GB2182524A - Picture image input/output system - Google Patents

Abstract

The conventional picture image input/ output system is defective in that it has a large number of complicated processing steps, and therefore requires sophisticated skill and a large amount of time and labor. A lay-out retouch system has been proposed for the processing step in printing plants in recent years, but it needs a medium of a large capacity such as a magnetic disk for storing data on original color pictures, and requires a high speed computer for editing process, thereby increasing the cost of the system inconveniently. The present invention correlates sets of co-ordinates in the apparatus i.e. those of the digetiser board 20 with the input drum 10 and the output drum 30. Thus enabling processor 53 and computer 51 to carry out manipulation of the selected images on the input drum and reproduce them with desired layout and sizes on the output drum. Layout is effected with the aid of graphic display 52. <IMAGE>

Description

1

SPECIFICATION

Picture image inputloutput system GB 2 182 524 A 1 This invention relatesto a picture image input/ outputsystem, and more particularlyto a picture Image inputloutput system which enlarges or reducesata designated magnification and colorseparateseach colororiginal picture by means of an imagescanning device such as a colorscannerto obtain colorseparation signals, processesthe color separation signals for appropriate color correction, sharpness enhancement and gradation conversion, and sequentiaily outputs images on a recording material in a lay-outwhich is commanded by lay-out command data inputted by a digitizer orthe like.

There has been proposed a method for laying out each one of co[or separated films of plural original color pictures by preparing a screened color-sepa- rated film in a predetermined magnification out of respective original pictures by means of a color scanner, laying out and composing a masked separation printerwhich is prepared separately from the screened color- sepa rated films on lay-outsheets, and contact-exposing them to obtain a laid-outcolorseparated film for each color. This method, however, is defective in that it requires a large number of complex processing steps, much time and labor, materials and a high degree of skill for registering the color-separated films at predetermined positions on the lay-out sheetfor composing. There has also been proposed another method for reproducing a laid out color picture image by color-printing plural original color pictures respectively at predetermined magni- fication, cutting outthethus prepared reproductions of the original pictures in a predetermined roughsketch form, and laying out and composing them at a predetermined position on a base paper. This method, however, is problematic in its image quality because, since it involves the use of a photographic technique, it is impossible to arbitrarily change the processing conditions for color correction, sharpness enhancement, gradation conversion and so on. A device to output rectangular picture images in laid-outform by means of plural input devices, for instance thatwhich is disclosed in Japanese Patent Publication No. 3176211977, has been proposed,; however, the device cannot deal with arbitrary diagrams, requires manual operation for preparing masked plates,and needs plural inputscanning means forthe color manuscript input.

In recentyears a total-system or a lay-out reto.uching system has been proposed for processing in printing. Graphic patterns are inputted by a digitizer so asto display the graphic and image patterns on a cQlGrCRT (Braun tube) according to the system. Original color pictures are color-scanned at a predetermined magnification, AD-converted and stored in a memory such as a magnetic disk. The stored original color picture data are displayed on the color CRT in accordance with inputted graphic data, edited in a main memory unit of a computer by interactive input, and stored again on a magnetic disk orthe like in a format corresponding to an output picture image.

Thus edited color picture image data are then Da-converted, and inputted to an output control circuit of a color scannerto obtain a pictu re image in a desired lay-out. The above mentioned lay-out retouching system, however, is defective in that it requires a magnetic disk or other medium of a large capacityfor storing data for original color pictures, as well as a high-speed computerforthe editing process,thereby increasing the cost of constructing such systems, and extending thetime required for editing.

The address of the graphics used in such a color scanner is controlledjor instance as shown in Fig. 1, by giving the addressthe maximum valuexnax and the minimum value xnin on the x axis and the maximum valueym.x and the minimum value Ymin Of the y axis, reading an original picture 1 of a rectangular shape, and outputting the original picture 1 in a rectangular shape. The address supervising method mentioned above is inconvenient as the picture imagesto inputloutput are limited to rectanguiar orsquare shapes, because coordinates are determined bythe maximum andthe minimum values on the axes. An example of such a methodfor supervising coordinates is described in the Japanese Patent Application laid-open under No 6043811981.

In the case when an original picture is not rectangular, for instance in the case of the original picture 2 shown in Fig. 2, the address of a picture element 3 for read-out and output needs to have the address of the beginning and the end of the pattern for respective scanning lines. In the case of Fig. 2, accordingly, the memory content of the graphic position data becomes as shown in Table 1. Table 1 xl Y5 X2 Y4 X2 Y5 X3 Y3 X3 Y5...... X9 Y1 X9 5 and the address data required forthe graphic position processing becomes excessive in amount. Ifthe coordinates are controlled by the absolute address fromtheorigin (0,0), the computation load becomes too large as coordinates must be computed anew everytime the graphic pattern moves.

Accordingly, an object of this invention is to provide methods of operation of a picture image inputloutput system which obviates the aforementioned defects.

The present invention provides a method of correlating two sets of coordinates in a pictu re imageloutput apparatus which comprises:

a) a digitizer board for inputting graphic informa tion; b) a console which inputs necessary information and operation commands; c) a read means which optically reads original The drawings originally filed were informal and the print here reproduced is taken from a later filed formal copy.

2 pictures mounted on a predetermined position on an input drum; d) a color processing section which stores the picture image data of said original pictures; e)a picture image output means which outputs picture images using the picture image data from said color processing section on a recording material mounted on an output drum; and f) a computer system which iscoupled respectively to said digitizer board, and said console; wherein said method comprises correlating the coordinates of the digitizer board with the coordinates of the inputdrum,the method comprising positioning an input original picture on a flexible base for mounting saidflexible base on said input drum or digitizer board, whereby said base is positioned at predetermined positions on said input drum and on said digitizer board, and said computersystem affects a correlation between the coordinates ofthe two positions of the base.

Reference is hereby made to ou r co-pending application 8318366, from which this application is divided.

In orderthatthe present invention may be more readily understood, embodiments thereof will now be described byway of example, with reference to the accompanying drawings, in which:- Figs. 1 and 2 are diagrams to explain the conventional coordinate supervising method; Fig. 3 is a block diagram to show an embodiment of the apparatus for use in this invention; Fig. 4 is a viewto showthe aspect of scanning of an input or output drum; Figs. 5 and 6 are explanatory diagramsto describe the relation between coordinates of a digitizer and thatof an output drum; FIGs. 7A and 713 are explanatory diagrams for processing hidden surfaces; FIGs. 8Aand 813 areviewsto explain the relation between original input pictures and laid-out picture images; FIGs. 9Ato 9D, FIGs. 10 and 11 and FIGs. 12A and 12B are diagramsto explain the coordinate relation among respective devices, respectively; FIG. 13 is an explanatory diagram for an input picture image and the conditions of picture image memory; FIGs. 14A and 14B are diagramsto showthe inputfoutput aspectfor picture images and picture image data; FIGs. 15Ato 15C are diagrams to showthe manner of magnification conversion of inputfoutput picture images; FIG. 16is an explanatory diagram to describethe aspect of an input picture image and picture image memory; FIGs. 17A and 17B are diagrams for explaining the method to synchronize an input drum and an output drum; GB 2 182 524 A 2 density relation between input picture image and output picture imagefor preparing a gradation table. This invention will now be described referring to attached drawings. 70 FIG. 3 is a biockcliagram of an embodimentof a picture image input/output system accordingto this invention. In thefigure, original color picturesA, B, C and D are mounted on a transparent base 11 of an input drum 10 are outputted on a recording material such as a color paper31 on an output drum 30 as laid-out picture images A', B', C'and Win accordance with the datewhich has been inputted graphically by a digitizer 20 as a graphic pattern input device. The inputdrum 10 and the output drum 30 both have a cylindrical structure as shown in FIG. 4 and are rotated in one direction (main scanning direction) by a motor 12. The rotating position (main scanning position) of the drums are detected by a rotary encoder 13 connectedto an output shaft. The original color pictures Athrough D mounted on the input drum 10 are color-separated by read-head 16 which is moved tothe direction x (sub scanning direction) through a pulse motor 14 and a lead screw 15to read out image information, and a colorseparation signals PS (3-color separation signals and unsharp signal) are inputtedtoa logarithmic circuit 40, converted to densitysignals DS, and then converted to digital signals in an AD converter41. The density signals DS which have been converted to digital signals in the AD converter41 are inputtedto a color processing circuit 42 for color correction, sharpness enhance ment,gradation conversion, etc. and the color processed image data are memorized in a memory 43.The data memorized in the memory43 are converted into analog signals bya DA converter 44, and inputted to a modulator 45 in a laser beam printer to modulate the laser beam (the laser beam of blue, green and red orthree laser beams of different wavelengths in false color) outputted from a laser46 soastoexposethecolorpaper31 mountedonthe output drum 30 via an output-head 32. The output head 32 is moved to the direction of X (sub scanning direction) via a pulse motor 33 and a lead screw 34 connected thereto.

