Facsimile and static presentation processing – Facsimile – Specific signal processing circuitry
Reexamination Certificate
1998-12-23
2001-06-19
Grant, II, Jerome (Department: 2624)
Facsimile and static presentation processing
Facsimile
Specific signal processing circuitry
C250S208100
Reexamination Certificate
active
06249358
ABSTRACT:
FIELD OF THE INVENTION
This invention relates generally to photographic film image scanners, and in particular to the use of selectively determined system response gain correction factors.
BACKGROUND OF THE INVENTION
In a typical scanning system for producing an electronic representation of visual material, the subject is illuminated by some light source and a resulting image is projected by a lens onto a photosensor (imager) to produce an electrical signal which is proportional to the brightness of the image. This electrical signal is typically converted to a digital representation for display and manipulation using a computer. For example, a charge coupled device (CCD) imager contains an array of light detecting sites (photosites) which accumulate charge depending on the light energy projected onto them. After some charge accumulation time, the charges in the photosites are transferred to a charge shifting structure so that the charges may be shifted out of the CCD and measured by some means in order to form an electronic representation of the image projected onto the CCD. Because of factors such as manufacturing variability in the CCD, dust or contaminants in the scanner optical imaging path, light source non-uniformity, or other source of variation, the system response for individual photosites may not be the same from site to site. Some means of compensation for this site to site variation is required so that the resulting electronic representation is not affected by the particular variation.
If the response of a given array of photosites to light energy is modeled as an offset representing the response when no light is present and as a proportionality constant which represents the effectiveness of the photosite when light is present, then that photosite may be normalized to a nominal responsiveness by subtracting the offset from the original signal and multiplying the result by the inverse of the proportionality constant. The inverse of the proportionality constant is referred to as gain. If an offset and gain are determined for each of the photosites in the imager array and if these offsets and gains are applied to the corresponding photosites each time the photosite outputs are measured or read, then the responsiveness of all the photosites will appear to be equal.
Typically, the system response for a given photosite does not change in the short term. Hence, the gain and offset values required to adjust the system response for a given photosite back to some ideal response can be determined by a calibration process at one time and then applied whenever the signal for that photosite is measured at some later time. A typical calibration process obtains samples of the system response for each photosite at two nominal signal input levels (using a white and a black card or using full illumination and dark, for example) at some nominal gain and offset values (typically 1 and 0, respectively) and then calculates the necessary gain and offset correction values for each of the photosites.
Offset correction is utilized in a practical scanner in order to account for inevitable small signals which arise even in the absence of light. The offset correction values are also referred to as dark offset correction. The offset correction values are subtracted on a pixel-by-pixel basis from the signal obtained from the image scanning system in order to make the result “zero based”, i.e., zero signal for zero light.
Gain correction is used in a scanner in order to account for inevitable variations in the light responsivity of the scanner on a pixel-by-pixel basis. Typically, the gain is the inverse (in a proportional sense) of the offset corrected signal from a uniformly white or gray image in the case of a reflection scanner, or a clear or minimum density (Dmin) piece of film in the case of a transmission scanner. Assuming that the system responds in a straight line fashion (i.e., the proportionality between the offset corrected signal and the illumination level is the same at all illumination levels), then multiplying each offset corrected pixel by its associated gain will normalize all the pixel responses to a common value.
The concepts and practices of determining gains and offsets on a pixel-by-pixel basis during calibration phase and employing those gains and offset during a later scanning phase in order to compensate for pixel-by-pixel variations in system responsiveness is well known in the art. U.S. Pat. No. 5,563,723—Beaulieu et al describes an approach to determining pixel-by-pixel gains and offsets, for example.
In the case of a negative film scanner, it is common practice to collect scan data without film (a so-called “open gate” condition) in order to determine the pixel-by-pixel gains. Although determining the pixel-by-pixel gains by collecting scan data through the Dmin of the film may be closer to the desired “white card” condition of maximum light from the subject material to be scanned, it leaves the gain determination process susceptible to scratches, dust, fogged or exposed areas and other imperfections in the selected Dmin area of the film which may corrupt the resulting gain values and consequently induce artifacts in the resulting electronic image. Open gate gain determination avoids the problems of these imperfections. However, the difference in overall illumination between open gate and Dmin must be taken into account. This can be done by reducing the illumination level or the exposure (integration) time for the open gate scan in order to yield the same average signal as a Dmin scan. The concept of determining pixel-by-pixel gains using an “open gate” condition is well known in the art, as demonstrated also by the above-mentioned '723 patent which describes a method for determining open gate gain correction factors.
Open gate calibration for a negative film scanner has been found to work well for large scale non-uniformities in the system response, such as the general fall-off of illumination that is typically seen toward the edges of the illumination source. However, it has been found that narrow, sharp non-uniformities in the illumination source are not corrected as well and show up as streaks in the subsequently scanned images. Things that can cause such narrow, sharp non-uniformities are dust particles at the film end of the illumination source or a small dark particle embedded in the film end of the light diffuser. If open gate derived gain correction factors are applied to the open gate scan data, the result will be uniform response across the scan. Similarly, if Dmin derived gain correction factors are applied to image scan data, the result is a fairly uniform photosite response over all image densities. This assumes that no localized imperfections were detected in the film Dmin area during the derivation of Dmin gain correction factors. If such localized imperfections are detected during the Dmin scan, they will have an effect on the Dmin gain correction factors which will show up as streaks on the resulting image reproduced from the image scan data even though no streaks exist in the film in the image frame area. However, if open gate gain factors are applied to image data rather than Dmin gain factors, it has been found that narrow, sharp non-uniformities resulting from anomalies, e.g. hair or dust particles, in the optical imaging path are not corrected properly in the resulting image data and show up as streaks in images produced from such image data. When the open gate gains are compared to the Dmin gains, the Dmin gains have relatively lower peaks in the areas of narrow, sharp illumination non-uniformities. There is a difference between open gate gain correction and Dmin gain correction which is not accurately described by a pure gain and offset model.
In summary, in a film (transmission) scanning system there has been found to be some differences in system response characteristics between a so-called “open gate” condition (which is ideal for determining pixel-by-pixel gains and offsets) and the “film present” condition when actual image scanning
Bryant Steven M.
Compton John T.
Boos, Jr. Francis H.
Eastman Kodak Company
Grant II Jerome
LandOfFree
Method of scanning photographic film images using... 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 of scanning photographic film images using..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of scanning photographic film images using... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2508897