Related Experiment Video
Updated: Jun 24, 2026
06:25
Motion-Acuity Test for Visual Field Acuity Measurement with Motion-Defined Shapes
Published on: February 23, 2024
Individual camera identification using correlation of fixed pattern noise in image sensors
Kenji Kurosawa1, Kenro Kuroki, Norimitsu Akiba
1Physics Section, Second Forensic Science Division, National Research Institute of Police Science, 6-3-1, Kashiwanoha, Kashiwa, Chiba 277-0882, Japan. kurosawa@nrips.go.jp
Journal of Forensic Sciences
|March 24, 2009
Summary
This study demonstrates that fixed pattern noise (FPN) in image sensors can identify individual video cameras. Analyzing the correlation coefficient of FPN allows for precise camera identification, with more frames improving accuracy.
Area of Science:
- Digital Image Forensics
- Sensor Technology
- Signal Processing
Background:
- Individual video camera identification is crucial for forensic analysis.
- Fixed Pattern Noise (FPN) is an intrinsic characteristic of image sensors.
- Previous methods for camera identification have limitations.
Purpose of the Study:
- To investigate the feasibility of using FPN correlation for individual CCD camera identification.
- To assess the impact of temporal changes and frame integration on identification accuracy.
Main Methods:
- Examined five identical color Charge-Coupled Device (CCD) modules.
- Captured 100 frames per module using a 12-bit monochrome video capture board.
- Integrated frames to obtain FPN and calculated normalized correlation coefficients.
Main Results:
- Successfully distinguished individual CCD modules based on FPN correlation.
- Temporal variations in correlation coefficients showed negligible impact on identification.
- A positive correlation was observed between the number of integrated frames and identification accuracy.
Conclusions:
- FPN analysis provides a reliable method for individual camera identification.
- Acquiring a larger number of frames enhances the precision of camera identification.
- This technique offers a robust solution for forensic video analysis.
