Digital Forensic Based Object Recognition for Enhanced Crime Scene Interpretation


Vikash Kumar Singh1, Durga Sivashankar2, Siddharth Sriram3, Manish Nagpal4, Warish Patel5, Shweta Loonkar6

 

1 Principal Software Engineer, IT, Societe Generale, India

2 Technical Lead, IT, Siemens Healthineers, India

3 Centre of Research Impact and Outcome, Chitkara University, Rajpura- 140417, Punjab, India

4 Chitkara Centre for Research and Development, Chitkara University, Himachal Pradesh-174103 India

5 Associate Professor, Department of Computer Science and Engineering, Parul Institute of Engineering and Technology, Parul University, Vadodara, Gujarat, India

6 Assistant Professor, Department of ISME, ATLAS SkillTech University, Mumbai, Maharashtra, India

Emails: vikashsd@gmail.com; durga.sivashankar@yahoo.com; siddharth.sriram.orp@chitkara.edu.in; manish.nagpal.orp@chitkara.edu.in; warishkumar.patel@paruluniversity.ac.in; shweta.loonkar@atlasuniversity.edu.in

 

Corresponding Author: Vikash Kumar Singh, Email- vikashsd@gmail.com


 


Text Box: Abstract This research introduces a novel and comprehensive framework for digital forensics-based crime scene interpretation. The proposed framework comprises five algorithms, each serving a distinct purpose in enhancing image quality, extracting features, matching, and constructing a database, recognizing, and reconstructing objects in 3D, and conducting context-aware analysis. An ablation study validates the necessity of each algorithmic step. The framework consistently outperforms existing methods in terms of accuracy, precision, recall, and processing time. A detailed comparative analysis of parameters further highlights its cost-effectiveness, moderate complexity, superior data integration, and scalability. Visualizations underscore its dominance across multiple metrics and parameters, positioning it as an advanced solution for digital forensic-based object recognition in crime scene interpretation.


Received: September 18, 2023 Revised: February 02, 2024 Accepted: June 17, 2024

Top of Form

 

Keywords: Digital forensics; Crime scene interpretation; Object recognition; Preprocessing; Feature extraction; Database construction; 3D reconstruction; Context-aware analysis; Decision support; Performance comparison.