There is provided a console 50 with a keyboard as a data and command input device. The data and so on fromthe console 50 are inputted to a computer 51 (e.g. a mini-computer) and the information processed in the computer51 is displayed on a graphic display 52 of an interactive type. The computer51 isfurther coupled to a micro processor53 of an interiorsystem, this micro processor 53 being mutually connected to a color processing circuit42 and a memory43 by a bus line 54. The computer 51 andthe micro processor 53 form the computer system to display commands for operators on a graphic display 52 according to the stored program. The position x of the read-head detected by a linear encoder 17 which is engagedwith a guide rail 18 and the position data thereof am FIG. 18 is a flow chartto explain the operation of the 125 inputted to a timing control circuit 55. The position X computer system according to this invention; FIG. 19isa blockcliagrarntoshowthe method of forming parameters for setting the image process conditions; and FIGs. 20Athrough 20C are diagramsto show 1 T of an output-head 32 is detected bya linearencoder 35 engaged with a guide rail 36 and the position data thereof are inputted to the timing control circuit 55. The positions on y axis of the input drum 10 and those 130 on Y axis of the output drum 30 are therefore detected 3 t GB 2 182 524 A 3 by the rotary encoder 13 coupled with the rotating shaft thereof, and the position information thereof is inputted to the timing control circuit 55. The timing control circuit 55 drives, through the computer 51 and the micro processor 53, the pulse motor 33 at a constant speed atthe time of inputloutput of picture images, controls the driving speed of the pulse motor 14 and controls the AD converter 41, the color processing circuit42 and the memory 43 in respect of 1() timing.

The above description briefly explains the structure of the picture image inputloutput apparatus according to this invention and the coordinate relation among respective devices will now be described below.

Coordinate transformation on the digitizer 20 is first described.

The digitizer 20 has its own proper origin and X-Y axes, butthe origin can be moved to an arbitrary point by operation and the coordinates can be rotated 80 easily. In FIG. 5, assuming thatthe proper origin of the is OD, XD device 0 the abscissa, and the ordinate Y', that after inputting through operation in the digitizer 20 a new origin 01 and a point X, on a new abscissa X, the coordinate values of the points 01 and X, in the proper coordinate system of the device are (00, flo) and (A1, y'l), an arbitrary point (x'n, y'n) in the proper coordinate system of the device will become transformed to a point (xn, Yn) on a new coordinate system according to the formula shown below:

Xn COSO SinE) X'n - A0 D_ D lYn -sinO cosO Yn YO wherein 0 is an angle formed between the axis XD properto the device and a straight line 01 X,, and the counter-clockwise direction becomes positive. All of the computations according to the above formula can be conducted bythe computer 51.

The supervisory on the coordinate of the input/ output drurnswill now be described.

In respect of the input drum 10 and the output drum 30, the main scanning (rotation) direction is on y and Y axes while the sub scanning (transversal shift) direction is x and X axes. The coordinates of the read head 16 are measured in the timing control circuit 55 according to the following method. The method comprises the steps of multiplying the output of the rotary encoder 13 coupled with the rotational shaft of the input drum 10 in a PLL (Phase-Locked Loop) circuit, resetting the counter atthe origin of the y axes, and computing the output pulses from the PLL circuitto obtain an ordinate. the multiplication constantforthe period of the output pulse f rom the PLL circuit is determined so as to make it 50 [lim] or 10 [Liml on the input drum 10. The abscissa coordinates are controlled by resetting the counter atthe origin on the abscissa, counting the output pulses from the linear encoder 17 to knowthe position of the read-head 16 on the abscissa. Abscissa coordinates of the output drum 30 are controlled in a manner similarto thatforthe input drum 10. Ordinates Y of the output drum 30 can be controlled in a manner similarto that of ordinates V as the input drum 10 and the output drum 30 are synchronized in rotation.

The concrete method to position an input original picture and to correspond the coordinate system of the digitizer 20 and the coordinate system of the input drum 10 according to this invention will be explained below.

The digitizer 20 is corresponded with the input drum 10 in the coordinate systems by using, for instance, a rectang u lar transparent base 11 as a medium. The transparent base 11 is made of flexible transparent material in a form of a rectangular sheet which is providedwith two register pin holeswhich engagewith register pins along the upperside thereof. The digitizer 20 is provided with two register pins 62,62 to be inserted into the register pin holes of the transparent base 11. When the transparent base 11 attached with original pictures Ato D is to be mounted on the digitizer 20, it can be positioned correctly by registering the register pin holes of the transparent base 11 with the register pin 62,62. As a pair of register pins 61 A and 61 B are provided on the input drum 10 in a manner similarto that of the digitizer 20, when the transparent base 11 attached with the original pictures Ato D is to be wound around the input drum 10 for mounting, it can be positioned correctly by inserting the register pins 61 A and 61 B in the register pin holes provided on the transparent base 11. As the transparent base 11 is mounted at a rig Int position u pon the digitizer 20 and the input drum 10 by registering the pin holes with the register pins, the coordinates on the digitizer 20 of the color original pictu res A to D u pon the transparent base 11 can easily be corresponded with that on the input drum 10.

Although two register pin holes are bored along the upper side of the transparent base 11 in the above description, the position, shape and number of the pin holes may be determined arbitrarily so far asthe digitizer 20 and the input drum 10 are provided with pinsto correspond therewith. The base may not necessarily betransparentso far as it allows base manuscript input and picture image input.

Twotypes of inputs concerning graphic patterns such as coordinate are available. One isthe graphic patterns inputto designatethe shape of output pictures and the other isthe base manuscript inputto designate which graphic pattern of the output pictures should be correspond to the color original picturesAto D for read-out. The graphic pattern input is the operation to receive a graphic pattern from the digitizer 20 as an input andto correspond itto the pictureframe on the output drum 30 and is quite similarto the pattern inputwhich is carried out usually by a rough sketch plooter. The base manu- script input has a function mainlyto correspond to the coordinates of thetransparent base 11 attached with plural original color pictures (Ato D) and that of the input drum 10 and to relatetothe respective original color pictures on the transparent base 11 with the above mentioned input graphic patterns in respect of position and magnification.

Thefollowing will explain howto operatethe graphic pattern input.

Firstof all, the coordinates of the digitizer M and those of the output drum 30 will be correspond. In 4 short, as shown in FIG. 6, the reference symbol OD 0 denotes the origin properto the digitizer 20, XD0and YD 0 points on the abscissa and ordinate proper to the digitizer20, OD a point on the digitizer 20which 1 corresponds to the origin of the output drum 30 and XD yD land points on the digitizer 20 which correspond to the points on the abscissa X and the ordinate Y of OD XD the output drum 30. If the points land laresetto make the straight linesUO,-, XD and UD JXD 0 1 parallel to each other and proper coordinate of the point OD on thedigitizer20 is (XD, yD) (XD, D) 0 0 an arbitrary point n Yn on the digitizer 20 will be transformed into a point (Xn, Y,j on the coordinate system of the output drum 30 as expressed bythe formula shown below.

