Image analysis – Image enhancement or restoration – Intensity – brightness – contrast – or shading correction
Reexamination Certificate
1999-04-08
2003-07-08
Rogers, Scott (Department: 2624)
Image analysis
Image enhancement or restoration
Intensity, brightness, contrast, or shading correction
C382S299000, C348S164000, C348S251000, C250S347000, C250S252100
Reexamination Certificate
active
06591021
ABSTRACT:
FIELD OF THE INVENTION
The invention relates to a method for correcting the gray levels or pixel intensities of images-produced by a digital infrared camera having a two-dimensional infrared detector array, wherein the gray level value for each image pixel is corrected by a stored correction coefficient.
BACKGROUND INFORMATION
The German company AEG Infrarot Module GmbH (AIM) has been producing high grade infrared detectors and associated components such as coolers and output electronics since the 1970s. Such detectors are typically used in military, research, medical, and industrial applications. In older one-dimensional or line detectors, a one-dimensional line of photosensitive detector elements cooperated with a mechanical scanner so as to successively scan a scene line-by-line so as to generate a complete image therefrom. In the newest generation of detectors, the individual photosensitive elements are arranged in a two-dimensional array over a surface plane of the detector. Since the array of elements directly receives a complete image over the two-dimensional array plane, such two-dimensional detectors no longer need a mechanical scanner for generating a complete two-dimensional image. In other words, using a two-dimensional detector, a scene can be directly imaged onto a photosensitive array chip using a suitable optics system. The respective output signals of the individual photosensitive elements are serially read-out and then reconstructed into a two-dimensional data structure representing the corresponding image, by means of a suitable electronic image processing circuit and procedure.
Due to technical reasons and limitations, any two-dimensional infrared detector comprising a two-dimensional array of individual photosensitive elements necessarily exhibits inhomogeneities or non-uniformities among the several individual elements. These non-uniformities can include constant or invariable non-uniformities as well as time-varying non-uniformities. In any event, such non-uniformities cause imaging errors or aberrations in the electronic capture and reproduction of the true or actual image. This arises because a difference in the output signal of a given detector element relative to the outputs of the other elements would initially be regarded as arising from a feature of the real scene being viewed, even though such a difference might actually arise do to a non-uniformity of the respective detector element.
The problem of distinguishing between scene-based features and detector-based non-uniformities is most readily apparent when all of the detector elements are detecting a neutral uniform scene without any characteristic features. In such a case, it can be assumed that any difference in output signals of the several detector elements arises from detector-based non-uniformities. However, the imaging errors resulting from non-uniformities become especially problematic and difficult to recognize and correct when the detector is “seeing” or imaging a scene that includes features having different image pixel intensities, because then the detector elements will produce respective differing output signal intensities resulting from both the true differences in the scene as well as the non-uniformities of the individual detector elements. In this regard, features in the scene that move relative to the detector can be recognized as such real features, because the variation in pixel intensity associated with such a moving feature will move across several detector elements. On the other hand, stationary scene-based features will constantly affect the output intensity of a particular one or more detector elements of a stationary detector, and the output signals are therefore prone to be incorrectly interpreted as resulting from a detector-based non-uniformity in the respective affected element or elements.
European Patent Application 0,600,742, published on Jun. 8, 1994, and corresponding U.S. Pat. No. 5,323,334 (Meyers et al.), issued on Jun. 21, 1994 discuss the above mentioned problem relating to the recognition and correction of imaging errors resulting from non-uniformities among the individual photosensitive elements of a detector or sensor array. The entire disclosure of U.S. Pat. No. 5,323,334 is incorporated into the present application by reference, excepting any subject matter that may be further incorporated by reference into U.S. Pat. No. 5,323,334.
EP 0,600,742 and U.S. Pat. No. 5,323,334 also both disclose a system for carrying out such correction or suppression of imaging errors resulting from sensor element non-uniformities. In the known system, the sensor or detector array is mounted on a sensor positioner, which physically moves the sensor relative to the incident image of the external scene falling on the sensor. The sensor is physically moved in order to distinguish between detector-based non-uniformities on the one hand and features in the real scene being viewed by the sensor on the other hand. As the sensor is moved by the sensor positioner, a sensor-based non-uniformity will move with the sensor, i.e. will always remain associated with a particular element of the sensor, while a real feature in the scene will move relative to the sensor, i.e. will successively fall on different elements of the sensor (unless the feature happens to follow exactly the same motion as the sensor, which can be avoided by proper choice of the motion pattern). A non-uniformity compensator is the used to recognize and correct or compensate the detector-based non-uniformities by processing the output signal of the sensor. Thereafter, a position corrector circuit electronically corrects the output signals by the inverse or opposite of the physical motion of the sensor, so that the image information is electronically shifted back to its true position, i.e. to counteract the physical motion of the sensor.
While the known system of the above discussed references uses a valid concept or theory for recognizing and correcting sensor-based non-uniformities, problems arise in connection with the physical movement of the sensor itself. The mechanical actuators, gimbals and the like used for mounting and moving the sensor are subject to wear as well as mechanical tolerances and inaccuracies. Also, especially for larger sized sensor arrays, the size and mass of the sensor array makes it difficult to accurately and rapidly physically move the entire array.
German Patent 197 15 983, published on Sep. 24, 1998 and corresponding U.S. patent application Ser. No. 09/061,147, which is cross-referenced and incorporated by reference herein, are not prior art relative to the present application. Instead, the present application represents a further development that is preferably used in combination with the method and apparatus for correcting the gray levels of images produced by a digital infrared camera by means of electronic signal processing as disclosed in German Patent 197 15 983 and corresponding U.S. Pat. application Ser. No. 09/061,147.
In the system according to German Patent 197 15 983 and U.S. patent application Ser. No. 09/061,147, a signal processing arrangement includes a memory in which respective correction coefficients K for each detector element j of the detector are stored, and the associated signal processing method comprises the following steps. First, the respective gray levels or pixel intensities Uj of a scene are acquired as image data by the detector. Next, the gray levels Uj of the image are corrected through use of the correction coefficients K in the image processing system, and the resulting corrected gray levels Uke are intermediately stored. Then the corrected gray levels UkG are filtered through a locally effective adaptive filter M, to produce filtered corrected gray levels F(Ukj). The remaining image error or aberration ej is then determined by using the gray levels Ukj of the unfiltered corrected image and the gray levels F(Ukj) of the filtered corrected image. Then the correction coefficients K are progressively updated or improved in connection with this determine
Breiter Rainer
Cabanski Wolfgang
Mauk Karl-Heinz
AEG Infrarot-Module GmbH
Fasse W. F.
Fasse W. G.
Rogers Scott
LandOfFree
Method and apparatus for correcting the gray levels of... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for correcting the gray levels of..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for correcting the gray levels of... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3001502