The use of AI has started to become more and more common in everyday life, from using advanced AI for parallel parking and self-driving cars to the use of ChatGPT for writing letters and emails to others. Due to the free access nature of many of these services more and more people are starting to use AI technologies for their everyday tasks. With the advent of AI technologies in the everyday lives of people, new and improved techniques have also started to emerge to tackle various uncertainties.
The introduction of AI into the field of forensic science was inevitable as with any other techniques. From image processing to machine learning techniques, various AI subfields are being currently researched for use in the forensic field. Some of them are beyond science fiction novels and movies.
For example, recent research by the Federal Bureau of Investigation has found that it is possible to make automatic face recognition from skeletal remains that are completely decomposed [1]. Studies are trying to create a non-invasive post-mortem procedure by integrating computer vision with various other radiography techniques like CT scans [2]. No longer are the incision stitches or alternative incisions needed for post-mortem which need to be concealed from the beloved. Various imaging techniques like MRI, CT, X-ray, and ultrasound are used to retrieve the basic information about the patient, using AI technology an analysis of the patient can be done without any surgical invasive procedures. These advances can even diagnose mental disorders like dementia, Alzheimer’s disease, schizophrenia, and depression [3]. Even the field of ballistics is flourishing with the use of various techniques for the identification of firearms[4] and also for documenting the striation marks on the bullet for later comparison [5].
Another advancement in crime scene investigation is the automation of evidence collection. Crime scene reconstruction and documentation are time-consuming processes in which even the slightest negligence can be crucial to an investigation. Usually, there is huge manpower required for the proper documentation of the crime scene along with strong analytical skills for reconstructing the chain of events that have unfolded in the crime scene. Various studies have found multiple methods chained together with AI to document and recreate the crime scene. These include LiDAR (Light detection and ranging), AR (augmented reality), VR (virtual reality), and MR (mixed reality) [6]. This type of 3D documentation and recreation of the crime scene enables the complete documentation of the crime scene and later reusability in courts and during further investigation procedures [7]. In the future, even the chain custody will be more technology oriented like the use of RFID in documenting evidence. RFID will enable the easy tracking of evidence at the crime scene and even beyond it [8].
Soon it will be possible to have small flying drones and walking robots in a crime scene for evidence collection, analysis, documentation, and reconstruction of crime scenes with VR, AR, and MR. The evidence will be tracked with RFID and the paperwork will be soon changed into online web-based reports. Even the analysis will be with various AI and machine learning techniques incorporated. Allowing for a more efficient and high-resolution screening and testing of evidence. One day the crime scene itself can be projected in the court for the judges and jury to view.
Reference
P. Tu, R. Book, X. Liu, N. Krahnstoever, C. Adrian and P. Williams, "Automatic Face Recognition from Skeletal Remains," 2007 IEEE Conference on Computer Vision and Pattern Recognition, Minneapolis, MN, USA, 2007, pp. 1-7, doi:
10.1109/CVPR.2007.383060. keywords: {Face recognition;Image reconstruction;Skull;Layout;Law enforcement;Humans;Automation;Image databases;Spatial databases;Computed tomography},
Wan, L., Song, Y. X., Li, Z. D., Liu, N. G., Wang, Y. H., Wang, M. W., Zou, D. H., Huang, P., & Chen, Y. J. (2020). The approach of virtual autopsy (VIRTOPSY) by postmortem multi-slice computed tomography (PMCT) in China for forensic pathology.
Forensic Imaging, 20, 200361. https://doi.org/10.1016/j.fri.2020.200361
Sullivan, S. O., Holzinger, A., Zatloukal, K., Saldiva, P., Sajid, M. I., & Wichmann, D.
(2017). Machine Learning enhanced virtual Autopsy. Autopsy and Case Reports, 7(4),
3–7. https://doi.org/10.4322/acr.2017.037
Li, D. (2006). Ballistics projectile image analysis for firearm identification. IEEE
Transactions on Image Processing, 15(10), 2857–2865. https://doi.org/10.1109/tip.2006.877530
Handique, S., Saha, S., Suresh, R., & Mahanta, L. B. (2024). Development of an ai-enabled video capturing device for bullet trajectory analysis and ballistic research.
ITEGAM- Journal of Engineering and Technology for Industrial Applications
(ITEGAM-JETIA), 10(47). https://doi.org/10.5935/jetia.v10i47.1105
Maneli, M. A., & Isafiade, O. E. (2022). 3D Forensic crime scene Reconstruction
Involving Immersive Technology: A Systematic Literature review. IEEE Access, 10,
88821–88857. https://doi.org/10.1109/access.2022.3199437
Villa, C., Lynnerup, N., & Jacobsen, C. (2023). A virtual, 3D multimodal approach to victim and crime scene reconstruction. Diagnostics, 13(17), 2764. https://doi.org/10.3390/diagnostics13172764
Bolic, M., Borisenko, A., & Seguin, P. (2012). Automating evidence collection at the crime scene using RFID technology for CBRN events. Forensic Science Policy &
Management an International Journal, 3(1), 3–11. https://doi.org/10.1080/19409044.2012.710294
Written By:
Mr. Atul Raj
Assistant Professor
Forensic Science Department
Harsha Institute of Management Studies