Xn _XDn - A0) (yD _ yD Y. n 0 ....... 1 (2) In this manner, the coordinates on the digitizer 20 can betransformed to the coordinates on the output drum 30. If the output sizeon the output drum 30 is commanded first by the console 50, the output size framewhich has been transformed at an appropriate ratio will be displayed on the graphic display 52. Then, if a graphic code (such as a rectangular or a circle) and necessary coordinates are inputted bythe digitizer 20 as a rough sketch pattern, the computer 51 will compute the coordinate transformation de- scribed above, the magnification and soon transformation necessary for the display on the graphic display 52, and consequently a graphic pattern will be displayed ata position and in size designated on the graphicciisplay 52. Everytime a new graphic code and a new coordinate pointare inputted,the computer 51 controlsthe graphic display52 so asto multiplexthe frame andthe graphic patterns which have been inputtedso long. As rough sketches are inputted in this manner, itis adapted thatan operatorcan visually confirm the display by using the graphic display52. When graphic patterns overlap each other,the hidden surface processing to be described hereinbelow must beconducted by inputting the commandforthe hidden surface in the digitizer20 and the console 50, and causing the computer 51 to carry outthe processing to complete a rough sketch information. In the case thatthe output picture on the graphic display 52 consists of overlapping patterns G1 through G3 as shown in FIG. 7A, the hidden surface processing will be conducted as shown in FIG. 713 by inputting, for instance G1 <G2, G2>G3" by the console 50.

The base manuscript inputwill now be described referring to FIGS. 8A and 8B.

The base manuscript input is carried out by the digitizer 20 as follows: i.e. thesteps of corresponding coordinates of thetransparent base 11 attached with original color pictures Ato D and those of the input drum 10, and detecting through a console 50 the corresponding relation between respective original colorpictures on thetransparent base 11 and the rough sketch pattern which has been inputted bythe graphic pattern inputoperation described above. The transparent base 11 attached with plural original colorpictures Ato D is fixed on the digitizer 20 by positioning it with register pins 62,62. Coordinates of GB 2 182 524 A 4 the transparent base 11 fixed on the digitizer 20 are transformed to coordinate of the input drum 10 in a manner similar to that explained concerning the transformation from the digitizer 20 to the output drum 30 in the graphic pattern input operation.Then, the rough sketch pattern which has been inputted is corresponded with coordinates of the original pictures Ato D on the transparent base 11 in respect of magnification. In otherwords, in FIGs. 8A and 8B, the output pattern Kcorresponds with the original picture A, but in orderto correspond tothe broken line A1 in the original picture A to the pattern A', a point in the pictu re A should be corresponded in coordinates with a point in the pattern A'and the magnification necessary for enlarging or reducing the broken line A1 to the graphic pattern A'should be determined. If these are satisfied, the coordinate relation will become absolutely determined. This is conducted simply by position-inputting a point in the graphic pattern A'and a point in the original pictureA for coordinates correspondence bythe digitizer 20 and inputting the magnification value bythe console 50. The original pictures Bto D and the output pictures B'to Ware also identical with the operation described above.

The coordinate supevisory between the coordinates of the digitizer 20 and those of the input drum 10 and the output drum 30 will be explained.

The position, shape and size of the picture to be laid out in output on the color paper 31 mountedonthe output drum 30 as a rough sketch are inputted bythe digitizer 20, defining the pattern on the coordinate system (the rough sketch coordinate system) which definesthe rough sketch, and thus defined graphic patterns are respectively supervised.

FIGs. 9A and 913 showthe case where the picture images defined bythe hatched areas 101, 201 of the input original pictures 100, 200 mounted on the transparentbase 11 on the input drum 10 are outputted in lay-outto the hatched areas 1 01A, 201A defined onthe color paper31 mounted on the output drum 30. However, forfacilitating understanding, explanation will be given on the case where an input original picture 100 is outputted in lay-out in the area 101A of the rough sketch orthe scope defined by hatched lines on the color paper 31 mountedonthe output drum 30. It is assumed that respective coordinate systems have the origin at left upper point,the abscissa axis X extends from leftto right and the ordinate Y axis extends from top to bottom in the description given below. The origin OH and a point 0 either on XH or yH axis ofthe rough sketch 1 01A shown in FIG. 9C are designated by the digitizer 20.

Coordinates ofsuch designated points on the digifizer20 are inputted to the computer 51 andthe computer51 computes the discrepancy between the coordinates ofthe digitizer 20 and the coordinate of the rough sketch. At this time, the coordinate origl.n OH 0 ofthe rough sketch pattern on the digitizer 2Gand the point XH on the abscissa are inputted in coordin- p atethereofas expressed in theformula below, point OH=XD yD 0 0, 0 pointXH=XD yD p 1,1 the coordinates ofthe rough sketch are displaced in XD XD yD yD parallelby Ointhe directionandby Ointhe 1 1 GB 2 182 524 A 5 direction on the coordinate system of the digitizer20, and itis rotated bythe angle 0 aroundthe point.

xD - xD - 1 0 cos E) V-(2 + _XD yD,)2 YD - YD 1 0 sin E) V(7_ XD )2 + (y? _ yD 0)2 ............... (3) The above mentioned instruction is recognized bythe computer51. If an arbitrary point of the rough sketch which is read bythe digitizer 20 is assumed to be (XD, yD) in the coordinates on the digitizer 20, the (XH yH) coordinates 1, 1 on the rough sketch are expressed in the formula below; (XH yH 1) 1 1 = (XD YD 1 1 1 0 0 0 1 0 -xD -YD 0 0 1 cosO +sinO 0 -sin() cosO 0 (4) 0 0 1 1 The computer 51 transforms the coordinates of 65 respective graphic patterns inputted from the digitiz er 20 into the coordinates on the rough sketch pattern by using the above mentioned formula (4). When the computer 51 receives as input respective coordinates of the pattern 101 A defined by hatched lines of the rough sketch pattern from the digitizer 20, the computer51 recognizes it as the pattern transformed to the coordinates on the rough sketch coordinate by operation as expressed in the formula (4). The pattern 101 A which has been recognized as the pattern on the rough sketch coordinate is recognized bythe compu ter51 as a pattern 101 B on the shape coordinate by defining a rectangle 102 which circumscribes the pattern 1 01A and is parallel to XH_yH axis and defining a coordinate system which hasthe origin (point 010' ') at thetop point 102Awhich isclosestto the origin of the rectangular rough sketch coordinate and axes (XK_yK axis) respectively parallel to XH_yH axis of the rough sketch coordinate. The above process can be express ed by the formula (5). If it is assumed thatthe coordinates on the rough sketch coordinate of the OK origin 0 (point 102A) of the shape coordinate system is (XH, Y), and the coordinates of an arbitrary point of 2 2 the rough sketch pattern 1 01A on the rough sketch is (XH yH), (XK yK) coordinate system 1, 1 thecoordinate 3, 3 of this point on the shape coordinate system will be represented bythe formula below. (XKyK 1) 3 3 =(X H Y H1) 1 1 1 0 0 -xH _YH 1 0 2 1 2 0 1 ................. (5) Contraryto the above, the rough sketch pattern 1 01A can be recognized as a graphic pattern which is obtained by parallel-displacing the pattern 101 B which passes through the origin OK of the shape 0 coordinate system and which inscribes the rectangular parallel to XH_yH axis to a designated position. The transformation from the shape coordinate to the rough sketch coordinate is carried out according to the following formula: (XH yH 1) 1 1 K K1) =(X3Y3 1 0 0 1 xH YH 2 2 1 1................

0.. (6) 1 The graphic pattern 101 A which has been inputted from the digitizer 20 is processed for transformation as described above to control the rough sketch 1 01A with the graphic data of the pattern 101Bontheshape coordinate system and the parameters XH, yH to be 2 2 used for transformation from the shape coordinate system to the rough sketch coordinate system. In this case, if the rough sketch coordinates and the positional coordinates on the output drum 30 are corresponded ata ratio 1: 1,the rough sketch data stored in the computer 51 will be reproduced on the coordinate of the color paper31 mounted on the output drum 30. As a result, the lay-out conditions designated in the rough sketch pattern of the digitizer 20 can be outputted in a picture image on the color paper31.

The position, shape and size of the original pictures mounted on the transparent base 11 are inputted from the digitizer 20 to define the original pictures on the base coordinate systems (the coordinate system to definethe original picture) according to the method described hereinbelow.

