The evaluation was conducted under two principal approaches: (1) image retrieval in a 1:N setting and (2) person verification in a 1:1 setting. The experimental results for image retrieval (1:N) show that the system achieved a Top-1 Accuracy of 86.59% and a Top-5 Accuracy of 98.0%, with an average response time of only 2.12 seconds per query—nearly 100 times faster than human experts, who required about 3.5 minutes per case. Although the experts’ highest accuracy (≈95.8%) remains superior, the proposed system proved highly effective as a preliminary screening tool. For the person verification (1:1) task, testing with similarity scores on the ID-All-99 dataset yielded an ROC–AUC of 0.706, reflecting the challenges posed by cross-pose and cross-quality variations. In comparison, evaluations under standardized test conditions produced a significantly higher ROC–AUC of 0.92, further emphasizing the system’s sensitivity to input image quality. In summary, the proposed system demonstrates strong potential for integration into forensic workflows, particularly in urgent scenarios such as locating missing persons or identifying disaster victims. However, practical deployment requires well-defined preprocessing procedures and carefully calibrated threshold settings to ensure consistent accuracy and reliability in real-world environments.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
เนื้อหาและข้อมูลในบทความที่ลงตีพิมพ์ใน วารสารวิชาการอาชญาวิทยาและนิติวิทยาศาสตร์ โรงเรียนนายร้อยตำรวจ ถิอว่าเป็นข้อคิดเห็นและความรั้บผิดชอบของผู้เขียนบทความโดยตรงซึ่งกองบรรณาธิการวารสาร ไม่จำเป็นต้องเห็นด้วยหรือรับผิดชอบใดๆ
บทความ ข้อมูล เนื้อหา รูปภาพ ฯลฯ ที่ได้รับการตีพิมพ์ใน วารสารวิชาการอาชญาวิทยาและนิติวิทยาศาสตร์ ถือว่าเป็นลิขสิทธิ์ของวารสาร วารสารวิชาการอาชญาวิทยาและนิติวิทยาศาสตร์ หากบุคคลหรือหน่วยงานใดต้องการนำทั้งหมดหรือส่วนหนึ่งส่วนใดไปเผยแพร่ต่อหรือเพื่อกระทำการใดๆ จะต้องได้รับอนุญาตเป็นลายลักษณ์อักษรจาก วารสารวิชาการอาชญาวิทยาและนิติวิทยาศาสตร์ ก่อนเท่านั้น
References
Albiero, V., Oliveira, L. S., & Bowyer, K. W. (2020). Analysis of gender effects on face recognition accuracy. IEEE Transactions on Information Forensics and Security, 15, 260–271. https://doi.org/10.1109/TIFS.2019.2927108
Ashton, J. (2009). Digital forensics and cyber crime. Wiley.
Caplova, A., Ferrara, M., Franco, A., Maltoni, D., & Rossi, M. (2017). Cross-quality face recognition: New challenges. Proceedings of the International Conference on Biometrics (ICB), 9–16.
Casey, E. (2019). Digital evidence and computer crime: Forensic science, computers and the Internet (4th ed.). Academic Press.
Cavazos, J. G., et al. (2021). Demographic bias in deep face recognition models. IEEE Access, 9, 79180–79190. https://doi.org/10.1109/ACCESS.2021.3084230
Deng, J., Guo, J., Xue, N., & Zafeiriou, S. (2019). ArcFace: Additive angular margin loss for deep face recognition. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 4685–4694). https://doi.org/10.1109/CVPR.2019.00482
Dinis-Oliveira, R. J., & Magalhães, T. (2016). Forensic science and the revolution in crime investigation. Journal of Forensic Research, 7(4), 1–2. https://doi.org/10.4172/2157-7145.1000e126
Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT Press.
Grother, P., Ngan, M., & Hanaoka, K. (2019). Face Recognition Vendor Test (FRVT): Part 3—Demographic effects (NISTIR 8280). National Institute of Standards and Technology.
He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. IEEE Conference on Computer Vision and Pattern Recognition, 770–778.
Institute of Forensic Medicine, Police General Hospital. (2025). Anonymous corpse statistics report 2019–2024 [Internal document]. Institute of Forensic Medicine, Police General Hospital. (In Thai).
Jain, A. K., Ross, A. A., & Nandakumar, K. (2011). Introduction to biometrics. Springer.
Lee, J. D., & See, K. A. (2004). Trust in automation: Designing for appropriate reliance. Human Factors, 46(1), 50–80. https://doi.org/10.1518/hfes.46.1.50.30392
LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521, 436–444.
Mayer, J. (2019). Forensic similarity: A new paradigm in face recognition evidence. Journal of Forensic Sciences, 64(4), 1010–1021.
Michalski, R. (2024). AI-assisted identification methods in mass fatality incidents. Forensic Science International, 356, 111789.
Office of the Personal Data Protection Committee. (2022). Handbook of the Personal Data Protection Act B.E. 2562 (2019). Office of the Personal Data Protection Committee. https://www.pdpc.go.th (In Thai).
Parasuraman, R., & Riley, V. (1997). Humans and automation: Use, misuse, disuse, and abuse. Human Factors, 39(2), 230–253. https://doi.org/10.1518/001872097778543886
Phillips, P. J., Jiang, F., Narvekar, A., Ayyad, J., & O'Toole, A. J. (2018). An analysis of face recognition accuracy by demographic groups and face image quality. NIST Technical Report.
Piotrowski, A. (2025). Machine learning tools in DVI operations: A global perspective. Journal of Forensic Identification, 75(2), 102–118.
Raposo, V. L. (2024). Automated facial recognition and privacy: Legal and ethical concerns. Computer Law & Security Review, 49, 105924.
Schroff, F., Kalenichenko, D., & Philbin, J. (2015). FaceNet: A unified embedding for face recognition and clustering. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 815–823. https://doi.org/10.1109/CVPR.2015.7298682
Schouten, R., & Jacobs, B. (2013). Humanitarian forensic action: A new model. International Review of the Red Cross, 95(890), 651–665. https://doi.org/10.1017/S1816383114000168
Taigman, Y., Yang, M., Ranzato, M., & Wolf, L. (2014). DeepFace: Closing the gap to human-level performance in face verification. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 1701–1708. https://doi.org/10.1109/CVPR.2014.220
Thailand. Ministry of Digital Economy and Society. (2022). Personal Data Protection Act B.E. 2562: Implementation guideline. (In Thai).
UNODC. (2019). Guidelines on the identification of missing persons. United Nations Office on Drugs and Crime.
UN/OHCHR. (2010). Guidelines on human rights and the use of biometrics. United Nations.
