Method and Apparatus for the Correction of a Non-Uniform Sensitivity of Detector Elements in Thermographic Cameras

A method for correcting non-uniform sensitivity of detector elements in a thermographic camera comprising:

• recording background frames for internal reference during shutter states corresponding to a beam path of the thermographic camera being closed,
• determining a function describing variations between the background frames for the shutter phases, and
• generating an updated background frame based on the function approximating and describing a long-term drift of the values of at least two background frames recorded during at least one shutter state, the long-term drift being determined from variation in values of individual detector pixels,
• assigning each image frame to be corrected a counting index within a sequence of image frames following an end of a prior shutter phase,
• using the function to assign a correction value to every image frame based on its counting index, and
• correcting at least one image frame using the updated background frame.

A thermographic camera comprising:

• a shutter unit in a beam path of the thermographic camera for electively opening and closing the beam path,
• a pixel-based detector unit comprising detector elements for pixel-by-pixel detection of thermal radiation, and
• a processor unit for pixel-by-pixel processing, storage and providing data from the detector unit,
• wherein the detector unit is capable of recording background frames during a shutter state when the shutter unit is closed,
• wherein the detector unit is capable of recording image frames during a shutter state when the shutter unit is open,
• wherein the processor unit is capable of generating an updated background frame from a newly recorded background frame and a background frame to be updated by adding a value of each pixel of the newly recorded background frame multiplied by a first factor and a value of each pixel of the background frame to be updated multiplied by a second factor, wherein the first factor is a value between zero and one, and the second factor is the difference between one and the first factor, so that a value of each pixel of the updated background frame is an accumulated weighted average, and
• wherein the processor unit is capable of generating at least one output image by applying said updated background frame to at least one image frame of the detector unit.