Referring nowto FIG. 8D, a transparent base 11 attached with an original picture 100 is mounted on the digitizer 20. In a mannersimilarto inputting a rough sketch, the coordinate on the digitizer 20 for an OB origin 1 of the coordinate system (or base coordin XB ate system) of the transparent base 11 and a point 9 onthe XB axisorthe yB axis of the base coordinate are inputted into the computer 51, the transformation parameters 0,, _XD,'_yD, fortransforming the coor0 0 dinates on the digitizer 20 to thecoordinates (the base coordinates) of the transparent base 11 are obtained in a manner similarto that used for obtaining the parameters 0,-A0,-Y'0 fortransforming the coordinates of the digitizer 20 to the coordinates of the rough sketch. The position and size of the original picture 100 on thetransparent base 11 attached on the digitizer20 are recognized asthe position, shape and size of the original picture 100 on thetransparent base 11 by reading the coordinates of the break points 103 to 106 on the outer periphery of the original picture on the digitizer 20, and transforming thus read coordinates on the digitizer 20 into the coordinate on the base coordinates system through computation mentioned Ey'_XD as (3) and (4) and using the parameters Wand _yD, 0 in the computer 51. In a manner similarto that used in the case of rough sketch, a rectangle 110 which circumscribes the original picture 100 and which has sides parallel to either axis XB ory 13 of the base coordinate is defined bythe computer51, and a manuscript coordinate system orthe coordinate system which has axis (x'- y'axis) respectively paralleltox'-y B axis of the base coordinate as an origin (point OcO at a vertex 11 OA orthe point closest to the origin O, of the rectangular base coordinate so as to recognizethe size and the shape of the original picture 100 on the base coordinate asthe diagram 6 GB 2 182 524 A 6 100A on the manuscript coordinate system. In this processing as in thatforthe rough sketch, if it is assuredthatthe origin OPof the manuscript coordin ate system of the original picture 100 on the transparent base 11 is expressed bythe coordinate value (X2', YD on the base coordinate system, the original picture 100 on the transparent base 11 can be recognized asthe result of parallel displacement of the original picture defined on the manuscript coor dinate system byX2' in the direction of the axisxl and byY213 in the direction of the axis y13 on the base coordinate. Then thetransparent base 11 attached with an input manuscript 1 00thus recognized is to be mounted on the input drum 10, butas pin holes are bored on thetransparent base 11 for mounting the inputdrum 10 on the register pins 61A and 61 B,the coordinates on thetransparent base 11 can be recognized as the coordinates on the input drum 10.

Asa result of aforementioned processing, the output position, shape and size of the original picture on the output drum 30 and the position, the shape and 80 size of the input manuscript 100 mounted on the transparent base 11 ontheinputdrum 10are recognized in the computer 51.

There arises the need for defining an image output scope and output magnification of the input manu script for outputting the original picture100tothe output drum 30 in the shape designated by the rough sketch 101A. The explanation thereforwill be given hereinbelow.

In the case where the output magnification S is designated in advance by the console 50 atthe time of inputting the rough sketch 1 01A, the rough sketch diagram 101 B defined on the coordinate shown in FIG. 8C is transformed in magnification and projected on the shape 100A ofthe original picture defined on the manuscript coordinate shown in HG.9D.Conse quently, the coordinate (XG4, YG4) on the manuscript oi (XK yK coordinate of an arbitrary p nt 3, 3)ontherough sketch 101C defined on the shape coordinates shown inFIG. 10 can be expressed bythe formula below:

(XG yG 1) 4 4 =(X KyK1) 3 3 VS 0 00 0 0...... (7) VS 0 The state of the input manuscript 1 OOA and that of the rough sketch pattern 101Cdefinedonthe manuscript coordinate are transformed from the manuscript coordinates to the screen coordinates and displayed on the screen of the graphic display 52. As thus displayed rough sketch which is defined on the manuscript coordinate is not established in correspondence with the image output scope, it is necessary to read the coordinates on the input manuscript mounted on the transparent base 11, and to transform it into the coordinates on the manuscript coordinate using the parameters E),,_Xffi,_yD'and 0 0 according to the transformation formulae shown in (3) and (4). By transforming the coordinates into that of the screen coordinate system and displaying a cursor on the screen corresponding to the position designated by the digitizer 20, as shown in FIG. 11, the position designated is displayed on the digitizer 20 and the reference point SP1A is designated forthe rough sketch pattern 101 D on the screen. Similarly, on the input manuscript mounted on the digitizer 20, the reference point ST1 is designated forthe input manuscript 1 00Awhich corresponds to the reference point SP1 of the rough sketch 101 C as shown in FIG.

(CD yD) of 10. Bythese designations, the coordinates 5, 5 the reference point SP1 of the rough sketch pattern 101 C on the manuscript coordinate is parallel displaced to the coordinate (XD yD) 6, 6 of there reference point ST1 of the input manuscript so as to define the image output scope 101 E on the input manuscript 1 OOA. Simultaneously, the reference point SP1A of the rough sketch pattern 1 01A is parallel displayed to the reference point ST1A of the input manuscript 100B on the screen of the graphic display 52 corresponding to the reference points SP1 and ST1 so as to definethe image output scope 101 E in the input manuscript. Through the above processing the coordinates (X13c, Y'3) of an arbitrary point on the rough sketch pattern 101 B which is, for instance, defined by the shape coordinate is transformed to the coordinates on the manuscript coordinates (XG, yG) i n 7 7 accordance with the following form u [a:

(xGYG 7 71) =(XKyKj) 3 3 lls 0 0 VS 0 0 0 1 0 X, A Y1 1 01 0 0 1 0 (8) whereinAx, =XD_XD Ay, =yD_yD 6 5r 6 5.

Therefore,ifthe rough sketch pattern 101Bdefined in the shape coordinate is transformed to the base coordinate by the coordinate transformation processing newly,the rough sketch diagram 101E defined on the base coordinate will indicate the image output scopeofthe input manuscript 100. The coordinates of point BS1 on the base coordinate which corresponds tothe point102Aonthe rough sketch coordinatecan be expressed as XB =Xll + A X,, yB =yB 2 2 + A yl. The coordination on the rough sketch coordinate of the point 102A can be expressed as XH =XH yH -yH 2, - 2.

On the other hand, if the output magnification S is not designated atthe inputtime of the rough sketch pattern 1 01A, a magnificaton which reduces the original picture defined on the manuscript coordinate should be considered (for instancethe magnification of 70%). In otherwords, if it is assumedthatthe maximum values of the coordinates of the manu script 1 OOA defined on the manuscript coordinate in XG yG XG the directions of axes and and 8andYG8,the maximum values ofthe coordinates on the diagram ofthe rough sketch pattern 101 B defined on the shape coordinate in the direction of axes XK and yK are XK and yK a magnification S, computed bytheformula 9' below isselected andthe rough sketch 101 B onthe shape coordinate is defined onthe manuscript coordinate.

1 7 1 GB 2 182 524 A 7 1ifXKjXG>yKlyG magnification S, becomes S,:X K lXG8 9 8= 9 8 - 9 K G< K G, Sl? K G fX9/X8=Y9/Y8 magnification S, becomes _Y9/Y8 By this processing, the coordinate (XG, yl) 4 4 on the manuscript coordinate which corresponds to the (XK yK) coordinate 3, 3 of an arbitrary point on the rough sketch pattern 101 B can be expressed on the shape 5 coordinate as in the formula (10).

(XG yG 1) 4 4 =(X KYK 3 31) llS, 0 0 1 0 llS, 0 30 0 0 1 In this waythe state of defining the shape 1 OOA and the rough sketch 101 C of the original picture are displayed on the graphic display 52 after transform ing them from the manuscript coordinate to the screen coordinate of the graphic display 52. However, the correspondence of the magnification and the output scope between the displayed rough sketch 101 D and the original picture 1 OOB is not desig nated, 40 it is necessary to read the coordinates of the input manuscript on the transparent base 11 attached on the digitizer 20 and transform them to the coordinate value of the manuscript coordinate using the para- if 1 XG _ XG 1/1 XG _XG1 1 yG - YG ill YG - YG 1, 12 13 10 1 12 13 10 11 1/1 XGO _ XG1 1 magnification S2 becomes S2 XG12 - XG13 1 1 if 1 XG _ XG J/1 XG _XG1 1 > 1 yG _ yG 111 yG _ yG1 1, 12 13 10 1 12 13 10 1 1 yG _ yG 1/1 yG yG magnification S2 becomes S2 = 12 13 10 -11 Inthis matterthe rough sketch 101Conthe manuscript coordinate istransformed in magnification and parallel displaced thereon. As a result,the coordinates (Xl,yG) onthe manuscript coordinate 3 3 cooresponding to the coordinates (XK yK of an 3, 3) abritrary point on the rough sketch 101 B on the shape coordinate can be represented bythe formula (13).

