Low-Altitude Infrared Slow-Moving Small Target Detection via Spatial-Temporal Features Measure

Area of Science:

• Computer Vision
• Infrared Imaging
• Target Detection

Background:

• Robust detection of infrared slow-moving small targets is critical for infrared search and tracking (IRST) applications.
• Existing methods struggle with missed detections and false alarms in complex low-altitude infrared scenes.

Purpose of the Study:

• To propose a novel algorithm for detecting low-altitude, slow-moving small targets in infrared imagery.
• To improve detection accuracy by addressing challenges posed by complex backgrounds and target dimness.

Main Methods:

• A spatial-temporal features measure (STFM) algorithm is developed.
• It incorporates a circular kernel for local grayscale difference (LGD) to suppress background noise.
• Short-term energy aggregation (SEA) and long-term trajectory continuity (LTC) mechanisms enhance target detection and reduce false alarms.

Main Results:

• The proposed STFM algorithm integrates spatial and temporal features effectively.
• Experimental results on diverse infrared sequences demonstrate superior performance compared to state-of-the-art methods.
• The method shows significant improvements in detecting dim and small targets in challenging environments.

Conclusions:

• The STFM algorithm offers a robust solution for infrared small target detection in complex low-altitude scenarios.
• The combination of LGD, SEA, and LTC effectively mitigates common detection issues.
• This approach enhances the reliability of IRST systems for security and guidance applications.