Television – Camera – system and detail – Combined image signal generator and general image signal...
Reexamination Certificate
1996-12-19
2004-02-10
Garber, Wendy R. (Department: 2612)
Television
Camera, system and detail
Combined image signal generator and general image signal...
C348S237000
Reexamination Certificate
active
06690418
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to an image sensing apparatus suitable for using a sequential scanning type solid-state image sensing device having a color filter of so-called Bayer arrangement, and the like.
Recently, an image sensing device, such as CCD, capable of sequentially reading signals of all the pixels (referred as “non-interlace scanning type image sensing device”, hereinafter) has been developed with the progress of semiconductor manufacturing technique. The non-interlace scanning type image sensing device has an advantage in that a higher resolution image can be obtained with less blurring than an image sensed by using a conventional interlace scanning type image sensing device even when sensing a moving object. In the interlace scanning type image sensing device, a frame image is composed of two field images which are sensed at different times, usually at a field period interval. Accordingly, there is a problem in which, when sensing a moving object, there are notches on edges of the object and perhaps of the background in a frame image because of the time gap between the two field images composing a frame image. If a frame image is made of image data of a single field image to overcome the aforesaid problem, there would not be notches on edges, however, the vertical resolution of the obtained frame image is halved. In contrast, with a non-interlace scanning type image sensing device, it is possible to sense a frame image in a field period, thus, the non-interlace scanning type image sensing device is expected to be applied to a still image camera and a camera for EDTVII, for example.
A general image sensing apparatus using an image sensing device which outputs signals after adding two vertically adjacent pixel charges will be explained with reference to FIG.
22
.
Referring to
FIG. 22
, an image sensing device
901
output signals after adding two vertically adjacent pixel charges in accordance with timing signals t
1
and t
2
generated by a timing signal generator (TG)
909
. The output image signals are inputted to a correlated double sampling (CDS) circuit
903
via a buffer
902
, and reset noises of the image sensing device
901
are removed from the output image signals by the CDS circuit
903
., then the image signals enter an automatic gain controller (AGC)
904
. In the AGC
904
, the image signals are amplified by a gain set in accordance with a control signal c
2
from a microcomputer
908
(gain control). The gain-controlled image signals are converted into digital signals by an analog-digital (A/D) converter
905
, then transmitted to a camera processing circuit
906
, where predetermined processes are applied to the digital image signals and a luminance signal Y and a color difference signal C are outputted. Further, the microcomputer
908
generates a control signal c
2
for controlling the gain in the AGC
904
in accordance with the gain information c
1
detected by a camera processing circuit
906
.
Most of the non-interlace scanning type image sensing devices used at the present time are provided with R, G and B filter chips arranged in so-called Bayer arrangement as shown in FIG.
23
. In this color arrangement, G signal is used as a luminance signal. Further, the non-interlace scanning type image sensing devices output image data of one frame in a field period, thus the speed for transferring charges is two times faster than the transferring speed of an image sensing device, as shown in
FIG. 22
, which outputs image signals obtained by adding two vertically adjacent pixel charges. Accordingly, it is preferred to design an image sensing device to have two horizontal registers which respectively transfer image signals of odd and even lines simultaneously to be first and second output signals, instead of transferring by one line through a single horizontal register.
In this case, the buffer
902
, the CDS circuit
903
, the AGC
904
, and the A/D converter
905
shown in
FIG. 22
are needed for each of the two horizontal registers. In a case where image signals are obtained from an image sensing device provided with a color filter of Bayer arrangement as shown in
FIG. 23
, the G signals are obtained from all the corresponding pixels by horizontally interpolating by using signals obtained at G filter chips (i.e., G signals) as shown in FIG.
24
A. The G signals obtained as above are converted into luminance signals to be displayed. However, if image signals to be displayed are obtained in this manner, difference in characteristic between the two AGCs will affect the quality of an image.
The AGCs change gains to be applied to image signals when a gain is provided, and the two AGCs have different characteristics from each other in general. Therefore, even though the same gain is provided to the two AGCs, the levels of amplified image signals may differ from each other. If this occurs, when an object of a uniform color (e.g., a white paper) is sensed, a variation in output signal level of the two AGCs appears in a stripe pattern of odd and even lines. Therefore, when an image of the object is displayed, the variation in output signal level appears as the difference in output signal level between odd and even line fields on a display as shown in
FIG. 24B
, which causes field flicker. This noticeably deteriorates the quality of the image.
To overcome this problem, a method for interpolating an average of pixel values of the G signals in a vertical row can be considered. However, in this method, when an object of a uniform color (e.g., a white paper) is sensed, a variation in output signal level of the two AGCs may cause vertical stripes which alternatively have different output levels on a display as shown in FIG.
24
C. The difference in output level in the vertical stripes also noticeably deteriorates the quality of an image.
Further, in order to compensate a variation in output signal level of the two AGCs, feed-back control is considered. Feed-back control can be performed in the following sequence, for example. First, a test signal is inputted to the two horizontal registers of the image sensing device at predetermined timing, then the test signals outputted from the two horizontal registers are processed in the same manner of processing image signals. Then, the difference in output level between the two AGCs is detected. On the basis of the detected difference in output level, new gains to be provided to the AGCs are adjusted. Thereafter, a test signal is inputted to the horizontal registers again so as to confirm that the difference in output level between two AGCs are corrected.
As described above, if there is a variation in characteristic between two AGCs, it is possible to compensate a variation in output signal level between the two AGCs by performing feed-back control.
However, in the aforesaid method, a problem in which the size of a hardware increases is posed in providing a feed-back control function for compensating a variation in output signal level between two AGCs, since a circuit necessary for performing the feed-back control has to be added.
Further, because of an arrangement of pixels for luminance signals (i.e., pixels with “G” filter chips) as shown in
FIG. 23
, the following problem arises.
FIG. 25
is a graph showing a spatial frequency plane. In
FIG. 25
, the abscissa denotes a spatial frequency in the horizontal direction, the ordinate denotes a spatial frequency in the vertical direction, and the oblique axis denotes the spatial frequency in the oblique direction. In
FIG. 25
, the spatial frequencies at ∘ marks are the spatial sampling frequency decided in accordance with the number of pixels, and so on, of CCD (the spatial sampling frequencies in the horizontal, vertical and oblique directions are respectively referred as “fs-h”, “fs-v”, and “fs-o”, hereinafter). When signal components having these sampling frequencies are sampled, the obtained signals appear as a direct current component signal, which gives bad effects on an image. Therefore, it is necessary to remove signal
Hattori Yuichiro
Terasawa Ken
Canon Kabushiki Kaisha
Garber Wendy R.
Morgan & Finnegan
Nguyen Luong
LandOfFree
Image sensing apparatus image signal controller and method does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Image sensing apparatus image signal controller and method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image sensing apparatus image signal controller and method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3281275