New Results on Small and Dim Infrared Target Detection

Area of Science:

• Computer Vision
• Infrared Imaging
• Aerospace Engineering

Background:

• Real-time small infrared (IR) target detection is crucial for aircraft situational awareness.
• Existing IR detection systems struggle with hardware constraints and performance in cluttered backgrounds, especially with clouds.

Purpose of the Study:

• To develop an efficient and robust IR target detection method for high-altitude aircraft.
• To address the limitations of current systems in challenging environmental conditions.

Main Methods:

• An efficient method involving candidate target extraction in pre-processing.
• Fusion of local scale, blob-based contrast maps, and gradient maps for detection.
• Development and use of mid-wave infrared (MWIR) and long-wave infrared (LWIR) cameras for data collection and validation.

Main Results:

• The proposed method achieves high detection accuracy, with mean AUC at least 22.3% higher than comparable methods.
• Demonstrated superior performance on both public datasets and custom-acquired image sequences.
• Significantly reduced computational cost compared to existing approaches.

Conclusions:

• The developed IR target detection method offers a robust and efficient solution for aircraft situational awareness.
• The fusion strategy effectively enhances detection in complex backgrounds.
• The system shows promise for real-world deployment in demanding environments.