(XG yG 1) 3 3 =(X M1) 3 3 1/S2 0 1 0 01 (13) 0 1 0 1 0 0 1/S2 0 0 0 1 A X2 A Y2 1 Consequently the rough sketch 101 Con the manuscript coordinate is transformed to a diagram 101 E which is made to correspond to the shape 1 OOA of the manuscript. FIG. 12A shows the state. The resu It is transformed i nto the screen coordinate, displayed on the g ra phic display 52 to conf irm whether the displayed rough sketch 10 1 E and the orig inal pictu re 1 OOB are located at the desi red positions. FI G. 12B indicates such a state. If they a re not correctly corresponding to each other, correspondence between the reference points SN1 and SN2 and the reference points PN 1 and PN2 are designated again.

FIGS. 12A and 12B showthe relation between the ................. (9) 0,'_X[) _yD meters Jand 0 in accordance with the formulae (3) and (4). Then, the cursor is displayed on a position on the screen designated bythe digitizer 20 after transformation tothe screen coordinate, and is moved to the reference points SN1 and SN2 of the rough sketch pattern 101 C as displayed in FIG. 12B so as to designate the point newly asthe reference point. In a similar manner, the reference points PN 1 and PN2 which correspondtothe reference points SN1 and SN2 of the rough sketch are designated for input in the manuscript on the transparent base 11 mounted on the digitizer 20. By this designation the computer 51 makes the referenced points SN1 andSN2ofthe rough sketch 101 Con the manuscript coordinate correspond with the reference points PN 'I and PN2 of the input manuscript 1 OOA. If it is assured that the coordinates of the reference points PN1 and PN2 of the input manuscript and the reference points SN1 and SN2 of the rough sketch 101 C are SN1 = (XG,O, YG (XG,), yG1), = (XG yG (XG lo), SN2 = 1 1 PN1 12, 12), P N2 = 13, yG), the average displacement A X2. A Y2 and the 13 magnification S2 will be expressed as below.

XG XH AX2= 12- 10 XG y1A......

AY2= 12- 10 (12) manuscript coordinate and the screen coordinate of the original picture 100. A similar relation holds for the original picture 200.

(XK yK) In short, each coordinate 3, 3 of the rough sketch pattern 101 B on the shape coordinate is transformed incoordinate according to the formula (14) (XB yB 1) 7 7 (XK yK 1) 3 3 1/S2 0 0 0 1/S2 0 0 0 1 1 0 X 0 XB + 2 A X2 V2' + A Y 2 1 0 (14) 1 1......

into the base coordinate, and the rough sketch pattern 101 E defined on the base coordinate indicates the image output scope of the input manuscript. The output magnification S is shown as S= S2, and the origin OK of the rough sketch 101 B on the shape 0 coordinate or the coordinate of the point BS 1 on the base coordinate corresponding to the point 102A on the rouqh sketch coordinates becomes as 8 GB 2 182 524 A 8 Xs= X92 +A X2,Y" =Y2B +A Y2. The coordinates of the =XH'yh origin on the rough sketch coordinate are X" 2 =YH2. Accordingly,the position, shapeforthe picture image output of the output drum 30 designated by the output of the rough sketch pattern 1 01A and the image outputscope and the output magnification of the original picture 100 when it is attached on the transparent base 11 and mounted on the input drum 10 are determined. Then, coordinates are correspon- dingly selected forthe origin 102A on the rough sketch coordinate corresponding to the shape coorOK dinateorigin 0 which transforms the rough sketch 101 B on the shape coordinate to the rough sketch 1 01A on the rough sketch coordinate and forthe origin BS1 on the base coordinate which makes it correspondto the image outputscope of the original picture on the base coordinate.

Coordinate transformation and coordinate control between devices are carried out in the manner shown above. The operation of the picture image input/ outputsystern according to this invention will now be explained.

Graphic codes and positional information necessary forthe iay-out of the output picture images are inputted into a computer 51 by using the digitizer 20 and the console 50, the computer 51 produces a g raphic pattern according to thus input graphic data, and the produced graphic data is transmitted to the graphic display 52 for display. The operator discri- minates the g raphic pattern while watching the picture frame displayed on the g raphic display 5.2 and if there is any correction or addition to be made, the operator corrects it by using the digitizer 20 and the console 50. The input of the rough sketch is carried out in a manner simiia rto the one used for rough sketch plotters, and alight pen may be used.

Original color picturesAto D for lay-out output are nextmounted on a transparent base 11, and the transparent base 11 is placed ata predetermined position on the digitizer 20. The original color pictures Ato D corresponding to the input graphic pattern are sequentially selected according to the commands given to the digitizer 20 and by key operation on the console 50, and the output scopeforthe original color pictures Ato D and the hidden surface processing for the diagram are designated. The outputscope can be instructed by designating two points on diagnosis if the pattern is a rectangle and bydesignating a center point if it is a circle. Then the magnification which is necessary for correspondence of the original color picturesAto D with the graphic patterns Kto Won the output picture designated and inputted, and simultaneouslythe parametersfor necessary process conditions such as color correction, sharpness en- hancement and gradation conversion are inputted by the console 50.

Then the computer 51 computes bythe main scanning line Xi, the start position Y-5,and end position Y91of the color picture CP on the output drum as shown in FIG. 13 for each unit of the input graphic pattern and stores the result in a memory such as a magnetic disk. Namely, the X-axis position X, of which a scanning fine of the main scanning direction firsttraverses the color original picture CP is stored, andthen the start point Ysi and the end position YE1 on the Y axis where the scanning line X, passes on the color original picture CP are stored, respectively. Similarly forthe scanning line X2, the start point YS2 and the end point Y5of the position X2 on the X axis and the original color picture CP are stored. The graphic data of the original color picture CP stored in the memory becomes as shown in Table 2forthe case of Fi G. 13.

Table 2

X, ys YE YE YE 1 1 X2 LYS2 xi... ixn YSn 1 n 11 1 YP YP 1 Such graphic data of the original color pictures are sequential ly stored at predetermined bits (for instance, 16 bits) foreach of the pictures attached on the transparent base 11. In the storage shown in Table 2, as it is difficult to judge whetherthe data on an arbitrary location relates to an abscissa or an ordinate, the most significant bit (MSB) of the address data of the graphic pattern is used to discriminate Xfrom Y or vice versa. In otherwords, if the address data is, for instance,---110... 01 ", it is judged to be an abscissa while it is "001... 11 -, it is to be an ordinate. Explanation will now be given forthe absolute coordinate and the relative coordinate. The absolute coordinate is to control respective coordinates for inputloutput drums and isto be counted using respective origins thereof as reference points.

The aforementioned points such as BS1, BS2,102A and 202A are supervised bythe absolute coordinate. It is also used to control the coordinates of the output picture elements as shown in FIG. 14B. The relative coordinate, on the other hand, is used for controlling input picture elements. In the device described herein and shown in FIGS. 14A and 1413, the output picture element size is constant and magnification transformation is conducted by varying the sampling pitch of the input picture elements while the sizes on abscissa and ordinate axes are not changed. In short, the size of the input picture elements becomes variable. The sampling timing control is conducted by the relative coordinate for the input picture elements with varied sampling pitches. The unit of the relative coordinate thus changes depending on the magni-fication. The magnification transformation for abscissa is applicable the method generally used is a color scanner. The magnification for the ordinate direction of the input drum 10 and the output drum 30 is transforemed by driving a pulse motor33 on the side of the output drum at a given speed and changing the rotational speed of the pulse motor 14 on the side of the input drum 1.0. The positions of the read-head 16 and-the output-head 32 in the sub scanning direction can be learnt by computing the output of the linear encoders 17 and 35 by the address register in the timing control circuit55.

The method forvarying the magnification of the inputloutput picture imageswill now be specifically described referring to FIGS. 15Ato 15C.

