CORRECTION OF NON-UNIFORMITY OF RESPONSE IN SENSOR ARRAYS

摘要

A microbolometer comprises an array of sensor elements or 'pixels' each individually responsive to generate an output indicative of the intensity of infrared radiation incident thereupon. In practice there are significant non-uniformities in the response of each pixel and further processing of the outputs must take place to generate an accurate image. These non-uniformities are dealt with by means of individual correction factors to be applied to the outputs of each pixel in the image. Commonly, the correction used provides a constant offset (1pt) and a linear gain correction (2pt). As the optimal values of these corrections drifts over time it is still necessary to provide further correction. In the present invention this is achieved by calculating one or more common scalar correction parameters that when applied to all of the individual correction factors before they are used to correct the output of the sensor elements will minimize the differences between the corrected outputs of adjacent sensor elements in the array when summed over a particular selection of sensor elements in the array.