JPS60185469A - Reading method of picture signal - Google Patents

Abstract

PURPOSE:To prevent an unnecessary picture signal from being picked up from the beginning by decreasing the bit number at each compression unit region of other picture signals while leaving one picture signal among digital picture signals converted from a color separation picture signal. CONSTITUTION:Plural adjacent picture elements on a color original are set as the compression unit region, a scanning light beam is projected to photo-electric converting means 8, 9, 10 via a color separation optical system provided with a single pickup lens 2 and a light beam split means 3 or the like and each color separation picture signal is obtained at the same time. The color separation picture signal is subject to digital conversion at the same time, and the other digital picture signals leaving at least one signal among the digital picture signals in the converted compression unit region are omitted while keeping a prescribed relation and the picture data with decreased bit number at each compression unit region is obtained.

Description

[Detailed Description of the Invention] C) Industrial Application ψ] This invention relates to a method for reading image signals in an rl image scanning and recording device such as a color scanner, and particularly relates to a method for reading image signals in an rl image scanning and recording device such as a color scanner. In order to be able to perform data editing at the same time as signal hunting, Z. regarding the method for extracting W from the i+ image signal.

(B) (/Y=Next technology For example, the color tone corrected image data corresponding to the 9th original number ii!+i obtained by the scanner input unit is temporarily registered in a memory device such as a magnetic disk, and then rear,
The central processing unit (CPU) performs editing processing according to a predetermined layout, and the scanner output section copies the data onto film, etc. based on the processed adjacent image data.
I image exist d self-record 7.7 like 1' [i! In 1+ image processing equipment, Color-Light II! By adopting II family's II-regeneration technology,
Treated as 4ke1. The dramatic decrease in digital visual data and the editorial department: DJ! 3. Increased performance and hardware speed. Jundan 4 due to faster speed on i14 side
Shortening of working time is shown in [Nu1 et al.71].

The compression and reproduction technology of such color gradation images is such that the gods are happy with this technology.
-227n8, we are devising a method for EF, pairing, and reproduction of 1iIji combined signals by utilizing human visual perception and the correlation between image signals. In other words, while this method improves the ability of human vision to distinguish between light and dark areas, which is sensitive to changes in brightness and darkness in minute areas, this method is sensitive to changes in color.
1) η of about L? ↑y)・Sho rice unit 61 area of image signal consisting of 11 pixels All 11 about the Ipa issue that meets all the brightness! 1. In order to reduce the number of pixels in the entire image data for each compressed unit area by omitting the signals responsible for colors in a predetermined relationship, 1. It is something.

According to this method, if it is necessary to put I[ilI image 1- into a memory device row -tT, the VC will store its contents [11] in dog[
Since it is possible to make a contract in one report - ('as well as the compression unit area and the desk: the basic unit of the fruit work is the arrangement of the two parts that should be done'), the editing Senbokudai・Kano-1
The speed can be increased in proportion to the number of flights at the f@ rate. Side],
-1 As shown above, in order to output the 51iA issue, the final image ( Young, 11f1
5. Next, the concave and image signals of the editing section are successively outputted from the memory device, and the duplicate image is recorded using the back image signal.5. In order to
I don't think output time will be an issue.

On the other hand, the input unit 1111 inputs 1 to the 1j51 image data picked up for the entire traverse without yoke.
If you perform FfE 4IX starting from 1 in 1, once for the yA stroke of the phase number i)',, ii;
+I image signal is included in the store-! Is the memory device used extremely dog'? ■
It's in season.

Therefore, while reading the original using a conventional rotary cylinder type scanner input device, it has been considered to write the image signal that has been reduced to 11th order into the memory device in real time. .

However, in order to do this, for example, if the compression unit area is 2 x 2 pixels, a panzer memory for at least 1 scanning line x 4 colors is required before writing to the memory device from the scanner entrance, and L = Hara 1tji Niji Linda 2
For each revolution, the compression ζ1. Since the image signal was transferred, it was not possible to turn on the input section +11j at all.

In this way, compression and reproduction technology of color gradation images σ)l-
Even though such an image scanning and recording device can significantly shorten the time required for editing, it is not possible to increase the reading speed of the input section, so it is difficult to increase the overall processing speed and seat size. However, there is a growing demand for faster thread removal speeds.

In order to solve the technical problem of increasing input speed, Honwade and others published JP-A-58-123540 and JP-A-58.
-137361, a method has been proposed to simultaneously read the gate and negative signs corresponding to several analytical cylinders in one rotation of the 1000 cylinder (1), and these methods are applied. I could think of something to do, but it cannot be applied to Usumi. This is because it is usually necessary to change the magnification between the original image and the duplicate image.
M, part 1. This is because it is necessary to take into consideration the tone and strength of the slopes at the foot of the mountain and on the mountainside side.

0 勺 Confession Riko's invention is based on the patent application filed in 1983 by the applicant.
-123540 or Vi JP-A-1983-'l5756
Using the technology disclosed in 1 [-, the image signal is picked up from the original image, and at the same time, the
08, the IF sum, ) method is implemented, and the unnecessary it↓z1 combination r is picked up from the most ,
The purpose is to provide information on how to read the images that can be obtained immediately due to changes in the reproduction magnification of U2.

ni) (J/) Naruko's invention Shiko-kaker Z1 ryo fIJ: (,; The reading method for the issue is to set multiple adjacent jiIJ: element of color hara-hi, for example, 2I2 ryo-murasaki as the compression unit area. Then, in this area, sub-scanning direction VC@contacting rU number one example (・-1
', saves 71 tons of color originals with a single scanning light beam that covers the two scanning lines (+tfl The image is projected onto each photoelectric conversion means through the color separation optical system obtained by the camera, and by t'1., each color separation signal 7 corresponding to the two signals is simultaneously obtained, and the color separation signal 7 is simultaneously obtained. Song 1 image number is digitally transformed j2, transformed σ1
1) Each digital signal in the 1st area 4) At least one of the 1 image signals has brightness. (1ti'i L, the number of bits is reduced for each compression unit area)9. A special exception is the acquisition of image data.

Furthermore, by using a zoom lens as the big-a-tube lens, when converting the reproduction magnification, it is possible to quickly change the sub-scanning width and width when reading a color original image to obtain 2.

(e) Embodiment 1 An embodiment of the method according to the present invention will be described below based on the drawings.

Figures 1 and 2 are schematic diagrams for explaining an example of the image reading method according to the present invention, and the original picture Schilling (
]) VrC attachment σ1 Attachment force 1 - The original image is scanned by a scanning light beam for the entire number of scanning lines (two in the example) on the original image.
A minute scanning pixel corresponding to 12 is scanned, and then the scanning light beam is focused fW on the digit of the light beam splitting means (3) via the biliqualib lens (2). This light beam split biomembrane (3) was previously published in Japanese Unexamined Patent Publication No. 123540/1988 [
A mirror (14) having an oval and ladder-shaped opening 0;
11.Constructed by combining roof type mirror (1υ).
The scanning light beam is divided by the light beam splitting means (3) into scanning light beams (111) and (1i 2) which are equal to 1 for each scanning pixel on each scanning line. Through this scanning light beam (111), condensing lens (41), dichroic mirrors (5), (61), (7), and each color separation filter (18R), (18G), (IIllB), photoelectric conversion is performed. The element (8), ('+9, Lli) photoelectrically converts the 7"L image signal corresponding to each color separation signal Ro, Go, BOνζ corresponding to the three additive primary colors. t12) is the condensing lens (
42), for example, only the green light beam is focused by
rI dichroic mirror (62), green component j
The signal is photoelectrically converted by the photoelectric conversion element (92) through the llfr filter (18G), and an image signal corresponding to the color separation signal 01 is obtained.

Dimensions?・7. Piece ・Perhaps not shown in Figure 1, the light beam is reflected by the mirror θ of the light beam splitting means (3) and is approximately split by the condenser lens ( 7I3),
By photoelectrically converting the photoelectric conversion element Q through the filter (18U), an unsharp signal (machigain et al. 21.
Ru.

As described above, color gon: By scanning an image, two scans %% that are tangent in the sub-scanning direction;'! The corresponding signals Ro, Go, Bo, image signals, jG+, and unsharp [No. In addition, as a matter of course, the RO and G of these Jl et al.
o, BO, Gl and Uf, 1 photoelectric transformer 1 seiko, even though the filters used are different, 1 for loquat and 1 for light suspension.
- Adjusted to have the same photoelectric change) color characteristics σ11
It is assumed that

In the method according to the present invention, two-dimensional compression is applied to the third file 1. A known color calculation circuit (2) for color scanners that outputs both color saturation and lightness signals.
7), an A/D converter (for example), a former, an l-conversion circuit (c), etc.

As shown in FIG.
image signals Ro, Go, Bo Vi, then color 7"
-/Necessary processing is performed using arithmetic circuits No. 12, C1
It is converted into a color plate signal corresponding to the amount of ink of four colors M, Y, 1]. Furthermore, the unsharp signal C) is added to M, Y, and each color signal r of M (as shown in the figure, and the portion corresponding to the detail of the color original 1i1j is emphasized). Ml produces n.

In addition, the image signal G1 of the scanning line F adjacent to the pixel of the scanning line where the image signal RO1GO1DO was originally removed is taken, and the pixel corresponds to the signal 01. However, if the difference is about 1iII]I elements, this type of image signal iij (7') phase r9J nature also indicates that C, M, The degree to which each color plate signal changes suddenly in Y, is small, and even if it does change, the human sense of perception is not capable of that much in minute parts [2], so the image The color plate signal corresponding to G1 is the image signal G
Assuming that there is no practical zodiac sign even if a color plate signal corresponding to O is substituted, the image signal G1 is read into the color plate signal1.
11. Detail is emphasized by adding the detail emphasis signal Δ1 of JLIJ. A color plate signal M1 is created.

Next, in order to perform sampling at a pixel pitch that corresponds to the replication magnification of the output image i+ij for the analog signals of and n et al. 1. 1. A basic clock signal generated by an encoder (not shown) has its timing adjusted by a known timing pulse generator. A/D converter (
g) ~-Input σn, ru. 'Cf1. V(, more Co
, Mo, Yo, KO, J: and Ml, for example, an 8-bit digital signal is generated for each pixel pitch according to the reproduction magnification of the output image. The format conversion circuit with the input section is a kind of data selection circuit, and corresponds to the output pixel pitch. Each time the clock signal is input, the output signal 21, the color plate signal Mo, and the color plate signal M+ are outputted in a predetermined order.

In other words, in the unfaithful example, as schematically shown in Figure 4,
The compression unit area is a 2 x 2 ii elemental area, and only the color plate signal M is output for all pixels as a value representing the brightness, and the color plate signals C and Y are the representative image tip position within the work type unit t1 area. It is output only for pixels and omitted for other pixels. Note that the subscripts (00), (ol), σC), and αη in the figure indicate the 6% relationship between the pixel of 1t and the representative surface element. Therefore, as shown in FIG. 5, each compression unit area of 4 pixels is grouped together into a format. In order to perform editing work such as layout processing, for example, using a subsequent image processor, it is desirable that the pixel density of the read and refined image signal corresponds to the pixel density of the output image.

Normally, the pixels of the output image are constant; for example, as in JP-A-54-35613 and JP-A-54-65601,
In order to perform magnification that is faithful to the original image, instead of duplicating the image signals of the same pixel or recording the image signal by thinning out the image signal for each predetermined pixel, it is necessary to At the input stage, it is necessary to set an appropriate sampling pitch and unit reading area (the range that is actually read as one pixel) commensurate with the duplication magnification.In conventional color scanners, this unit reading area is set by adjusting the aperture of the input section. , - This is handled by appropriately selecting the diameter.

In other words, in an actual color scanner, the magnification setting range is divided into several small ranges, and an aperture with an appropriate diameter is selected and used for each division. (For color scanners that read 1irll image signals at -, by using a zoom lens as a pickup lens, it is possible to adjust the unit reading for continuous changes in duplication magnification as well as for the zumbling pitch in the main scanning direction and sub-scanning direction.) The area can be adjusted.

However, in the method according to the present invention, in which one scan, for example, two adjacent scanning lines are simultaneously scanned [-], the scanning line can be read by selecting the aperture diameter. It is possible to adjust each unit reading area, but the pitch interval of each unit reading area, that is, seconds,
It is difficult to adjust the distance between ii+1 elements to H.

Therefore, if a zoom lens is used as a pickup lens, the distances between the unit reading area, aperture 1t, and pixel output 1 can be adjusted at the same time, so there is no need to change the aperture diameter, and the reproduction magnification of 1-history V is sufficient. can be accommodated.

In this way, the optical system after the light beam splitting means (3) does not need to be touched, but as the set magnification of the zoom lens is changed, the photoelectric conversion elements (81), (91)
, (101), (92) Since the gain adjustment of the valley amplifier installed after (92) or the adjustment of the illumination light amount is multi-axis, even if there is a change in magnification, the automatic aperture mechanism synchronizes with the change in magnification. I am using a zoom lens with a .

゛Also, the adjustment of the corresponding distance on the original image that the image signals Go and Gi read n, that is, the distance between one element per reading cycle, is not strictly 1 to the set magnification, but, for example, in stages. If you decide that variable is fine, prepare multiple lenses with different imaging magnifications for the song while keeping the distance between the original image and the light beam splitting means constant, and use a clever magnification setting ((For preparations, 71 (H', Fil lenses, choose the one closest to you) (You may choose to use a small one. In this case, change the aperture value for each lens to increase the light weight and the light intensity that hits the conversion element. Just keep it constant.

In the above, the rdl-
, the mechanical sub-scanning feed pitch of the sub-scanning section is determined by multiplying the magnification by 4 times, then calculating the compression unit area (honjitsu), and in the case of a hook example, it is necessary to have a feeding distance of 2 strokes. It's not even a moat.

In addition, in this embodiment, the image reading section is
Although the explanation has been made using the optical system disclosed in JP-A-58-137361, the present invention can be implemented at the same time by using the optical system disclosed in JP-A-58-137361 instead.

(f) Effects According to the image signal reading method of the present invention, unnecessary signals t are not read from the beginning, and two-dimensional compression of 7':Jn image signals in real time during one rotation of the original cylinder is performed. This simplifies the structure of the scanner input section, and even if it is necessary to once write the obtained image signal to a memory device, the capacity can be significantly reduced, and it is possible to Since the editing process can be performed, the editing speed can be increased in proportion to the reciprocal of the compression rate. Therefore, the overall processing speed of image data processing can be increased.

It is possible to easily change the reading pitch in the sub-scanning direction to match the magnification setting, and there is no need to change the magnification between the original image and the output image. There is no problem.

[Brief explanation of drawings]

Figures 1 and 2 eJ This brother 1! This is a schematic diagram illustrating an example of a method of reading an image by one thread in Ji, and ti′
Figure C3 (Sl) is a block diagram showing an example of an image signal processing circuit used in the method according to the present invention, and Figure 4 schematically shows an image signal input and processed by the method according to the present invention. 5bj is a schematic diagram showing the data format of the signal output from the format conversion path 1rjl in FIG. 3. (1) Original image cylinder (2) Pick amplifier lens ( 3)...Light beam splitting means (4t), (42)-...Lens (5), (6+), (62), (7)...Gicroic mirror (8), (9i), (92) , (
1, c4... Light military switching element fgi da... Color rendition circuit required... Format conversion circuit Figure 1 Figure 2

Claims (1)

[Claims] 1. The image of each iif++ element corresponding to the scanning line of the frame adjacent in the fi11 scanning direction on the color original image is scanned at least by those Jl. The image of each pixel scanned by the scanning light beam is sequentially scanned by a single scanning light beam covering the area. Through the optical system 1. When simultaneously obtaining required image signals corresponding to the number of scanning lines by projecting them onto a photoelectric conversion element, [2. Set a compression unit area, and each 1)! in the compression unit area. l
At least one of the image signals is left, and the others are left!
l Statue 1M' month of abuse! A method for reading both iW signals, characterized in that '6 is abbreviated to i while giving a phantom value. 2. Single pickup, 9 is Tomohiroji 5. Scope of visit: 1. The method for reading an image signal according to item 1. 6. The image according to claim 11, in which a single image pickup lens can be selected from pre-prepared lenses with an imaging magnification of i4 and a hook number of 8. How to read.