Television – Format conversion – Changing number of lines for standard conversion
Reexamination Certificate
1999-04-16
2001-11-13
Eisenzopf, Reinhard J. (Department: 2614)
Television
Format conversion
Changing number of lines for standard conversion
C348S441000, C348S443000, C348S448000
Reexamination Certificate
active
06317159
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention relates to a scanning line number converting circuit for converting video signals of various formats of the different numbers of scanning lines into video signals of the predetermined number of scanning lines.
2. Description of the Related Art
As a standard television broadcasting signal, an NTSC (National Television System Committee) system and a PAL (Phase Alternation by Line) system have been known. Although the number of scanning lines of one frame is equal to 525 in the NTSC system, the number of scanning lines of one frame is equal to 625 in the PAL system. The numbers of scanning lines, therefore, in the NTSC system and the PAL system differ.
The development of television broadcastings of not only the standard system such as NTSC system or PAL system but also the HDTV (High Definition Television) system has been being developed in recent years. The number of scanning lines of one frame in the HDTV system is equal to 1125.
In computer images, further, a video signal of a format different from that of the television broadcasting is used, the number of pixels of VGA (Video Graphics Array) is equal to (640×480) dots, and the number of pixels of SVGA (Super VGA) is equal to (800×600)dots.
As mentioned above, in recent years, not only the video signal of the standard system such as NTSC system or PAL system but also video signals of various formats of the different numbers of scanning lines such as video signal of the HDTV system, video signal for computers, and the like are used. A display which can cope with the video signals of those various formats is demanded.
Hitherto, as a display, a CRT (Cathode Ray Tube) display has widely been used. In case of the CRT display, the number of scanning lines changes by changing a deflecting speed of an electron beam. It is, therefore, possible to relatively easily realize a display which can cope with the video signals of various formats. However, in the CRT display, an electric power consumption is large and it is difficult to miniaturize.
On the other hand, recently, in place of the CRT display, the development of an LCD (Liquid Crystal Display) display, a plasma display, and the like has been being progressed. In the LCD display and plasma display, since the size is small and an electric power consumption is small, it is presumed that they will be further spread in future.
In the LCD display and plasma display, however, the positions and the number of pixels are fixed. Therefore, to allow the LCD display or plasma display to cope with video signals of various formats, it is necessary to convert the number of scanning lines.
As a method of converting the number of scanning lines, there have been proposed a nearest neighborhood interpolating method of extracting data of a line existing at the position nearest to the position of a line after completion of the conversion of the number of scanning lines from inputted data of one scanning line, a bilinear interpolating method of extracting data of two lines existing at the positions nearest to the position of the line after completion of the conversion of the number of scanning lines from inputted data of one scanning line and linearly interpolating from the data of two lines, a filter switching interpolating method of converting the number of scanning lines by using an FIR filter in accordance with a conversion ratio, and the like.
Although the nearest neighborhood interpolating method can be realized by an extremely simple logic arithmetic operation on a hardware construction, there are problems such that a picture quality after the conversion fairly deteriorates, thin lines are extinguished and a small figure is distorted at the time of reduction, and a notched portion appears in a peripheral portion at the time of enlargement.
According to the bilinear interpolating method, although the deterioration of the picture quality is less than that of the nearest neighborhood interpolating method, when an image is reduced into (2:1) or less, a phenomenon called a line dropout occurs and the picture quality remarkably deteriorates. According to this method, since a gentle low pass filter is performed, particularly, a picture quality of a vertical edge portion (lateral fringe) becomes a picture quality of a slightly blur image.
On the other hand, in the filter switching interpolating method, the conversion of the number of scanning lines is performed by using an FIR filter in accordance with a conversion ratio. According to the filter switching interpolating method, although a construction becomes complicated, the conversion of the number of scanning lines can be performed at a high picture quality as compared with the nearest neighborhood interpolating method and the bilinear interpolating method.
The filter switching interpolating method will now be described hereinbelow. The conversion of the number of scanning lines of a non-interlace image will be first described. In the non-interlace image, a process of a frame period is performed and even after the conversion of the number of scanning lines, so long as the non-interlace, there is no need to separate the processes for the first field and the second field. Therefore, the processes are relatively more simple than those of an interlace image. To show an outline of an idea, explanation will be first made with respect to a non-interlace signal as an example.
For example, a principle of the conversion of the number of lines for the (2:3) enlargement (hereinafter, also referred to as a (2:3) enlargement line number conversion) such as to form three output lines for two input lines will now be described.
FIG. 1
shows a diagram for explaining the principle of the line number conversion for the (2:3) enlargement. In
FIG. 1
, values of each input line are set to Ri−1, Ri, Ri+1, Ri+2, Ri+3, . . . and values of each output line are set to Qi, Qi+1, Qi+2, Qi+3, . . . , respectively. In the diagram, P
1
, P
2
, P
3
, P
1
, . . . indicate deviations (line phase information) of the phases of the input lines and output lines.
In the (2:3) enlargement line number conversion, as shown in
FIG. 1
, three output lines are formed for two input lines and between the input line and the output line, there is a relation such that the values of the output line are calculated from the input line near them. Various interpolating methods exist in dependence on which range is used as a neighborhood range to form the output lines, which coefficient values are used as values of coefficients when the output lines are calculated by the interpolation from the input lines, or the like. However, an example of a cubic interpolation for interpolating from ranges of four points (corresponding to four lines) as neighborhood ranges will now be described hereinbelow.
A cubic interpolation function Cub(x) which is used in the cubic interpolation is shown in FIG.
2
and its functional equations are shown in equations (1). It is assumed that an axis of abscissa of the cubic interpolation function shown in the equations (1) is normalized by a sampling interval when an original image is sampled to a digital signal.
Cub
(
x
)=|
x|
3
−2
|x|
2
+1 (when |
x|≦
1)
Cub
(
x
)=−|
x|
3
+5
|x|
2
−8|
x|+
4 (when 1<|
x
|≦2)
Cub
(
x
)=0 (when 2<|
x
|) (1)
In case of the enlargement line number conversion, an interpolation value of each output line is expressed by a convolution arithmetic operation of the values of the four input lines and the cubic function and the interpolation values of the output lines can be expressed as shown by the following equations (2).
Qi=Cub
(
x
11
)*
Ri−
1+
Cub
(
x
12
)*
Ri+Cub
(
x
13
)*
Ri+
1
+Cub
(
x
14
)*
Ri+
2
Qi+
1=
Cub
(
x
21
)*
Ri−
1+
Cub
(
x
22
)*
Ri+Cub
(
x
23
)*
Ri+
1
+Cub
(
x
24
)*
Ri
Eisenzopf Reinhard J.
Sonnenschein Nath & Rosenthal
Sony Corporation
Yenke Brian P.
LandOfFree
Image processor for conversion of scanning lines and... 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 processor for conversion of scanning lines and..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image processor for conversion of scanning lines and... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2612346