Modeling and Compensating Temperature-Dependent Non-Uniformity Noise in IR Microbolometer Cameras

Area of Science:

• Infrared Imaging Technology
• Image Processing
• Sensor Calibration

Background:

• Uncooled microbolometer-based infrared (IR) cameras suffer from spatial non-uniformity (NU) noise.
• This NU noise creates fixed patterns in images, and its intensity fluctuates with camera temperature instability.

Purpose of the Study:

• To develop a novel model and compensation algorithm for spatial NU noise and its temperature-dependent variations.
• To improve the accuracy and reliability of IR imaging systems.

Main Methods:

• A model separating NU noise into constant and dynamic (temperature-dependent) components was developed.
• Offline characterization of NU parameters using a black-body radiator and temperature samples.
• Online estimation of temperature-dependent variations using Hammerstein-Wiener and pixelwise least mean squares (LMS) estimators.

Main Results:

• The proposed algorithm effectively compensated for NU noise in long-wave IR camera images.
• Achieved a root mean square error of less than 0.25 °C.
• Demonstrated excellent NU correction performance despite a ~15 °C array temperature variation.

Conclusions:

• The novel model and algorithm provide effective compensation for spatial NU noise and its temperature-dependent fluctuations.
• This approach significantly enhances the quality of images from uncooled microbolometer-based IR cameras.