Factors that Influence PRNU-Based Camera-Identification via Videos

Area of Science:

• Digital Image Forensics
• Sensor Pattern Noise Analysis

Background:

• The Photo Response Non-Uniformity (PRNU) pattern serves as a unique digital fingerprint for cameras, enabling image source identification and camera matching.
• PRNU patterns are susceptible to various external factors, including camera settings, which can impact their reliability.

Purpose of the Study:

• To investigate the influence of resolution, video length, and compression on the reliability of PRNU pattern-based image and video comparisons.
• To evaluate the effectiveness of computation and comparison methods under varying conditions.

Main Methods:

• Utilized a single iPhone 6 to capture images and videos under controlled variations of resolution, video length, and compression levels.
• Analyzed and compared PRNU patterns derived from these captured media using established forensic techniques.

Main Results:

• Increased image resolution positively correlates with more reliable PRNU comparisons.
• Longer reference videos yield a more robust and accurate PRNU pattern.
• Compression, exemplified by Snapchat's algorithm, significantly degrades the correlation value of PRNU patterns, reducing comparison accuracy.

Conclusions:

• Camera settings like resolution and video length critically influence PRNU pattern reliability.
• Compression poses a significant challenge to PRNU-based forensic analysis.
• Further research is essential to quantify the individual impact of numerous factors on PRNU patterns for improved forensic applications.