It is performed by eitherchanging the pitch of the picture elements 5for output in respect of the picture elements4which scan the input manuscript as j 9 GB 2 182 524 A 9 h 10 1 shown in FIGS. 15A and 15B or byvarying the size of the output picture element 6 as shown in FIGS. 15A and 15C. In the lattercase, byvarying the size of the output picture elements 6, the pitch isto be changed correspondingly thereto. The size and pitch of the inputloutput picture elements are determined based upon the magnification input in the computer 51 by the console 50 and transmitted to the micro processor 53. Based on that, the micro processor 53 generates a command to the timing control circuit 55, and the timing control circuit 55 determines the sampling pitch of the input drum 10, the output pitch of the ouput drum 30 and the feeding speed of the read-head 16 and the output-head 32 in the sub scanning direction. Then, the memory of the compu ter 51 transmits the graphic data shown in Table 2 to the timing control circuit 55 via the micro processor 53. According to the present invention, addresses such as---1 " or "0" are assigned to both inputloutput data in the scanning direction, but as the magnifica tion is variable in pitch and size and the number of picture elements is constant, as shown in FIGS. 15A through 15C, by storing the reference position Ps of the inputloutput graphic pattern in absolute address and storing the picture elements 4to 6 on the periphery thereof in the address relative to the reference positions Ps adress of the picture elements on the periphery may be made identical for both input and output. In this case, the computer 51 supplies first 65 the magnification to the timing control circuit 55 via the micro processor 53, determines the pitch and size of the picture elements in the timing control circuit 55 and synchronizes with the head byfeeding the reference positions Ps of inputloutput as described hereinafter. The relative address of the graphic pattern is transmitted and the timing control circuit 55 feeds an identical signal both to inputloutput driving systems.

The most significant bit of the address data forthe graphic pattern of the color original picture can be used to discriminate X-Y coordinate in a manner described below. It is assumed thatthe outer periphery of the inputioutput drum are 600 mm respectively and the magnification scope is from 50 to 400%. Underthis condition, the relative coordinate scope is not dependent on the size of the input picture elements and remains within the size of the output drum. If the output picture element size is assumed to be 50[cm, it is determined by the formula below.

600(mm) X 20flinelmm) = 12,000 The absolute coordinate scope, on the other hand, becomes as expressed by the formula below, as in the case of this invention, the absolute coordinates are controlled by 10jum.

600(mm) X loo(line/mm) = 60,000 Value 60,000 can be expressed by 16 bits and 12,000 by 14 bits. If 16 bits are assigned to the relative coordinate,the high-order 2 bits can be used freely and one bit of thetwo is used for discriminating X from Y. All of the data shown in Table 2 are the addresses using relative coordinate.

Explanation will be given forthe case when a graphic pattern more complexthan the one shown in FIG. 13 is inputtedloutputted; for instance the case where a graphic data of the pattern shown by hatched lines in FIG. 16 is to be stored. In this case, the ordinate coordinates Y1 to Y44 are stored in respect of the scanning lines X, to X12 as listed in Table 3.

Table 3

X1 Y1 Y2 X2 Y3 Y4 X3 YE; Y6 Y7 Y8 X4 Y9 Y10 Y11 Y12 X5 Y13 Y14 Y15 Y16 X6 Y17 Y18 Y19 Y20 X7 Y21 Y22 Y29 Y24 Y25 Y26 X8 Y29 Y30 Y31 Y32 X9 Y33 Y34 Y35 Y36 Y37 Y38 X12 X43 Y44 Ordinates in an even number always existfor one X scanning line and the odd numbered one is judged as the start point while the even numbered one is judged as the end point. By designating a graphic information, it becomes possible to inputioutput a graphic pattern of an arbitrary shape. In this case, if the pitch on the X scanning lines is set dense enough, more accurate graphic pattern may be selected.

The capacity of the memoryfor storing coordinates can be reduced by controlling absolute coordinate and the relative coordinate corresponding thereto. The graphic patterns can be displaced simply by re-computing the reference position on the absolute coordinate without varying the address (or the relative coordinate) shown in Table 2, thereby reducing computation load conveniently.

Thetransparent base 11 attached with original color pictures Ato D is mounted on the input drum 10 by positioning it with the register pins 61A and 61 B. When the motor 12 is driven,the input drum 10 (the output drum 30) is rotated in one direction. The rotary encoder 13 is connected to the rotating shaft of the input drum 10, and the output pulse therefrom is inputted to the two address registers via the PLL circuit in the timing control circuit 55 which is controlled bythe micro processor 53. One of the address registers is to control the absolute coordinate of the rotational direction (main scanning direction) and the other one is to control the relative coordinate of the input picture elements.

If it is assumed thatthe read-head 16 of the input drum 10 is separated from the start position SP by xl, the output-head 32 of the output drum 30 is separated from the start position SP by X, and the magnification is M, and while the read-head 16 moves by x, the output-head 32 moves by M.x. In otherwords, the GB 2 182 524 A 10 ratio of the distances covered by the read-head 16 and the output-head 32 in the sub scanning direction for a unittime is the magnification M. The controlling method varies depending on the dimensional rela tion between xl, and X,/K When the relation is 70 expressed by X, X, >-- --- ------ --- (15) M as shown in FIG. 17A, the read-head 16 is controlled singly to move by (xl -X,/M) and then to move together with the output-head 32 simultaneously. In this arrangement, by the time when the read-head 16 co m es to th e sta rt p os itio n, th e o utput-h ea d 32 co m es to the start position SP coincidentally so as to synchronize the sub scanning direction.

When the following relation holds X, xl<............................................. (16) m asshown in FIG. 17B,the output-head 32 is controlled singlyto move by (X, - M-xl) andthen to move togetherwiththe read-head 16 coincidentally.

The inputloutput picture image data inthe main scanning direction issupervised asfollows.As shown in FIGS. 14Aand 1413, the pointP(xl,V1) is designated asthe point closest to the origin of a rectangle circumscribing thegraphic pattern onthe inputdrum 10 and is expressed byrelative coordinate specified by picture elements of a predetermined unit. A point G (X,, Y1) corresponding to the point P is expressed bythe absolute coordinate specified by the picture elements of a predetermined unit on the output drum 30. In this way, the picture element points of inputloutput picture images can be expressed bythe lattice points shown in FIGS. 14A and 1413. The picture element data of the densitywhich has been converted to a digital value by theAD converter 41 is processed bythe colorprocessing circuit42, and then the memory43 stores successivelythe start points and the end points thereof at a timing _ increased from the time when the address registerfor they direction of the input drum 10 becomes "yl". When the memory 43 is used in an output mode, it is made to become effective f rom the timethe address register becomes "Y, " in the direction Y, and is controlled to outputthe picture elements for the du ration from the start point Ysi to end point Y9,using the point Y1 as the origin. The memory 43 comprises two systems for each line, and if one of them is used in a n input mode, the other assu mes an output mode. The output picture image, therefore, is delayed in output by one line than the input picture image.

A chain of operation of the computer 51, the micro processor 53 and thetiming control circuit 55 will now be described with reference to FIG. 18.

Atthefirst step S1, the standard color processing conditions incorporated to the color processing circuit42 are set with the console 50 byan operator, and then stored in the memory in the computer 51. At the next processing step S2, graphic pattern inputto decidethe lay-out of the output picture images is carried out by using the digitizer20 and the console 50, and the input data are stored in the memory in the computer 51 as well as displaying on the graphic display 52. Next, base manuscript input is carried out bythe digitizer 20 and the console 50 in the same manner. Processing of this base manuscript input comprises trimming inputfor inputting the trimming conditions of output image position and output magnification forthe input original images and processing condition inputfor inputting the colour and the gradation processing conditions of output picture images. The input result is memorized in the memory in the computer 51 as well as displaying on the graphic display 52.

Then, the computer 51 produces the data of the scanning lines. This producing process is conducted by inputting the trimming conditions obtained at the above mentioned step S3 and the memory data memorized atthe step S2, and the scanning-line data obtained as listed in Table 3 are stored in the memory in the computer 51. Atthe processing step S5, the standard conditions which are stored atthe proces- sing step S1 in advance are read out and are transmitted to the color processing circuit 42through the micro processor 53. Atthe same time, the scanning-line data which are produced atthe processing step S4are read outfromthe memory and then aretransmitted to thetiming control circuit 55 through the micro processor53. As a result of this processing, the timing control circuit 55transmitsthe pulse signal to the pulse motor 14and drives it, thereby moving x-position of the input drum 10. At loothistime, y-position of the inputdrum 10 is detected bythe rotary encoder 13, and the detected data are stored in the address register inthetiming control circuit 55. The timing control circuit 55,therefore, is able to control the position forthe input drum 10 of the read-head 16, and the color-separation signal data of the read-head 16 are successively stored in the memory 43. The position data stored in the memory 43 are transmitted to the computer 51 through the micro processor 53 and to stored in the memory of the computer 51, and the above-mentioned processing is repeated at the times corresponding to the numberof the input original pictures mounted on theinput drum 10. The processing conditions used atthe next processing step S7 is obtained according to FIG.

19 as described hereinafter by using the picture image data obtained at the above-mentioned rough scanning and the processing conditions produced at the step S3, and the obtained processing conditions are stored in the memory in the computer 51.

The color and the gradation processing conditions which are stored atthe step S6 are read and are transmitted to the color processing circuit 42 through the micro processor 53, respectively, and furtherthe scanningline data obtained at the step S4 are transmitted to the timing control circuit 55 through the micro processor 53. Thetiming control circuit 55 transmits the pulse signalsto the pulse motor 14 and 33 and moves the read-head 16 and the output-head 32 to x- direction and X-direction, respectively. The condition setting atthe processing step S6 and fine 11 0 GB 2 182 524 A 11 scanning atthe processing step S7 are repeatedly executed atthe times corresponding to the number of the input original pictures mounted on the input drum 10.

Although the inputoriginal pictures comprise 4 types,Ato D, in the above embodiment, the shape or number may be chosen arbitrarily and the lay-out of the output picture image may also be arbitrarily inputted. Although color correction and gradation conversion are carried outdigitally in the above embodiment,they may be conducted analogously.

Although in the above mentioned embodimentthe scanning speed on the inputside is varied for magnifying the picture image output,the scanning speed on the output side may be varied instead. 80 The picture image signal which has been AD converted by the AD converter 41 is color-processed in the color processing circuit42 and thus color processed signal is stored in the memory 43 in the above embodiment, butthe AD-converted picture image signal may be stored in the memory, and color-processed in a color processing circuitwhen it is outputted.

An automatic setting method for picture image processing (such as for color, sharpness and grada tion) in the picture image inputloutput systems according to this invention will now be described.

In orderto automatically set above conditions, two types of data, i.e. the attribute information forthe input color original picture and rough-scanned data are used. The attribute information is inputted according to the base manuscript input step which relates each original picture with the position and magnification of output graphic patterns. In these steps, the digitizer 20 and a menu sheet placed thereon or a functional key board are used to input such data in orderto determine the type of photo 9 raphic materials, image type, highlight point coor dinates, shadow point coordinates, skin-color point coordinates, g ray point coordinates, background color point coordinates, color fogging coordinates, color correction amount, unsharp mask amount, the curve to be chosen from the preset gradation setting curves. The rough-scanned data comprises 13(blue), G(green) and R(red) densities of picture elements for each original color pictu re obtained in the manner described hereinafter. When the base manuscript 11 is mounted on the input drum 10, the computer 51 prepares a drum position control information as shown in FIG. 13 and Table 2 as the outputscope of each original color picture has been determined by the coordinate supervising method. The sampling interval for rough-scanned data may be set at 500 [pm]. If the interval is set at a value too small such as 50 [[im], the number of picture element data becomes 120 too large, providing disadvantageous operation timewise. The picture element data thus sampled are stored in an outside memorysuch as a magnetic disk of a computer system. Sophisticated skill is not required forobtain[ng such attribute information and rough-scanned data of original color pictures. Any workercan betrainedto conduct such an operation.

Advantages of using the attribute information of each original color picture are now discussed; (1) Re: Photographic material of color manuscript As spectroscopic property and base density of hues of respective photographic material vary, the color processing parameters should be adjusted for each material.

(2) Re: Imagetype Parametersfor gradation, color and sharpness processing vary depending onthetypeof imagesof a picture; such asa portrait, scenery, still lifeJor instance, processing fora too strong sharpness would notbe preferablefora picture centering arounda person because grains on the skin become toocoarse.Such a pictureshould beprocessedwith a lesserdegreeof sharpness. Parameter computation mayvaryforeach of the classified images. For instance, in an image centering around a person, parameters should be selected so as to emphasize tone reproduction mainly on the skin portion of the picture. They may be selected to emphasizetone reproduction of overall picture for other images. It would be very difficuitto judge the pattern from the picture element data of the picture image, and errors may occur even if such a technique as pattern recognition is used. Butthistype of data can be obtained instantaneously if an operator looks atthe picture. Thistype of data therefore should be inputted by an operatorto reduce mistakes and time. (3) Re: Positional coordinates of highlight points and shadow points If the highlight point position coordinates of a color manuscript is inputted, the density value corresponding to those coordinates can be selected by computation out of the picture image data which have been inputted and scanned to be set as a highlight setting density. The density can be set similarly for shadow points, too. Gradation characteristic can be varied by selecting the density for hig hlight points and shadow points out of the picture image in this manner. (4) Re: Positional coordinates of skin color point, gray point and background color point

The density of skin color point and background color point can be obtained respectively in the manner mentioned above. Those densities are used to selectthe color processing and gradation processing parametersfor reproducing the skin and gray color point on the output picture image, thereby remarkably improving the quality of the output picture image. Similar advantages are observed in respect of the background colors. Parameters can be selected so asto process the gradation in a manner notto intensify a particular background color, thereby enhancing the gradation reproduction of essential portions of a color manuscript. (5) Re: Colorfogging amount

Gradation conversion is conducted between input and output by inputting the colorfogging amount of the input color manuscript. Gradation conversion parameters are selected to maintain gray-balance of the output picture image. (6) Re: Selection of gradation conversion curves An operator looks at a color manuscript and inputs a curve approximated to a preferable gradation conversion characteristic. By this input, a picture image closerto the instruction given by an operator can be outputted than thatwhen the gradation is automatically set, thereby improving the quality of 12 GB 2 182 524 A 12 the output picture image. (7) Re: Unsharp mask amount (USM) An operator inputs an unsharp mask amount which he desiresto add to the color manuscript. The picture image which has a desired sharpness will be outputted. (8) Re: Color correction amount This is the parameterto designate the degree of color correction. The degree of sharpening hues of the color manuscript can be varied bythe color correction parameters.

FIG. 19 indicates how input information is processedto setcondition parameters.The color/gradation processing method used herein as an example is disclosed in German Patent Application P33 13 392.1orBritish Patent Application No. 8310010. "END" stands for Equivalent Neutral Density.

Referring to FIG. 19,the three-coior density corres- ponding to the skin coior point coordinates in the rough-scanned data is obtained from the skin color point coordinates and the rough-scanned data. It is preferable to computethe three-color density as the average of rough-scanned data nearthe skin color point coordinates. In a manner similarto above, the densityfor highlight, shadow, gray and background colors is computed, but if the coordinate input has not been carried out, such a density computation is omitted. In the END setting process, where a photo- graphic material of a color manuscript is inputted and an END-matrix is outputted, as different END-matrices existfor different photographic materials, it is preferableto obtain END-matricesfor respective photographic materials in advance and to resist them. In this step, a photographic material type is inputted and a registered END-matrix is retrieved. In the USM computing process, an unsharp mask amount and the pattern are inputted and if the unsharp mask amount is designated, priority is placed on the unsharp mask amount ratherthan the magnification and the pattern to compute USM condition-setting parameters. If the unsharp mask amount is notdesignated, theparameters forsetting USM conditions is computed from the magnification and the pattern. Then the result of computation for the pattern and the skin color density is inputted for processing the skin color cumulative histogram. But this computing process for skin color cumulative histogram is executed only when the skin color coordinates are indicated and when the pattern is instructed to focus mainly on a person, although the skin color is not indicated. In othercases computation forthe skin color cumulative histogram is no operated. A cumulative histogram is computed out of the skin color pointdata extracted from the roughscanned data bysetting a center on the result of the computation forthe skin color density if there is a skin color point coordinate indication, and on a predetermined value if there is notsuch an indication according to the method using the probability ellipse disclosed in Japanese Patent Laid-open No. 1566241 1977 and No. 15662511977 orthe method using a rectangle centered on the above mentioned value. Thetotal cumulative histogram is processed by in putting the computation result of the background color density and the rough-scanned data, but if there is no indication forthe background color points, all of the rough-scanned data are computed. If there is such an indication, the cumulative histogram is computed out of the rough-scanned data minus the background color data in a manner similarto that forthe skin color extraction. In the case of the computation for highlight point and shadow point, the results of the total cumulative histogram computation as well as the computation for highlight density point and shadow point density are inputted. When highlight point coordinates and the shadow point coordinates are indicated, the highlight points and shadow points are determined by the result of the density computa- tion, where if there is not such an indication, highlight point/shadow point maybe computed by, for instance, setting the density equivalentto 1 %of the total cumulative histogram as the highlight density and that to 99% as the shadow density. The colorfogging computation is operated by inputting the result ofthe gray density computation and the color fogging amount. If there is no indication for both gray point and the colorfogging, the colorfogging is assumed notto exist and the computation proceeds. If there is indicated a colorfogging amount, however, the amount of parallel displacement of the gradation curves is decided so asto correctthe colorfogging. If there is an indication fora gray point coordinates the amount of parallel displacement, is decided to make the combination of the densities of the result of gray density computation gray. The gradation computing process is divided into two steps, i.e. that for setting a gradation conversion parameter and that for preparing gradation table; as the gradation conversion parameters highlight density, shadow density and a curve number are used. The highlight density and the shadow density are obtained from the result of the computation for highlight point and shadow point. The pattern, the skin color cumulative histogram, the total cumulative histogram and thecurve number are inputted, and discriminated to selectthe most preferable curve for gradation reproduction outof several tens of standard curves which are preset to produce a gradation table by linear transformation (parallel displacement and enlargement/reduction) on thus selected curve using the result of the color fogging amount computation, highlight density and shadow density data.

The method of preparing the gradation table is described hereinbelow.

FIG. 20A indicates the group of standard curves which have been set in advance while FIG. 20B indicates a standard curve of fo (D) which is selected bythe method described above. The one-dot chain in FIG. 20C representsthe standard curvefo (D) while the solid line representsthe curvefo (aD+b) which is obtained by linear-transforming the input side of the standard curve from the highlight density and the shadow density data. In FIG. 20Cthe curves fo (D) and fo (aD+b) take an identical value dH at respective highlight densities DHO and DH, and an identical value ds at respective shadow densities Dso and Ds.

fo (DHO) = fo (aDH +b) = dH fo (Dso) = fo (a Ds +b) = ds ..... (17) ..... (18) 11 1 13 GB 2 182 524 A 13 There holdsthe above relation expressed bythe formulae. Then the formulae (19) and (20) subsequently hold.

aDH +b = DHO aDs +b = Dso ..... (18) ..... (20) Out of the aboveformulae, coefficients a and b are 65 computed as shown below:

4.

p a=(Dso -DHOffilDs -DH) (21) b=(DHO=Ds-DH-Dso)/(DS-DH) (22) The above coefficients a and b determined by the above formulae (21) and (22) are computed from the highlight density DHO and shadow density Dso of the standard curve and a gradation table g(D) is obtained 10 according to the following formula.

9(D)=fo(aD+b) ..... (23) In this manner, a gradation table which maintains the characteristic of standard curves and yet posses ses desired highlight density and the shadow density is obtained.

The colorcorrection computing process is ex ecuted by inputting the results of computation for images, colorcorrection and gradation. In other words, color correction parameters are selected to makethe colorsharpfor respective cases. The degree of colorcorrection must be determined by consider ing the particular conditions as some pattern needs sharper colorwhile others not. Further, a color may become turbid by gradation conversion which some times increasesthe output picture image density. The colorcorrection amount data thus determined is transformed into a color correction parameter. The colorcorrection parameters are computed to obtain a weighted mean using the results of the pattern, color correction amount and the gradation processing 95 finally.

Parameters of the color correction, the enhance ment of sharpness and gradation arethus deter mined. Each parameter is set in the computer 51 via the micro processor 53for each unit of the picture image outputon the output drum 30 either im mediately beforethe output orthe storage in the memory43. Then the inputloutput drums are control led in a manner described in theforegoing. Although one graphic pattern is singly controlled, as such controlling operation can be sequentially and con tinuously conducted, lay-out picture images are automatically outputted without requiring the in tervention of an operator.

As described in theforegoing, the system accord ing to the present invention does not need the preparation of the rough sketch base paper and masks, stripping after registering (positioning) and the multiple exposure which have been heretofore conducted manually. It can automatically control all of the operation sequentially for each of the output graphic patterns by inputting process condition parameters of the each of original color pictu re and the positional data of the original color picture mounted on a transparent base and the graphic pattern data,thereby remarkably saving labor, time and resources. Unlike the total lay-out re-touch system, it does not require a large scale exterior memory andlor a high speed central processing unit for editing and yet can construct an excellent system capable of a higher performance at a lower cost. The description above concerns the output of a color picture image, butthe system may be applied for a processing step at a printing plant by color separating again the color picture outputs in a color scannerto obtain a colorseparated film. The color picture image per se can be applied for graphic artfield aswell as various otherfields.

As described in detail in theforegoing, as the positioning method for inputting original picture according tothis invention comprises the steps of placing an input original picture on a transparent base, providing register pin holes on the transparent base while providing pins on an input drum and a digitizer and engaging the pins with the pin holes, it is possibleto make the coordinates on the input drum to correspond with the coordinates on the digitizer of the input original picture correctly and easily, thereby facilitating the coordinates controlling.

Asthe base position which is indicated by absolute address is set on a graphic pattern and the picture elements on the periphery of the pattern are indicated by relative address in respect of the base position, the coordinates of graphic data become controllable and

Claims (2)

the capacity of the memoryto store the graphic pattern data can be reduced. CLAIMS

1. A method of correlating two sets of coordinates in a picture image inputloutput apparatus which comprises:

a) a digitizer board for inputting graphic information; b) a console which inputs necessary information and operation commands; c) a read means which optically reads original pictures mounted on a predetermined position on an input drum; d) a color processing section which stores the picture image data of said original pictures; e) a picture image output means which outputs picture images using the picture image data from said color processing section on a recording material mounted on an output drum; and f) a computer system which is coupled respective- lyto said digitizer board, and said console; wherein said method comprises coordinates of the digitizer board with the coordinates of the input drum,the method comprising positioning an input original picture of a flexible basefor mounting said flexible base on said inputdrum or digitizer board, wherebysaid base is positioned at predetermined positions on said inputdrum and on said digitizer board, and said computersystem effects a correlation between the coordinates of thetwo positions of the base.

2. A method as claimed in claim 1, wherein said base is formed with a transparent material on a square sheetform.

Printed in the United Kingdom for Her Majesty's Stationery Office by the Tweeddale Press Group, 8991685, 5187 18996. Published at the Patent Office, 25 Southampton Buildings, London WC2A lAY, from which copies may be obtained.

i 4 i

2. A method as claimed in claim 1, wherein said base is formed with a transparent material in a square sheetform.

3. A method as claimed in claim 1 or 2, further including a method of inputting a graphic pattern including the step of inp;utting graphic patterns which receives graphic patterns from said digitizer board 14 GB 2 182 524 A 14 and corresponds said digitizer board with the pictures on said output drum, and the step of baseman uscript input which corresponds said original pictures with the pattern inputted by said step of inputting patterns 5 in position and magnification.

4. A method as claimed in claim 1, substantially as described with reference to Figures 5 to 8 of the accompanying drawings.

Amendmentsto the claims have been filed, and have the following effect:(a) Claims 1-2 above have been deleted or textually amended. (b) New ortextually amended claims have been filed asfollows:- CLAIMS 1. A method of correlating two sets of coordinates in a picture image inputloutput apparatus which comprises:

a) a digitizer board for inputting graphic informa- tion; b) a console which inputs necessary information and operation commands; c) a read means which optically reads original pictures mounted on a predetermined position on an inputdrum; d) a color processing section which stores the picture image data of said original pictures; e) a picture image output means which outputs picture images using the picture image data from said color processing section on a recording material mounted on an output drum; and f) a computer system which is coupled respectivelyto said digitizer board, and said console; wherein said method comprises correlating the coordinates of the digitizer board with the coordinates of the input drum, the method comprising positioning an input original picture on a flexible base having register pinholes provided at plural positions thereon, register pins being provided on said input drum and digitizer board for engaging with said register pinholes for mounting said flexible base on said input drum or digitizer board, whereby said base is positioned at predetermined positions on said input drum and on said digitizer board, and said computersystem effects a correlation between the coordinated of thetwo positions of the base.