Auto-minimum resolvable temperature difference method for thermal imagers

Electro-optical industries worldwide are producing thousands of thermal imagers based on transfer of technology from design agencies. Evaluation of a thermal imager is necessary prior to their effective field deployment. Traditionally, subjective minimum resolvable temperature difference (MRTD) is a parameter which is employed to evaluate the quality of an IR image. But due to human intervention, the usage of this parameter becomes time-consuming and slows down the production process. In the presented work, an accurate auto-MRTD method is proposed, which increases the efficiency of testing while minimizing human dependency cost. This is also a case study of how theory of objective or auto-MRTD can successfully be implemented in production facilities to increase their efficiency. Auto-MRTD method is described by measuring the modulation transfer function and noise-equivalent temperature difference of the thermal imager. Thereafter, calibration constant c ( f ) is computed, which can be used for faster MRTD computation. Proper experimentation was carried out to compute both auto-MRTD and subjective MRTD. Quantitative analysis and percentage error were carried out to evaluate the effectiveness of this method. It is shown that percentage error using auto-MRTD is within 5% of that of the conventional subjective MRTD method.