Motion deblurring of infrared images from a microbolometer camera

Abstract The infrared image of a microbolometer camera can show significant blurring effects if the object is moving. The blurring mechanism of a microbolometer is different to that encountered with classical CCD and CMOS cameras. The electrical signal in the pixel of a microbolometer detector decays exponentially with a time constant of 10–15 ms; therefore, the moving object is mapped to more pixels, resulting in a blurred image. Because of this blurring effect, the recorded infrared radiation of the object is dispersed and the contrast of the object corrupted. In this paper it is shown how the image can be restored and the blurring eliminated. The point spread function of the microbolometer camera is determined and the impact of the blurring from objects of different sizes is investigated. In order to suppress the noise in the restoration, a Wiener filter is used and it is demonstrated how objects of different sizes can be restored. Examples are presented for objects moving linearly and/or rotating; furthermore, an application is described where this technique has been used for automated thermographical testing.