JP3790877B2 - Image processing device - Google Patents

Description

【００４５】
図１０はγ補正・ＰＷＭ制御回路６１のブロック図、図１１はＰＷＭ制御の一例を示す波形図である。Ｃ，Ｍ，Ｙ，Ｋの各色に対応するγ補正・ＰＷＭ制御回路６１〜６４の構成は同一であるので、ここでは代表例としてＣに対応するγ補正・ＰＷＭ制御回路６１を説明する。
【００４６】
階調再現回路５０からの画像データＤ５０は、γ補正テーブル６０１で補正されるとともに８ビットのデータに変換される。そして、Ｄ／Ａ変換器６０２を経てアナログの濃度データとなって２個の比較器６０４，６０５に共通に入力される。一方の比較器６０４には２画素周期の三角波信号がリファレンス信号Ｓｒｅｆ１として入力され、他方の比較器６０５には１画素周期の三角波信号がリファレンス信号Ｓｒｅｆ２として入力される。各比較器６０４，６０５は、濃度データとリファレンス信号Ｓｒｅｆ１，２との大小判別信号をＰＷＭデータ（パルス幅変調データ）としてセレクタ６０６へ出力する。すなわち、濃度データの値に応じて各画素の露光パターンを決める。セレクタ６０６は、階調再現属性信号に応じて比較器６０４と比較器６０５のどちらかの出力を選択して第２のセレクタへ送る。階調再現属性信号が連続階調部を示す“Ｈ”のときには階調再現性をより滑らかにするために２ｄｏｔ周期のＰＷＭ制御が行われ、エッジを示す“Ｌ”のときには先鋭化に適した１ｄｏｔ周期のＰＷＭ制御が行われる。光変調周波数を自動的に切り換えて画質を向上させるのである。なお、２ｄｏｔ周期のＰＷＭ制御では画像の粒状性が向上するように４５度方向のスクリ−ン角を設けるため、ライン毎にリファレンス信号Ｓｒｅｆ１の位相を半周期ずつずらす。
【００４７】
一方、階調再現回路５０からの画像データＤ５０は強度変調回路６１０にも入力され、信号レベルが濃度に応じて変化する強度変調データに変換される。セレクタ６０７は、セレクタ６０６からのパルス幅変調データと強度変調回路６１０から強度変調データの一方をＣＰＵ３０１の指示に従って選択し、ＬＤ駆動信号Ｄ６１として出力する。このＬＤ駆動信号Ｄ６１を用いて半導体レーザ７１の発光制御が行われる。
〔本発明を適用したデータ処理〕
本実施形態の複写機１においては、Ｃ，Ｍ，Ｙ，Ｋの各色の濃度再現性を一定に保つために、ＡＩＤＣパターンを作成し、その濃度測定の結果を階調再現回路５０における入力階調レベル（２５６段階）と出力階調レベル（８段階）との対応関係に反映させる。具体的には、低値化回路５２０に与える閾値ｔｈ０〜６を変更するとともに、誤差検出テーブル５３０を書き換えて上述した特定入力階調レベルを変更する。電子写真プロセスは温度や湿度などの環境条件に左右される。また、複写機１の内部環境が不均一になることもある。そこで、感光体７１〜７４における主走査方向の複数箇所でＡＩＤＣパターンを作成し、画像全体の再現性を均一化する。
【００４８】
図１２はＡＩＤＣパターン９０の配置の一例を示す模式図である。
ＡＩＤＣパターン９０は、Ｃ，Ｍ，Ｙ，Ｋの４色のそれぞれについて、用紙搬送ベルト３４における主走査方向の一端部と他端部の２箇所に作成される。本例では各色毎に濃度の異なるＡＩＤＣパターン９０が６個ずつ副走査方向に並べて配置され、１回の動作設定に際して形成されるＡＩＤＣパターン９０の総数は４８（＝６×４×２）である。１個のＡＩＤＣパターン９０の大きさは１ｃｍ角程度である。ＡＩＤＣセンサ３７は各箇所に４個ずつ配置され、１つのＡＩＤＣセンサ３７は１色の６個のＡＩＤＣパターン９０の濃度を順に検出する。
【００４９】
図１３は階調変換の設定変更の要領を説明するための図である。
上述のとおり電源投入直後の初期状態において、０〜２５５の２５６段階の入力階調レベルのうち、８段階への変換に際して誤差を生じない特定入力階調レベルは０，３２，６４，９６，１２８，１６０，１９２，２２４である。これら８つの特定入力階調レベルのうちの０及び２２４を除いた６つの特定入力階調レベルについてＡＩＤＣパターン９０を作成し、その濃度を測定する。特定入力階調レベルの場合には誤差拡散処理において周辺画素に振り分ける誤差が０であるので、１つのＡＩＤＣパターン９０を構成する所定数の画素（４００ｄｐｉで約１６０² 個）の露光パターンは全て同一になる。特定入力階調レベルは、疑似階調手法を適用しない場合における出力階調レベル（０〜７）と見ることができる。
【００５０】
複写機１の設計段階で各特定入力階調レベルに対する再現濃度（目標値）が定められており、設計上の階調再現特性は線型特性である。例えば、特定入力階調レベル３２に対する再現濃度の目標値は０．２である。しかし、動作環境の変化や部品の経年変化に因り、実際の再現濃度（測定値）が目標値とずれる場合がある。図１３では主走査方向の一端側（例えば複写機１の前面側）の測定値が黒丸「●」で示され、他端側（背面側）の測定値が黒三角「▲」で示されている。前面側では総じて“濃いめ”に再現され、背面側では総じて“淡いめ”に再現されている（ただし、これは説明のための仮想の例示であって、必ずしも実際にこの特性が現れるとは限らない）。
【００５１】
複写機１は、濃度の測定値に基づいて特定入力階調レベル間の実際の階調再現特性を例えば直線近似し、目標値に対応する推定階調レベルを求める。そして、求めた推定階調レベルを新たな特定入力階調レベルに設定する。図の横軸における丸「○」（前面側）又は三角「△」（背面側）で示される位置の階調レベルが推定階調レベルである。例えば、目標値０．２に対応する前面側の入力階調レベルが２４であったとすると、出力階調レベル１に対する特定入力階調レベルを３２から２４に変更する。なお、前面側と背面側とで独立に特定入力階調レベルを設定してもよいし、前面側及び背面側の測定値の平均を求め、平均値に基づいて特性を近似して目標値に対応する特定入力階調レベルを算定してもよい。
【００５２】
新たな特定入力階調レベルについて改めてＡＩＤＣパターン９０を作成し、その濃度を測定する。目標値と測定値との差が許容範囲内になるまで、又は一定回数に達するまで、濃度測定に基づく特定入力階調レベルの修正を繰り返す。これにより、再現特性を理想に近づけることができる。
【００５３】
このような階調変換の設定変更において重要な点は、実際の再現特性を近似し、近似特性から目標値に対応する入力階調レベルを求める一連の演算に係わる数値（階調レベル及び再現濃度）のビット数を、出力階調データのビット数（本例では３）より多くすることである。仮にＡＩＤＣパターン９０で測定する階調レベルを出力階調データと同じ３ビットの数値（０〜７）とすると、特定入力階調レベルの変更の余地が無くなってしまう。数値演算のビット数が多いほど、特性の近似及び目標値に対応する階調レベルの算定の精度が高まる。例えば階調レベルを０．１刻みで表すようにした場合は、推定階調レベルの演算値が２３．５〜２４．４のときは２４を特定入力階調レベルとし、２４．５〜２５．４のときは２５を特定入力階調レベルとするといった設定が可能になる。
【００５４】
以下、フローチャートに基づいて、本発明に係るプリンタ部３０のＣＰＵ３０１の動作を説明する。
図１４は画像安定化処理ルーチンのフローチャートである。
【００５５】
このルーチンは、電源投入時、電源投入後の最初のコピー終了時、一定枚数のコピーを行った時、一定時間が経過した時、設定時刻など、あらかじめ定められた条件と一致する時に実行される。
【００５６】
ＣＰＵ３０１は、ＡＩＤＣパターン作成（＃１）、再現特性の測定（＃２）、階調レベル設定（＃３）、及びその他の安定化（＃４）の各処理を実行する。その他の安定化処理では、トナー補給や帯電チャージャの出力制御などを行う。
【００５７】
図１５はＡＩＤＣパターン作成サブルーチンのフローチャートである。
階調レベルメモリ３０２（図８参照）から主走査方向の一端側についての現時点の特定入力階調レベルＤｆ₀ 〜Ｄｆ₇ を読み出し（＃１１）、続いて他端側についての現時点の特定入力階調レベルＤｅ₀ 〜Ｄｅ₇ を読み出す（＃１２）。そして、図１２で説明したように主走査方向の２箇所にＡＩＤＣパターン９０を作成する（＃１３）。
【００５８】
図１６は再現特性の測定サブルーチンのフローチャートである。
各ＡＩＤＣセンサ３７の出力を周期的に取り込み、２箇所に作成された各色のＡＩＤＣパターン９０の濃度を測定する（＃２１）。測定結果に基づいて、主走査方向の一端側及び他端側のそれぞれについて、各階調レベル間の再現特性を近似する直線Ｙｆ_j ，Ｙｅ_j を求める（＃２２）。
【００５９】
図１７は階調レベル設定サブルーチンのフローチャートである。
まず、再現濃度の目標値Ｙ₀ 〜Ｙ₇ を設定する（＃３１）。先に求めておいた直線Ｙｆ_j ，Ｙｅ_j と目標値Ｙ₀ 〜Ｙ₇ とから、実際の再現特性に適合する特定入力階調レベルＤｆ₀ ’〜Ｄｆ₇ ’，Ｄｅ₀ ’〜Ｄｅ₇ ’を演算する（＃３２）。そして、得られた特定入力階調レベルＤｆ₀ ’〜Ｄｆ₇ ’，Ｄｅ₀ ’〜Ｄｅ₇ ’を新たな特定入力階調レベルＤｆ₀ 〜Ｄｆ₇ ，Ｄｅ₀ 〜Ｄｅ₇ として階調レベルメモリ３０２に格納する（＃３３）。
【００６０】
図１８は印字処理ルーチンのフローチャートである。
このルーチンは、イメージリーダ部２０からの原稿画像又は外部入力画像の印字要求に呼応して起動される。
【００６１】
主走査方向の前半部の画像データの印字のときには、階調レベルメモリ３０２から一方の特定入力階調レベルＤｆ₀ 〜Ｄｆ₇ を読み出す（＃５１、＃５２）。主走査方向の後半部の画像データの印字のときには、階調レベルメモリ３０２から他方の特定入力階調レベルＤｅ₀ 〜Ｄｅ₇ を読み出す（＃５１、＃５３）。
【００６２】
読み出した特定入力階調レベルに合わせて閾値ｔｈ０〜６を設定して誤差検出テーブル５３０の内容を書き換え（＃５４）、閾値設定レジスタ５２１の値を入れ換える。そして、階調再現回路５０を制御する多値誤差拡散処理（＃５６）、電子写真プロセスを制御する印字制御（＃５７）を実行する。
【００６３】
上述の実施形態によれば、低値化の閾値ｔｈ０〜６及び誤差検出テーブル５３０の内容を書き換えることにより、濃度再現特性を柔軟に調整することができる。すなわち、上述のようにＡＩＤＣパターンの測定結果に基づいて濃度再現特性を線型特性に近づけることができるだけでなく、例えば原稿のハイライト部分（明部）の階調再現を他の部分より優先させるといった調整も可能である。ハイライト側の閾値間隔を狭めるのである。
【００６４】
上述の実施形態において、ＡＩＤＣパターン９０を３以上の箇所に作成し、主走査方向の濃度傾斜をより詳しく測定し、結果を階調変換の設定に反映させてもよい。
【００６５】
階調再現回路５０における入力画像データＤ２０の階調数Ｎを２５６、出力画像データＤ５０の階調数Ｍを８として例を挙げたが、Ｎ＞Ｍの関係を満たす範囲内で画像データのビット数ｎ，ｍを任意に選定することができる。
【００６６】
【発明の効果】
請求項１乃至請求項３の発明によれば、階調再現性をより高精度に一定化し、品質の安定したプリント出力を実現することができる。
【図面の簡単な説明】
【図１】本発明に係る複写機の全体構成を示す図である。
【図２】イメージリーダ部の画像処理回路のブロック図である。
【図３】ＭＴＦ補正回路の要部のブロック図である。
【図４】図３の各部の出力波形の一例を示す図である。
【図５】空問周波数の補正関数を示す図である。
【図６】１次微分データとエッジ強度との関係を示すグラフである。
【図７】プリンタ部の信号処理系の概略図である。
【図８】階調再現回路のブロック図である。
【図９】低値化回路のブロック図である。
【図１０】γ補正・ＰＷＭ制御回路のブロック図である。
【図１１】ＰＷＭ制御の一例を示す波形図である。
【図１２】ＡＩＤＣパターンの配置の一例を示す模式図である。
【図１３】階調変換の設定変更の要領を説明するための図である。
【図１４】画像安定化処理ルーチンのフローチャートである。
【図１５】ＡＩＤＣパターン作成サブルーチンのフローチャートである。
【図１６】再現特性の測定サブルーチンのフローチャートである。
【図１７】階調レベル設定サブルーチンのフローチャートである。
【図１８】印字処理ルーチンのフローチャートである。
【符号の説明】
３００ データ処理系（画像処理装置）
Ｄ２０ 画像データ（入力画像データ）
Ｄ５２０ 画像データ（出力画像データ）
Ｄｆ，Ｄｅ 特定入力階調レベル（特定階調レベル）
３０１ ＣＰＵ（制御手段）[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an image processing apparatus for printing out multi-tone image data, and is suitable for image reproduction by an electrophotographic process that performs multi-value control of exposure.
[0002]
[Prior art]
In an image forming apparatus that prints out image data such as a digital copying machine or a page printer, multi-tone halftone reproduction is performed by combining multi-value density control in pixel units and area tone control in pixel matrix units. ing. As a density control method for each pixel, there are pulse width modulation and intensity modulation of an exposure light source in an electrophotographic process. An error diffusion method is generally used as a method for area gradation control.
[0003]
Usually, the number of gradations that can be reproduced in pixel units is about 4 (2 bits) to 8 (3 bits), which is smaller than the number of gradations of the original image data (generally 256: 8 bits). Therefore, in the image forming apparatus, input image data representing an original image is converted into output image data having a smaller number of bits, and image processing is performed in which an error in pixel values associated with the data conversion is distributed to surrounding pixels.
[0004]
Conventionally, the correspondence between each gradation level of the input image data and each gradation level of the output image data has been fixed. However, in order to ensure the sharpness of the edge part of the original image and smooth the gradation reproducibility of the other part, the number of gradations of the output image data is changed between the edge part and the other part. There was something.
[0005]
Also known as an electrophotographic image forming apparatus is an AIDC function (automatic image density adjustment) that creates a toner image of a predetermined pattern and adjusts image forming conditions such as the photoreceptor potential according to the measured density value. It has been.
[0006]
[Problems to be solved by the invention]
In the conventional image forming apparatus, since the correspondence between the gradation levels of the input image data and the output image data is fixed as described above, the gradation reproducibility is the temperature, humidity, developer toner ratio, photoconductor. There is a problem that the quality of the reproduced image is unstable because it depends on the image forming conditions such as the surface characteristics. That is, in the case where the number of gradations of the input image data is converted to a smaller number of gradations by the error diffusion method, the correspondence between the gradation level of the input image data and the gradation level of the output image data is fixed. Therefore, even if the print output density reproduced at the gradation level of the output image data changes due to image formation conditions or environmental fluctuations, error diffusion occurs based on the gradation level of the output image data. Since this is performed, there is a difference between the density indicated by the input image data and the density actually printed. In particular, in the case where error data between the gradation level of the input image data and the output gradation level is added to the input image data to be input next and converted to the gradation level of the output image data, the level of the output image data is converted. When an appropriate output density cannot be obtained at the gradation level, there is a problem that density deviation occurs in the entire image. That is, the density indicated by the input image data and the density actually printed are not constant. In a color image forming apparatus that superimposes toner images of a plurality of colors, a change in gradation reproducibility of each color becomes conspicuous as a change in reproduced color.
[0007]
SUMMARY OF THE INVENTION An object of the present invention is to make gradation reproducibility constant with higher accuracy and to realize a print output with stable quality.
[0008]
[Means for Solving the Problems]
In the present invention, the correspondence relationship between the input and output gradation levels is changed based on the measured value of the density reproduction state at at least one place in the print output means. At that time, the number of gradations in the calculation is set to be larger than that of the output image data so that the number of gradations corresponding to the deviation between the target density and the measured value is sufficiently large and the number of correspondence options is increased.
[0009]
In the apparatus of the first aspect of the invention, the number of gradations N is 2. ⁿ Applying the error diffusion method to the input image data (n is an integer of 3 or more), the gradation number M is 2 ^m An image processing apparatus for converting into output image data (m is an integer of 2 or more smaller than n), wherein an N-level gradation level of the input image data and an M-level gradation level of the output image data The gradation is determined according to print output density measurement data corresponding to a specific gradation level in which the correspondence is variable and an error does not occur in the conversion of the gradation level of the input image data to the output image data. Control means for setting a correspondence relationship between levels, and the control means determines in advance the gradation level of the output image data based on the measurement data corresponding to each of the plurality of specific gradation levels. An estimated gradation level corresponding to the target value of the printed output density is calculated, and the gradation level of the input image data closest to the calculated estimated gradation level is set as a new specific gradation level. It is intended to be set.
[0010]
In the apparatus of the invention of claim 2, the estimated gradation level is calculated by numerical calculation of a bit number larger than the bit number m of the output image data.
According to the method of the invention of claim 3, the number of gradations N is 2. ⁿ Applying the error diffusion method to the input image data (n is an integer of 3 or more), the gradation number M is 2 ^m An image processing method for converting into output image data (m is an integer of 2 or more smaller than n), wherein the input image data has N gradation levels and the output image data has M gradation levels. The print output density measurement data corresponding to each of a plurality of specific gradation levels that do not cause an error in the conversion to the output image data out of the gradation levels of the input image data is variable. Based on the estimated gradation level corresponding to the print output density target value determined in advance for the gradation level of the output image data, and calculating the level of the input image data closest to the calculated estimated gradation level. The tone level is set as a new specific gradation level, and the correspondence relationship of the gradation levels is set according to the new specific gradation level.
[0011]
DETAILED DESCRIPTION OF THE INVENTION
A digital copying machine will be described as an application example of the present invention. First, a configuration related to print output of image data will be described in general, and then a configuration unique to the present invention will be described in detail.
〔overall structure〕
FIG. 1 is a diagram showing the overall configuration of a copying machine 1 according to the present invention.
[0012]
The copying machine 1 is a digital color copying machine and includes an automatic document feeder (ADF) 10, a reduction projection type image reader unit 20, and an electrophotographic printer unit 30. The copier 1 also includes an interface 27 for data communication with an external device, and outputs image data read by the image reader unit 20 to the external device, or conversely, image data from the external device. Print data to the printer unit 30 for printing.
[0013]
The automatic document feeder 10 transports the document set on the document setting tray 11 to the reading position of the image reader unit 20 and discharges it onto the discharge tray 13 after the reading is completed. The document conveying operation is performed in accordance with an instruction from an operation panel (not shown), and the discharge is performed in response to a reading end signal from the image reader unit 20. When a plurality of originals are set, these control signals are continuously generated, and a series of operations of original conveyance, reading, and original discharge are performed efficiently.
[0014]
In the image reader unit 20, a document positioned on the document table glass 28 is irradiated from below by an exposure lamp 21, and reflected light from the document enters a CCD image sensor 24 through a mirror group 22 and an imaging lens 23. . The scanner in which the exposure lamp 21 and the first mirror are assembled is moved by the motor 29 in the sub-scanning direction at a speed V corresponding to the magnification. Thus, the original can be scanned over the entire surface. In accordance with the movement of the scanner, the second and third mirrors move in the same direction at a speed V / 2 and keep the optical path length constant. The position of the scanner is controlled based on the calculated value of the amount of movement (number of motor steps) from when the sensor 26 detects that the home position has been left. The reading resolution of the image reader unit 20 is, for example, 400 dpi.
[0015]
Light incident on the CCD image sensor 24 is photoelectrically converted for each of the R, G, and B color components. Analog signal processing, A / D conversion, and digital image processing are performed by the image processing circuit 25 on the photoelectric conversion signals of the respective colors. Image data corresponding to the document is sent from the image processing circuit 25 to the printer unit 30 or the interface 27.
[0016]
Note that a white plate for shading correction is arranged outside the document reading area of the document table glass 28, and the white plate is read in order to generate shading correction data prior to reading the document.
[0017]
In the printer unit 30, image data sent from the image reader unit 20 is converted into cyan (C), magenta (M), yellow by the data processing system of the print head 31 (corresponding to the image processing apparatus of the present invention). It is converted into image data (print data) for printing (Y) and black (K). The control unit of the print head 31 causes a semiconductor laser, which is an exposure light source, to emit light according to the print data of each color. The laser light is deflected in the main scanning direction by a polygon mirror, and is guided to the photosensitive members of a total of four imaging units 32c, 32m, 32y, and 32k by the mirror.
[0018]
Elements required for the electrophotographic process are arranged around the photoreceptor in the inside of each imaging unit 32c, 32m, 32y, 32k. In the figure, the photosensitive member rotates clockwise, and charging, exposure, development, and transfer processes are continuously performed. The imaging units 32c, 32m, 32y, and 32k are integrated independently for each color, and are configured to be detachable from the printer unit main body.
[0019]
The toner image obtained by developing the latent image on the photosensitive member is obtained by C, M, Y by the transfer chargers 33c, 33m, 33y, 33k arranged at positions facing the photosensitive member with the paper conveying belt 34 interposed therebetween. , K in the order of transfer onto the paper. However, in the case of monochrome copying using only the black imaging unit 32k, the sheet conveying belt 34 is separated from the other three-color imaging units 32c, 32m, and 32y as indicated by a chain line in the drawing. Placed in position. This is to prevent the photoconductor from being worn.
[0020]
A registration correction sensor 36, an AIDC sensor 37, a cleaner 38, and a fixing roller pair 39 are disposed on the downstream side of the imaging unit 32k in the paper transport path. The paper that has undergone the fixing process is discharged onto a paper discharge tray 49. The registration correction sensor 36 detects color misregistration, and the detection signal is used for drawing position correction and distortion correction. The AIDC sensor 37 is a photosensor that detects the density of the AIDC pattern according to the present invention. The AIDC pattern is a toner image for checking the density reproduction state, and is formed in an area without paper on the paper transport belt 34. The cleaner 38 erases the AIDC pattern.
[0021]
Below the paper transport belt 34, paper feed cassettes 41a, 41b, and 41c are mounted. A sheet of a predetermined size is supplied from the selected paper feed cassette to the transport path by the paper feed roller, and is sent to the transport belt 34 by the roller group. Waiting for the sensor 35 to detect the reference mark on the paper transport belt 34, transport by the paper transport belt 34 is started. Then, the paper that has undergone the electrophotographic process including fixing is discharged onto a paper discharge tray 49. In the case of double-sided copying, the paper that has passed through the fixing roller pair 39 is turned upside down by switchback conveyance. Both sides Sent to unit 48, Both sides The paper is fed again from the unit 48 to the paper transport belt 34.
[0022]
The image processing circuit 25 includes an A / D conversion circuit 251, a shading correction circuit 252, a log conversion circuit 253, a UCR / BP processing circuit 254, a masking circuit 255, a density correction circuit 256, an MTF correction circuit 257, and an additional function circuit (not shown). It is composed of A predetermined instruction is given to these circuits from the first CPU 201 that controls the image reader unit 20. Examples of the additional function circuit include a circuit for correcting chromatic aberration of the reading optical system, a circuit for realizing various image editing including enlargement / reduction, and a circuit for preventing counterfeiting of banknotes and securities.
[0023]
The A / D conversion circuit 251 performs offset and gain correction on the analog signal input from the CCD image sensor 24, and the corrected signal is 8 bits (256 gradations) for each of R, G, and B colors. Convert to image data. The shading correction circuit 253 corrects the image data of each color according to uneven light distribution of the exposure lamp 21 and variations in sensitivity between pixels of the CCD image sensor 24. The log correction circuit 253 converts image data representing luminance into image data representing density according to human specific visibility. The UCR / BP processing circuit 254 extracts dark color components to be reproduced with black toner from the image data, and corrects the R, G, and B data values according to the extracted values. The masking circuit 255 generates four-color image data of C, M, Y, and K based on the three-color image data. A density correction circuit 256 performs color correction by multiplying C, M, and Y image data by a predetermined coefficient. Then, the MTF correction circuit 257 performs processing for improving image quality such as smoothing. The output of the MTF correction circuit 257 is sent to the printer unit 30 as print data.
[0024]
Here, filter processing in the MTF correction circuit 257 will be described. The applied filters are a general first-order differential filter and a Laplacian filter.
The relationship between the input D and the output DD of the MTF correction circuit 257 is expressed by equation (1).
[0025]
DD = D × [F (ΔV) × G (ΔD)] (1)
ΔV in the equation is Laplacian data of lightness V obtained by conversion from the RGB color space to the VHC color space, and F (ΔV) is a spatial frequency correction function shown in FIG. Further, ΔD is a primary differential value of input data (density) from the density correction circuit 256, and G (ΔD) is an edge strength output shown in FIG. In this example, when the edge strength G is equal to or greater than a predetermined threshold, the target pixel is regarded as an edge, and the edge signal is activated. This edge signal is sent to the printer unit 30 together with the image data, and is used for halftone reproduction control.
[0026]
FIG. 3 is a block diagram of a main part of the MTF correction circuit 257, and FIG. 4 is a diagram showing an example of an output waveform of each part of FIG.
The image data (D) from the density correction circuit 256 is differentiated by the primary differential filter 2571, and an edge having a value corresponding to the differential value ΔD from the first table 2572 that receives the obtained differential value ΔD. Strength A signal G (ΔD) is output. On the other hand, the brightness V is differentiated by a Laplacian filter 2574, and the obtained differential value ΔV is input to the second table 2575. The table 2575 outputs a correction function F (ΔV) corresponding to the differential value ΔV. Image data (D), edge Strength The signal G (ΔD) and the correction function F (ΔV) are multiplied by a multiplier 2578. The differential value ΔD obtained by the primary differential filter 2571 is also input to the comparator 2579. The comparator 2579 outputs an edge signal when the differential value ΔD is larger than a predetermined threshold value.
[0027]
When an annular original image as shown in FIG. 4 is read along a line crossing the original image, four edges are present. The transition of the edge intensity along the line is less than that of the original image.
[0028]
FIG. 5 is a diagram showing an air frequency correction function F (ΔV).
The relationship between the differential value ΔV of lightness V by the Laplacian filter 2574 and the correction function F (ΔV) is expressed by the following equation.
[0029]
F (ΔV) = a · ΔV + b (ΔV> 0) (2)
F (ΔV) = a ′ · ΔV + b ′ (ΔV <0) (2 ′)
That is, the correction function F (that is, the image data value DD) is made positive or negative depending on whether the differential value ΔV is positive (image portion) or negative (background portion). Increases the contrast.
[0030]
FIG. 6 is a graph showing the relationship between primary differential data (ΔD) and edge strength G (ΔD).
A portion with a negative edge strength G corresponds to a halftone portion, and a positive portion corresponds to an edge. In the figure, the relationship regarding three lines A1, A2, and An is shown. An represents a function Gn (ΔD) of the nth main scanning line in the sub-scanning direction. When the primary differential data ΔD is obtained for the nth pixel in the sub-scanning direction, the edge intensity at that pixel is obtained from the function Gn (ΔD). The functions G1 (ΔD) to Gn (ΔD) are stored in advance as a table 2572 in the MTF correction circuit 257. If this value is negative for each density, a smoothing filter corresponding to that value is selected. An averaging filter or a median filter is used as the smoothing filter. The smoothing process may be performed with a moving average filter.
[0031]
In FIG. 6, when the primary differential value (ΔD) becomes larger than the threshold value and the edge strength G becomes positive, the image data value DD represents the edge component of the image. In this case, the edge signal is activated as described above. As the primary differential value (ΔD) is larger, G (ΔV) is also larger. As shown in FIG. 4, this component is usually curled with respect to the original image.
[Signal processing in the printer section]
The configuration of the printer unit 30 is a “tandem format” in which a color image is formed by simultaneously using four imaging units 32c, 32m, 32y, and 32k. Print data of four colors C, M, Y, and K are simultaneously transferred from the image reader unit 20 to the printer unit 30 during one scanning period for the document. Therefore, the signal processing in the printer unit 30 is basically four color parallel processing.
[0032]
In color copying, four color toner images must be transferred onto a sheet without deviation, but the four imaging units 32c, 32m, 32y, and 32k are arranged at equal intervals along the sheet conveyance direction. . For this reason, in order to align the transfer position on the paper, the process from exposure to transfer is performed at a timing shifted by a predetermined time for each color. Therefore, since the four-color print data input at the same time is used for exposure control at different timings, it is necessary to sequentially delay the print data for each color by a time determined by the arrangement interval of the photoconductors and the paper conveyance speed. In addition, since four laser beams are deflected by one polygon mirror, the main scanning directions are reversed between the upstream two colors (C, M) and downstream two colors (Y, K). become. Further, color misregistration occurs due to a deviation in exposure position in the main scanning direction, main scanning magnification distortion, sub-scanning bow distortion, and skew distortion when the axial direction of the photosensitive member is not parallel to the deflection direction. For these reasons, the printer unit 30 incorporates a delay memory having a predetermined capacity and a correction circuit for preventing color misregistration. In addition, the printer unit 30 includes a memory unit for storing a single-sided image when copying on both sides.
[0033]
FIG. 7 is a schematic diagram of the signal processing system 300 of the printer unit 30.
The C, M, Y, and K image data D20 transferred from the image reader unit 20 is input to the gradation reproduction circuit 50. In accordance with an instruction from the CPU 301 that controls the printer unit 30, the gradation reproduction circuit 50 converts the 8-bit image data D20 of each color into a 3-bit pseudo gradation in a character separation type multi-value error diffusion format as will be described later. Data is converted into data (for example, 256 gradations). However, for the edges, simple ternary processing is performed in place of the multilevel error diffusion processing in order to increase the sharpness. This switching of processing follows the gradation reproduction attribute signal (LIMOS signal) from the image reader unit 20.
[0034]
In the image distortion circuit (not shown), the main scanning magnification distortion of the C, M, and Y components and the sub-scanning are determined based on the detection results of the deviation amounts of C, M, and Y with respect to K in the resist detection test pattern in which four colors are superimposed. Scanning bow skew distortion is corrected by density distribution type interpolation processing. Then, the corrected data is replaced with black solid data as necessary based on the banknote recognition result.
[0035]
The image data of each color whose value has been lowered by the gradation reproduction circuit 50 is appropriately delayed as described above by the delay memories 59a to 59f, and the γ correction / PWM control circuits 61, 62, 63, provided for each color. 64. In terms of delay, since the image data is reduced to 3 bits, the required memory capacity is smaller than in the case of 8 bits. The γ correction / PWM control circuits 61 to 64 correct gradation distortion due to γ characteristics due to the electrophotographic process, and generate PWM pulses having a duty ratio corresponding to the corrected data value. On / off control of the semiconductor lasers 71 to 74 which are exposure light sources of the photosensitive members 81 to 84 is performed by the generated PWM pulses.
[0036]
FIG. 8 is a block diagram of the gradation reproduction circuit 50. Since the circuit configurations corresponding to the four colors C, M, Y, and K are the same, the circuit configuration for one color is shown here.
[0037]
Image data D20 is input to the gradation reproduction circuit 50 in synchronization with the pixel clock. The latched image data D20 is sent in common to the simple ternarization circuit 51 and the multilevel error diffusion processing circuit 52 according to the present invention, and undergoes predetermined processing in parallel in each circuit. Then, either the output of the simple ternarization circuit 51 or the output of the multi-value error diffusion processing circuit 52 is selected by the selector 55, and is output to the subsequent stage together with the 1-bit gradation reproduction attribute signal as the image data D50. . The selection control signal of the selector 55 is an edge determination signal (corresponding to the gradation reproduction attribute signal). That is, when the pixel of interest corresponds to an edge, a simple ternary value that takes one of the minimum value “0”, median value “128”, or maximum value “255” of the input dynamic range “0 to 255”. The output of the conversion circuit 51 is selected. In other cases, the output of the multi-value error diffusion processing circuit 52 described below is selected. The gradation reproduction circuit 50 is provided with selectors 53 and 54 in accordance with the effective image area signal. When scanning outside the effective image area having a margin around the original, the image data D20 and the edge determination are performed. Each signal is replaced with a constant value data indicating an invalid area.
[0038]
The multi-level error diffusion processing circuit 52 is a circuit that reduces the number of gradations by applying a known error diffusion method as a pseudo gradation reproduction method, and includes a low-value reduction circuit 520, an error detection table (look-up table). 530, a selector 550, an error diffusion matrix 560, a divider 570, and an adder 580. The outline of the function of the multilevel error diffusion processing circuit 52 is as follows.
[0039]
The adder 580 adds the error distributed from the peripheral pixels to the image data D20 of the target pixel. The lowering circuit 520 compares the output of the adder 580 with the threshold set by the CPU 301 and performs gradation conversion from 256 levels to 8 levels. The value of the error accompanying the gradation conversion can be calculated in advance. The error detection table 530 is a data set indicating errors corresponding to 256 input gradation levels. Here, the actual error takes a positive or negative value, but in order to facilitate the error diffusion process, the offset value corresponding to the maximum negative error is added or subtracted when creating data in the error detection table 530. First, the offset value (= 32) is subtracted from the input value (Din), and then an error in the gradation range corresponding to the threshold used in the lowering circuit 520 is obtained. Finally, an offset value is added. This series of calculations is performed by the CPU 301, and the calculation results are downloaded to the table memory as an error detection table 530.
[0040]
The output of the adder 580 is input to the error detection table 530 to obtain an error for the next pixel. An error corresponding to the input value is output from the error detection table 530. The selector 550 is provided to fix the error to a constant value (32 in the example) outside the effective image area.
[0041]
The error diffusion matrix 560 is a functional element that distributes gradation conversion errors by weighting peripheral pixels. Again, the weight is an integer for simplicity, and the divider 570 performs an operation to divide the output of the error diffusion matrix 560 by the sum of the weights. As described above, since the offset is subtracted and added so that the error is always positive at the stage of setting the contents of the error detection table 530, the calculation of a negative number in the error diffusion matrix 560 becomes unnecessary. As a result, the circuit configuration is simple and small-scale, and the operation becomes faster. In error diffusion, it is necessary to calculate the amount of error to be added before the image data D20 of the target pixel is input. Therefore, the error diffusion processing time defines the transfer rate of the image data. Therefore, the copy speed can be improved by increasing the speed of the error diffusion matrix 560.
[0042]
FIG. 9 is a block diagram of the lowering circuit 520.
The lowering circuit 520 includes a threshold setting register 521, seven comparators 522a to 522g, and seven selectors 523a to 523g. The threshold value setting register 521 is loaded with seven threshold values th0 to th6 by the CPU 301. Each of the comparators 522a to 522g is commonly input with the image data D580 to which the error is distributed, and is individually input with a corresponding one of the threshold values th0 to th6 from the threshold setting register 521. The selectors 523a to 523g select and output one of the two options (output gradation levels) according to the result of the size discrimination between the image data D580 and the thresholds th0 to th6 by the comparators 522a to 522g.
[0043]
The default values of the threshold values th0 to th6 are 17, 53, 90, 127, 164, 201, and 238 in order, and are set so that the dynamic range of the image data D580 is divided into approximately seven equal parts. Table 1 shows the correspondence between the input gradation level and the output gradation level in the default state. And a total of 8 input gradation ranges partitioned by thresholds th0-6 In The input gradation level (32, 64, 96, 128, 160, 192, 224) is a specific input gradation level at which the error becomes zero upon gradation conversion.
[0044]
[Table 1]

[0045]
FIG. 10 is a block diagram of the γ correction / PWM control circuit 61, and FIG. 11 is a waveform diagram showing an example of PWM control. Since the γ correction / PWM control circuits 61 to 64 corresponding to the colors C, M, Y, and K have the same configuration, the γ correction / PWM control circuit 61 corresponding to C will be described here as a representative example.
[0046]
The image data D50 from the gradation reproduction circuit 50 is corrected by the γ correction table 601 and converted into 8-bit data. Then, it passes through the D / A converter 602 and becomes analog density data, which is input to the two comparators 604 and 605 in common. One comparator 604 receives a two-pixel triangular signal as a reference signal Sref1, and the other comparator 605 receives a one-pixel triangular signal as a reference signal Sref2. Each of the comparators 604, 605 outputs a magnitude determination signal between the density data and the reference signals Sref1, 2 to the selector 606 as PWM data (pulse width modulation data). That is, the exposure pattern of each pixel is determined according to the density data value. The selector 606 selects one of the outputs of the comparator 604 and the comparator 605 according to the gradation reproduction attribute signal and sends it to the second selector. When the tone reproduction attribute signal is “H” indicating a continuous tone portion, PWM control of a 2-dot cycle is performed to make the tone reproducibility smoother, and when the tone reproduction attribute signal is “L” indicating an edge, it is suitable for sharpening. PWM control with a 1-dot period is performed. The light modulation frequency is automatically switched to improve the image quality. Note that in the 2 dot period PWM control, the screen angle in the 45-degree direction is provided so as to improve the granularity of the image, so the phase of the reference signal Sref1 is shifted by half a period for each line.
[0047]
On the other hand, the image data D50 from the gradation reproduction circuit 50 is also input to the intensity modulation circuit 610, and is converted into intensity modulation data whose signal level changes according to the density. The selector 607 selects one of the pulse width modulation data from the selector 606 and the intensity modulation data from the intensity modulation circuit 610 in accordance with an instruction from the CPU 301, and outputs it as the LD drive signal D61. The light emission control of the semiconductor laser 71 is performed using the LD drive signal D61.
[Data processing to which the present invention is applied]
In the copying machine 1 of this embodiment, in order to keep the density reproducibility of each color of C, M, Y, and K constant, an AIDC pattern is created and the result of the density measurement is input to the gradation reproduction circuit 50. This is reflected in the correspondence between the tone level (256 levels) and the output gradation level (8 levels). Specifically, the threshold values th0 to th6 to be applied to the lowering circuit 520 are changed, and the error detection table 530 is rewritten to change the specific input gradation level described above. The electrophotographic process depends on environmental conditions such as temperature and humidity. In addition, the internal environment of the copying machine 1 may become uneven. Therefore, AIDC patterns are created at a plurality of locations in the main scanning direction on the photoconductors 71 to 74, and the reproducibility of the entire image is made uniform.
[0048]
FIG. 12 is a schematic diagram showing an example of the arrangement of the AIDC pattern 90.
The AIDC pattern 90 is created at two locations, one end and the other end of the paper transport belt 34 in the main scanning direction for each of the four colors C, M, Y, and K. In this example, six AIDC patterns 90 having different densities for each color are arranged side by side in the sub-scanning direction, and the total number of AIDC patterns 90 formed during one operation setting is 48 (= 6 × 4 × 2). . The size of one AIDC pattern 90 is about 1 cm square. Four AIDC sensors 37 are arranged at each location, and one AIDC sensor 37 sequentially detects the density of six AIDC patterns 90 of one color.
[0049]
FIG. 13 is a diagram for explaining the procedure for changing the setting of gradation conversion.
As described above, in the initial state immediately after the power is turned on, among the input gradation levels of 256 levels from 0 to 255, the specific input gradation levels that do not cause an error when converted to 8 levels are 0, 32, 64, 96, and 128. , 160, 192, 224. The AIDC pattern 90 is created for six specific input gradation levels excluding 0 and 224 of these eight specific input gradation levels, and the density is measured. In the case of the specific input gradation level, the error assigned to the peripheral pixels in the error diffusion process is 0, so that a predetermined number of pixels (about 160 at 400 dpi) constituting one AIDC pattern 90. ² ) Exposure patterns are all the same. The specific input gradation level can be regarded as the output gradation level (0 to 7) when the pseudo gradation method is not applied.
[0050]
The reproduction density (target value) for each specific input gradation level is determined at the design stage of the copying machine 1, and the designed gradation reproduction characteristic is a linear characteristic. For example, the target value of the reproduction density for the specific input gradation level 32 is 0.2. However, the actual reproduction density (measurement value) may deviate from the target value due to changes in the operating environment and aging of parts. In FIG. 13, the measured value on one end side in the main scanning direction (for example, the front side of the copying machine 1) is indicated by a black circle “●”, and the measured value on the other end side (rear side) is indicated by a black triangle “▲”. Yes. The front side is generally “dark” and the back side is generally “light” (however, this is a hypothetical example for explanation, and this characteristic does not always appear in practice) Absent).
[0051]
The copying machine 1 linearly approximates the actual tone reproduction characteristics between specific input tone levels based on the measured density value, for example, and obtains an estimated tone level corresponding to the target value. Then, the obtained estimated gradation level is set to a new specific input gradation level. The gradation level at the position indicated by the circle “◯” (front side) or the triangle “Δ” (back side) on the horizontal axis in the figure is the estimated gradation level. For example, if the input gradation level on the front side corresponding to the target value 0.2 is 24, the specific input gradation level for the output gradation level 1 is changed from 32 to 24. The specific input gradation level may be set independently for the front side and the back side, or the average of the measured values on the front side and the back side is obtained, and the characteristics are approximated based on the average value to obtain the target value. The corresponding specific input tone level may be calculated.
[0052]
The AIDC pattern 90 is newly created for the new specific input gradation level, and the density is measured. The correction of the specific input gradation level based on the density measurement is repeated until the difference between the target value and the measured value falls within an allowable range or reaches a certain number of times. Thereby, the reproduction characteristic can be brought close to ideal.
[0053]
An important point in changing the gradation conversion settings is to approximate the actual reproduction characteristics and to obtain numerical values (gradation level and reproduction density) related to a series of operations for obtaining the input gradation level corresponding to the target value from the approximate characteristics. ) To be larger than the number of bits of output gradation data (3 in this example). If the gradation level measured by the AIDC pattern 90 is the same 3-bit numerical value (0 to 7) as the output gradation data, there is no room for changing the specific input gradation level. The greater the number of bits in the numerical operation, the higher the accuracy of approximation of characteristics and the calculation of the gradation level corresponding to the target value. For example, when the gradation level is expressed in increments of 0.1, when the calculated value of the estimated gradation level is 23.5 to 24.4, 24 is set as the specific input gradation level, and 24.5 to 25. In the case of 4, it is possible to set 25 as the specific input gradation level.
[0054]
Hereinafter, based on a flowchart, operation | movement of CPU301 of the printer part 30 which concerns on this invention is demonstrated.
FIG. 14 is a flowchart of an image stabilization processing routine.
[0055]
This routine is executed at the time of power-on, at the end of the first copy after power-on, when a certain number of copies are made, when a certain time has passed, and when a predetermined condition such as a set time is met. .
[0056]
The CPU 301 executes AIDC pattern creation (# 1), reproduction characteristic measurement (# 2), gradation level setting (# 3), and other stabilization (# 4) processes. In other stabilization processes, toner replenishment and charging charger output control are performed.
[0057]
FIG. 15 is a flowchart of the AIDC pattern creation subroutine.
Current specific input gradation level Df for one end side in the main scanning direction from the gradation level memory 302 (see FIG. 8) ₀ ~ Df ₇ (# 11), and then the current specific input gradation level De for the other end side ₀ ~ De ₇ Is read out (# 12). Then, as described with reference to FIG. 12, AIDC patterns 90 are created at two locations in the main scanning direction (# 13).
[0058]
FIG. 16 is a flowchart of a reproduction characteristic measurement subroutine.
The output of each AIDC sensor 37 is periodically taken and the density of the AIDC pattern 90 of each color created at two locations is measured (# 21). Based on the measurement results, a straight line Yf that approximates the reproduction characteristics between the gradation levels for each of the one end side and the other end side in the main scanning direction. _j , Ye _j (# 22).
[0059]
FIG. 17 is a flowchart of the gradation level setting subroutine.
First, the reproduction density target value Y ₀ ~ Y ₇ Is set (# 31). The straight line Yf that was obtained earlier _j , