Image analysis – Image sensing
Reexamination Certificate
1999-06-14
2002-04-23
Boudreau, Leo (Department: 2621)
Image analysis
Image sensing
C382S298000, C382S305000, C358S403000, C358S451000
Reexamination Certificate
active
06377714
ABSTRACT:
BACKGROUND OF THE INVENTION
The invention relates to a method for the archiving of master images, in particular of films, in which the master image is scanned and the detected image points are digitized, corrected and compressed, and to a device for carrying out the method.
Due to the finite durability of celluloid films and their complicated storage, there has long been a need for the image information to be archived simply-and permanently. There have also been many and various efforts to store the master images on other media.
The theoretical preconditions for loss-free digitization have long been known from telecommunication and information technology. It is necessary, for this purpose, to conform to Shannon's scanning theorem which states that scanning must be carried out at a scanning frequency at least twice as high as the highest signal frequency occurring, in order to avoid loss of information. Similar considerations also apply to the digitization of master images (Jähne, Bernd: Digitale Bildverarbeitung [Digital Image Processing], second edition 1991, Springer Verlag Berlin, pages 41-49). There is no appreciable problem in conforming to the scanning theorem in telecommunication and information technology, since the bandwidth of the useful signal is predetermined. In the case of master images, however, the resolution is usually not already known, so that there are considerable practical difficulties in conforming to the scanning theorem. If, by contrast, the master image is simply scanned with the maximum possible resolution in technical terms, the incidence of data is so great that, on the one hand, digitization takes up an inconceivable amount of time and, on the other hand, the storage space required becomes immense. As regards master images, therefore, the scanning theorem, in the widest sense, plays a part only if information on the master image is available beforehand.
With regard to the problems caused by the incidence of data, reference is made, for example, to WO 92/22 141, which discloses a data compression device having a hierarchical design. Here, a multiplicity of microprocessors are arranged at different levels, in each case the microprocessor of the higher level transmitting only some of the data to be compressed, so that, in particular, data records changing only little from master image to master image can be processed rapidly. Disadvantages of the known device are, besides the fact that loss-free archiving is not possible, the highly complicated design and the low speed of processing of rapidly changing master images.
JP 5-63912 A discloses a device for the digitization of master documents, in which digitization is carried out at a specific scanning rate. By means of the binary structure of the master document, aliasing effects can then be inferred relatively simply from the subscanning and the resolution can be increased until the aliasing effects have disappeared. In the case of more complex master images without previous information, such a procedure is ruled out, since the aliasing effects cannot be detected reliably.
In practice, therefore, a different technical direction has been adopted as regards master films, specifically optimized adaption to the reproduction medium at the expense of a loss of data.
EP 0,440,230 discloses, for example, a device, by means of which the image information of a cinema film is converted into a video signal for television. In this case, mechanical faults in the master image are also taken over. Specific faults are then subsequently detected during playback and are suppressed. Methods of this kind are known, for example, from DE 33 11 898 or DE 44 09 283. A disadvantage of this and of similar procedures is that conversion is tailored to television as the future medium, so that image information is lost from the master image during conversion. If the master image is subsequently destroyed, these faults can scarcely, or cannot at all, be reconstructed any more, so that the original master image can no longer be generated in full from the archive data.
SUMMARY OF THE INVENTION
The technical problem on which the invention is based is, therefore, to provide a method, and a device for carrying out the method, for the archiving of master images, in particular of films, so as to be free of loss.
Since the geometric and/or radiometric and/or spectral resolution of the respective master image is predetermined or previously detected and the respective resolution of the optical device is subsequently matched adaptively to the predetermined or previously detected respective resolution of the master image, in such a way that the master image is scanned with a resolution at least twice as high, it is possible for the master image to be scanned without any loss of information, and with pseudodata being avoided. This reduces both the time required for the scanning operation and the quantity of data to be stored. The partly cumulative and partly alternative formulation results from the fact that all the respective scanning theorems must actually be conformed to, in order to ensure loss-free archiving. It may also be envisaged, however, to have master images in which the infringement of one or more scanning theorems can be accepted without any loss of information. For example, in the case of master documents having a binary structure, the spectral and radiometric distribution may not be of any further use, so that corresponding adaptive matching is unnecessary. Moreover, instances may also be envisaged, in which adaptive matching of an individual resolution is unnecessary, without a substantial increase in the data quantity occurring. This may be the case, for example, when information is available beforehand, so that, if the resolution is permanently set, the scanning theorem is reliably conformed to, without the excess pseudodata assuming major importance. Further advantageous refinements can be gathered from the subclaims.
In this case, the maximum geometric resolution can be detected, for example, either from the autocorrelation function of the gray values or from the spectral distribution of the spatial frequencies with determination of the limit values.
In this case, the adaptive matching of the geometric resolution of the optical device may be carried out, for example, via macropixel formation of the photosensitive component and/or defocusing and/or compression. For the adaptive matching of the spectral resolution, the spectral sensitivity of the photosensitive component and/or the spectral luminous intensity of the radiation source may be modified. For the adaptive matching of the radiometric resolution, the amplification of the photosensitive component and/or the integration time and/or the irradiation intensity of the radiation source may be modified.
Advantageously, the photosensitive component is measured and correction values are determined, by means of which, for example, inhomogeneities of the radiation source, the edge decline of imaging optics and the sensitivity fluctuations of the photosensitive component itself can be corrected.
In order further to reduce the incidence of data, before data compression mechanical faults and/or other defects in the master image are detected and are corrected by means of adjacent image points and/or preceding and/or following master images or ignored. For example, a local deviation of the point spread function (PSF) with respect to the remaining master image may be utilized for detecting the edge of the mechanical faults in the master image and/or in the sound track of a master film, in which case aliasing effects are to be ruled out. In this case, the PSF of the remaining master image is predetermined on the basis of the known recording technique and the material of the master image. If, for example, there is a tear in a master image of a film sequence, the known methods would interpret this as a pronounced change from master image to master image and would store correspondingly large data quantities. By contrast, in the method according to the invention, this fault i
Driescher Hans
Eckardt Andreas
Hilbert Stefan
Reulke Ralf
Scheele Martin
Boudreau Leo
Cohen & Pontani, Lieberman & Pavane
Deutsches Zentrum für Luft-und Raumfahrt e.V.
Sherali Ishrat
LandOfFree
Method for archiving master images by digital image data... 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 for archiving master images by digital image data..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for archiving master images by digital image data... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2932798