JPS62128374A - Picture processing method - Google Patents

Picture processing method

Info

Publication number
JPS62128374A
JPS62128374A JP60269881A JP26988185A JPS62128374A JP S62128374 A JPS62128374 A JP S62128374A JP 60269881 A JP60269881 A JP 60269881A JP 26988185 A JP26988185 A JP 26988185A JP S62128374 A JPS62128374 A JP S62128374A
Authority
JP
Japan
Prior art keywords
picture
image
address
original
digital
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
JP60269881A
Other languages
Japanese (ja)
Inventor
Miyuki Enokida
幸 榎田
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Canon Inc
Original Assignee
Canon Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Canon Inc filed Critical Canon Inc
Priority to JP60269881A priority Critical patent/JPS62128374A/en
Priority to US06/902,320 priority patent/US4850028A/en
Priority to DE19863629984 priority patent/DE3629984A1/en
Publication of JPS62128374A publication Critical patent/JPS62128374A/en
Pending legal-status Critical Current

Links

Landscapes

  • Image Processing (AREA)

Abstract

PURPOSE:To obtain a picture having high quality with high speed by obtaining a coordinate corresponding to the original picture from the address of a generated picture, and obtaining the data of the position of the said coordinate from the original picture. CONSTITUTION:By using a digital differential analyzing method, the address of the main scanning line of a converted coordinate system is obtained. Then the coordinates corresponding to the said coordinate system of the original picture's coordinate system is obtained from the address of the picture center of a converted picture by using such a method that two suites of incremental differential values are selectively taken to be repeatedly added up. The weighted average over four or two picture-element digital values whose weights are the distances from such four or two picture-element centers of the original digital picture that surround the already obtained coordinates, is obtained. And the resulting values are made the values of the picture elements of the digital picture. As a result, no deterioration in the quality of picture occurs even after operating an optional revolution. Also, necessary mechanisms are made least in quantity. An eventually, the picture having the high quality can be obtained at the high speed.

Description

【発明の詳細な説明】 く分  野〉 本発明は画像の回転、変倍等を行う画像処理方法に関す
る。
DETAILED DESCRIPTION OF THE INVENTION Field: The present invention relates to an image processing method for rotating, scaling, etc. an image.

〈従来技術〉 従来、この種の装置は2値画像、特に黒白2 イ++’
i画像に応用されるようなものであったため、第2図の
ように1欣密には同じ位置にない画素、第2図において
は画素dと画素eの画素値を求める時、複雑な計算式を
使って求めたり、一定の増分(triを緑り返し加える
ことにより値を求めるデジタル微分解析法を用いて求め
ていた。したがって、このような方法を多値のカラー画
像処理等の高い画質が要求されるような処理装置に適用
すると、複雑な計算をハードウェアで実現する場合は、
ハードウェアが大きくなり、一定の増分値を繰り返し加
えるデジタル微分M’析法を用いる場合は+Bt差が大
木発明は+1ili述従来例の欠点を除去し、任、念の
回転の操作後も、画質の劣化を起こさず、そのために必
要な機構も極小とすることも可能になった画像処理方法
の提供を目的としている。
<Prior art> Conventionally, this type of device has been used to produce binary images, especially black and white 2 i++'
Since it was applied to i-images, complex calculations were required when determining the pixel values of pixels that are not exactly in the same position as shown in Figure 2, pixel d and pixel e in Figure 2. It was calculated using a formula, or using a digital differential analysis method that calculates the value by repeating and adding a certain increment (tri). When applied to processing devices that require
As the hardware becomes larger and a digital differential M' analysis method that repeatedly adds a constant increment value is used, the +Bt difference is +1ili. The purpose of the present invention is to provide an image processing method that does not cause deterioration of images and can minimize the necessary mechanisms.

〈実施例〉 本実施例は、元デジタル画像と結果デジタル画像との間
で、7iいに画素中心が重なり合わないような変換が行
なわれる場合、すなわち画像の回転を伴う拡大及び縮小
において、元の画像に対し、結果画像のサンプリングの
点が互いに異なっているような変換が行なわれる場合に
、結果デジタル画像を元デジタル画像に忠実に変換され
た画像にすることにある。再サンプリングの方法はいく
つか考えられるが、実施例では、まずデジタル微分解析
法を用いて、変換後の座標系の主走査線のアドレスを求
める。次に2組の増分(1rJを使い分けながら、繰り
返し加算していく方法を用いて、変換後の画像の画素中
心のアドレスから、元画像の座標系におけるそれぞれ対
応した座標を求める。ここで求められた座標を囲む4つ
あるいは2つの元デジタル画像の画素中心からの距離を
重みとするそれら4つあるいは2つの画素デジタル値の
加重平均を求め、その値を求める結果デジタル画像のそ
れぞれの画素の値とする。以下図を用いて説明するが、
説明のための実施例では簡単なために、直線の等倍な原
点を中心とする角度30”の回転を考える。
<Example> In this example, when conversion is performed between the original digital image and the resulting digital image in such a way that the pixel centers do not overlap, that is, when enlarging and reducing the image with image rotation, the original To convert a resulting digital image into an image that is faithfully converted to the original digital image when conversion is performed on an image such that the sampling points of the resulting images are different from each other. Although several resampling methods are conceivable, in the embodiment, first, a digital differential analysis method is used to obtain the address of the main scanning line in the converted coordinate system. Next, using a method of repeatedly adding two sets of increments (1rJ), calculate the corresponding coordinates in the coordinate system of the original image from the address of the pixel center of the converted image. The weighted average of the four or two pixel digital values surrounding the coordinates of the four or two original digital images, using the distance from the pixel center as the weight, is calculated, and the resulting value is the value of each pixel in the digital image. As explained below using the diagram,
In the illustrative example, for the sake of simplicity, we will consider a rotation by an angle of 30'' about the origin of the same size of the straight line.

第3図は、本実施例により画像データがどのように変換
されるかを示したものである。第3図には(A)、(B
)、(C)の3つの図があるが、各々の横軸20,21
.22はそれぞれの主走査方向を示し、白丸はサンプリ
ング点を、縦軸は画素の濃度等の多値データの値を示し
ている。図(A)は元デジタル画像、図(C)は本方法
で行なった場合の結果デジタル画像、図(B)は元デジ
タル画像を補間を用いず回転したもので、従来技術の結
果の一例である。これから図(C)のサンプリング点2
3の多値データの値を与える実施例を順を追って説明す
る。
FIG. 3 shows how image data is converted according to this embodiment. Figure 3 shows (A), (B
), (C), each horizontal axis 20, 21
.. 22 indicates each main scanning direction, white circles indicate sampling points, and the vertical axis indicates values of multi-value data such as pixel density. Figure (A) is the original digital image, Figure (C) is the resulting digital image obtained by this method, and Figure (B) is the original digital image rotated without using interpolation, which is an example of the results obtained using the conventional technique. be. From now on, sampling point 2 in figure (C)
An example of providing three multivalued data values will be explained in order.

まず、第3図(C)の結果画像の主走査線22のアドレ
スを与える実施例を第1図に示す。
First, FIG. 1 shows an embodiment in which the address of the main scanning line 22 of the resultant image shown in FIG. 3(C) is given.

この実施例では、F1点を中心とする30°の回転なの
で結果レジスタ30の整数部33と小数部32にOを、
増分レジスタ34にはj an30°=0.57735
0という値がセットされている。又、結果画像の主走査
線の画素数がカウントレジスタ35にセットされる。こ
れらハQ’S 1図には1図示されない装置によって、
ハードウェア的に又はソフトウェア的に計算され。
In this example, since the rotation is 30 degrees around point F1, O is set in the integer part 33 and decimal part 32 of the result register 30.
Increment register 34 has j an30°=0.57735
A value of 0 is set. Further, the number of pixels in the main scanning line of the resultant image is set in the count register 35. These Q'S are carried out by a device not shown in Figure 1.
Calculated by hardware or software.

セットされている。37.38に同じクロックを7jえ
ることにより、増分レジスタ34と結果レジスタ30の
小数部32が加算され、その時に結果レジスタ30の整
数部33のイ+tJ 3 gが主走査線22のY軸方向
のアドレスとしてi4図に示すように0 、1 、1 
、2、−−−−一と出力される。
It is set. By adding the same clock to 37.38, the increment register 34 and the decimal part 32 of the result register 30 are added, and at that time, the integer part 33 of the result register 30, i+tJ 3 g, is added in the Y-axis direction of the main scanning line 22. As shown in the i4 diagram, the addresses are 0, 1, 1
, 2, ----1 are output.

ここで近似される直線は第10図で100に示すように
1つのX方向のアドレスには必ず1つのY方向のアドレ
スが決′まるような軌跡となる。
The straight line approximated here becomes a locus such that one Y-direction address is always determined for one X-direction address, as shown at 100 in FIG.

この演算回路は加算回路だけで構成されるため、回路コ
ストの低下と、十分に速いスピードを得ることが出来る
Since this arithmetic circuit is composed only of adder circuits, it is possible to reduce the circuit cost and obtain sufficiently high speed.

次に第1図のキャリ31からの信号を入力して元画像の
座標系における変換後の画像の画素中心に対応する座標
を求める実施例を′1iIJ5図に示す。
Next, an embodiment in which a signal from the carrier 31 of FIG. 1 is inputted and coordinates corresponding to the pixel center of the transformed image in the coordinate system of the original image are determined is shown in FIG.'1iIJ5.

第1図31のキャリ信号が0、すなわちY座標方向のア
ドレスが変化しない時のX、Yの値xo =l*cos
30’ +0esin30”=Q 、866025 Y0=1 *C−s i n30’ )+0− co 
s30’=−0゜5 をそれぞれの増分レジスタ510 、530に、キャリ
信号が1、すなわちY座標方向のアドレスがlふえる時
のX、Yの値 X、=1*cos30’+l会5in30’=1.36
625 Yl =l*−5in30@+l*cos30’=0.
36625 をそれぞれの増分レジスタ520,540に。
The values of X and Y when the carry signal in FIG. 131 is 0, that is, the address in the Y coordinate direction does not change, xo = l * cos
30'+0esin30''=Q, 866025 Y0=1 *C-sin30')+0-co
s30'=-0°5 is stored in the respective increment registers 510 and 530, and the value of X and Y when the carry signal is 1, that is, the address in the Y coordinate direction increases by 1, is 1*cos30'+15in30'= 1.36
625 Yl =l*-5in30@+l*cos30'=0.
36625 into their respective increment registers 520, 540.

第5図には図示されない装置によって、ハードウェア的
に又はソフトウェア的に計算されて、セットされている
ものとする。またX、Yの結果レジスタ550,560
には今回の実施例は原点を中心とする回転であるので、
それぞれOがセラ)・されている。この状態の回路に第
1図の同期信号37.38と同じクロック信号を入れる
ことにより、第1図31のキャリの信号に伴なってX、
Yそれぞれの増分レジスタが結果レジスタに足し込まれ
元画像でのX、Y座標の値が590゜591から逐次的
に第6図に示すタイミングチャートのように出力される
。この@算回路も、加算回路だけで構成されるため、回
路コストの低丁と十分に速いスピードを得ることが出来
る。
It is assumed that the values are calculated and set by hardware or software by a device not shown in FIG. Also, X, Y result registers 550, 560
In this example, rotation is centered around the origin, so
Each O is Cera). By inputting the same clock signals as the synchronization signals 37 and 38 in FIG. 1 to the circuit in this state, X,
The Y increment register is added to the result register, and the values of the X and Y coordinates in the original image are sequentially output from 590°591 as shown in the timing chart shown in FIG. Since this @arithmetic circuit is also composed of only an adder circuit, it is possible to obtain low circuit cost and sufficiently high speed.

最後に第5図のX、Y結果レジスタから出力されるX、
Y座標590,591の値を入力して近傍2点での加重
平均を求める時の概念図を第7図に示す、第7図中のm
は、第5図の590で出力されるX座標の小数部である
。そこで座標70の多値デジタル値Yは元画像の画素中
心71.72の多値デジタル値をそれぞれX7.、Xg
とするとY7o=m@X71+ (1−m) ・Xgの
式で求めることができ、」二式のX71.x7.lに第
2図(A)の24.25の多値デジタル値を代入するこ
とにより、第2図(C)の画素23の多値デジタル値を
求めることが出来る。
Finally, the X, which is output from the X, Y result register in Figure 5,
Figure 7 shows a conceptual diagram when entering the values of Y coordinates 590, 591 and calculating the weighted average at two neighboring points.
is the decimal part of the X coordinate output at 590 in FIG. Therefore, the multi-value digital value Y at the coordinate 70 is the multi-value digital value at the pixel center 71.72 of the original image, respectively. ,Xg
Then, Y7o=m@X71+ (1-m) ・Xg can be obtained using the formula, and the two equations x7. By substituting the multi-value digital value of 24.25 in FIG. 2(A) to l, the multi-value digital value of pixel 23 in FIG. 2(C) can be obtained.

上記実施例は直線の等倍な回転であったが、画像データ
の回転は、第8図に示すように元画像の主走査方向のラ
インバッファを3本持ち、変換後の主走査方向の直線の
近似値を求める回路をfflll図のようにし、結果レ
ジスタ113から出力されるキャリ信号llOをカウン
トレジスタ114の値をアドレスに直しメモリ112上
のアドレス信号111が指す番地にセーブし、セーブさ
れたキャリの値を第5図に入力し、結果デジタル画像の
近傍補間を第9図に示すように2次元で考えることによ
り、近傍wJ間は Y90= (n ’ X91 + (1−n)  ψX
g3) e m+(n e Xg2+(1−n) ” 
Xg、) e (1−m)の式で求めることが出来る0
元画像での座標を求め、近傍補間する処理を変換後の画
像の副走査方向の画素数回繰り返すことにより簡単に対
応出来る。
In the above embodiment, rotation of the straight line was performed at the same magnification, but as shown in Fig. 8, the rotation of the image data is performed using three line buffers in the main scanning direction of the original image, and a straight line in the main scanning direction after conversion. The circuit for obtaining an approximate value of By inputting the carry value in Figure 5 and considering the neighborhood interpolation of the resulting digital image in two dimensions as shown in Figure 9, the distance between the neighborhoods wJ is Y90 = (n'X91 + (1-n) ψX
g3) e m+(ne Xg2+(1-n)”
Xg,) e (1-m)
This can be easily handled by repeating the process of finding the coordinates in the original image and performing neighborhood interpolation several times for the pixels in the sub-scanning direction of the converted image.

また回転角45@〜360°は第12図(B)の手法に
より、O°〜45@までの変形と考えることが出来るた
め、第11図の回路で生成するキャリのbit列をテー
ブルでメモリ上に持つことにより、処理スピードを上げ
ることも出来る。
Also, since the rotation angle 45@~360° can be considered as a transformation from 0°~45@ using the method shown in Figure 12(B), the bit string of the carry generated by the circuit in Figure 11 is stored in the table. By holding it above you, you can also increase the processing speed.

今回の実施例は等倍の回転を考えたが、拡大の時は元画
像の画像データの入力タイミングを倍率により遅くする
ことによって対応でき、縮小の時は第8図に示すライン
バッファの数を増すことによって対応出来る。
In this example, rotation at the same magnification was considered, but when enlarging, it can be handled by delaying the input timing of the image data of the original image depending on the magnification, and when reducing, the number of line buffers shown in Fig. 8 can be adjusted. This can be addressed by increasing the amount.

く効  果〉 以上の様に本発明の画像処理方法は生成画像のアドレス
から元画像の対応する座標を求め、その座標位置のデー
タを原画像から求める様に構成したものである。かかる
構成により、部用な演算式で回転後の座標位、’、!i
の正確なデータを求めることができる。従って回転に依
っても画像劣化の少い高品位の画像を高速に得ることが
できる。
Effects> As described above, the image processing method of the present invention is configured to obtain the corresponding coordinates of the original image from the address of the generated image, and to obtain the data of the coordinate position from the original image. With this configuration, the coordinate position after rotation can be calculated using a simple arithmetic expression, ',! i
Accurate data can be obtained. Therefore, high-quality images with little image deterioration can be obtained at high speed even when rotated.

【図面の簡単な説明】[Brief explanation of drawings]

第1図は本実施例のアドレス生成回路の前段の回路図、 第2図は従来法による画像の回転の説明図。 第3図は従来法と本実施例による1度データの変換の比
較を示す図、 第4図は第1図のタイミングチャートを示す図、 fJS5図はアドレス生成回路の後段の回路図。 第6図は第5図のタイミングチャートを示す図。 第7図は加重平均を行う時の模式図、 第8図は画像を扱う場合の装置構成図。 第9図はその時の加重平均の模°式図、ffllo図は
第1図で求める近似された直線の軌跡を示す図、 第11図は第9図の変換を行う構成例を示す図、 第12図(A)、(B)は座標変換の説明図である。
FIG. 1 is a circuit diagram of the front stage of the address generation circuit of this embodiment, and FIG. 2 is an explanatory diagram of image rotation according to the conventional method. FIG. 3 is a diagram showing a comparison of one-time data conversion according to the conventional method and this embodiment. FIG. 4 is a diagram showing the timing chart of FIG. 1. FIG. fJS5 is a circuit diagram of the subsequent stage of the address generation circuit. FIG. 6 is a diagram showing the timing chart of FIG. 5. Figure 7 is a schematic diagram when performing weighted averaging, and Figure 8 is a diagram of the equipment configuration when handling images. Figure 9 is a schematic diagram of the weighted average at that time, the ffllo diagram is a diagram showing the locus of the approximate straight line obtained in Figure 1, Figure 11 is a diagram showing an example of a configuration for performing the conversion in Figure 9, FIGS. 12A and 12B are explanatory diagrams of coordinate transformation.

Claims (4)

【特許請求の範囲】[Claims] (1)画像の回転を行う画像処理方法において、生成さ
れる画像の主走査方向の直線を近似 するために、生成される画像の主走査方向の1つのアド
レスに対し、必ず副走査方向のアドレスが1つ決まるよ
うにして求め、そこで求めたアドレスから元画像で対応
する座標を求め、その座標から、生成される画像の画素
データを元画像の画像の画素データの近傍 補間から求めることを特徴とする画像処理方法。
(1) In an image processing method that rotates an image, in order to approximate a straight line in the main scanning direction of the generated image, one address in the main scanning direction of the generated image must always have an address in the sub-scanning direction. is determined so that one address is determined, the corresponding coordinates in the original image are determined from the address determined, and from those coordinates, the pixel data of the image to be generated is determined by neighborhood interpolation of the pixel data of the original image. Image processing method.
(2)直線近似のためのアドレスを求める時は、デジタ
ル微分解析法を用い、元画像での座標を求める時は、2
つの増分レジスタを切り替えながら加算する回路を有す
る第1項記載の画像処理方法。
(2) When finding the address for linear approximation, use the digital differential analysis method, and when finding the coordinates in the original image, use 2
2. The image processing method according to claim 1, further comprising a circuit that performs addition while switching between two increment registers.
(3)近傍補間のためのパラメータが上記デジタル微分
解析法における小数部保持レジスタの内容である第1項
記載の画像処理方法。
(3) The image processing method according to item 1, wherein the parameter for neighborhood interpolation is the content of a decimal part holding register in the digital differential analysis method.
(4)補間が2点直線的内挿である第1項記載の画像処
理方法。
(4) The image processing method according to item 1, wherein the interpolation is two-point linear interpolation.
JP60269881A 1985-09-04 1985-11-29 Picture processing method Pending JPS62128374A (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
JP60269881A JPS62128374A (en) 1985-11-29 1985-11-29 Picture processing method
US06/902,320 US4850028A (en) 1985-09-04 1986-08-29 Image processing method and apparatus therefor
DE19863629984 DE3629984A1 (en) 1985-09-04 1986-09-03 IMAGE PROCESSING METHOD AND DEVICE

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP60269881A JPS62128374A (en) 1985-11-29 1985-11-29 Picture processing method

Publications (1)

Publication Number Publication Date
JPS62128374A true JPS62128374A (en) 1987-06-10

Family

ID=17478506

Family Applications (1)

Application Number Title Priority Date Filing Date
JP60269881A Pending JPS62128374A (en) 1985-09-04 1985-11-29 Picture processing method

Country Status (1)

Country Link
JP (1) JPS62128374A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH06180580A (en) * 1992-10-01 1994-06-28 Hudson Soft Co Ltd Image processor

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH06180580A (en) * 1992-10-01 1994-06-28 Hudson Soft Co Ltd Image processor

Similar Documents

Publication Publication Date Title
US4667236A (en) Television perspective effects system
JPH09326958A (en) Image processing unit and processing method
JPH03201078A (en) Method and device for hough transform
JPS62128374A (en) Picture processing method
JPH0481231B2 (en)
JPS61184687A (en) Control system of multilevel image magnification and reduction
JPS6125278A (en) Real time three dimensional image process device
US6674922B1 (en) Image processing method, image processing apparatus, and storage medium
JPH104358A (en) Signal processing method and equipment
JPS62139085A (en) Picture processor
JPH0445491A (en) Digital special effect device
JPS6219980A (en) Image processor
JP2712278B2 (en) Wipe pattern generator
JPH02118782A (en) Method for processing picture
JPH05127980A (en) Picture processor
JPS6231482A (en) Picture processor
JPH04241683A (en) Method for correcting picture
JPS60177779A (en) Solid-state image pickup device
JPS63262771A (en) Affine transformation processor
JPS63262772A (en) Affine transformation processor
JPH0863595A (en) Method and device for rotation processing of image
JP2002288649A (en) Image rotation processing method and device
JPS5937770A (en) Picture reduction device
JPS6195478A (en) Coordinate conversion processing device of picture
JPS63161774A (en) Expanding method for binary picture by picture element density conversion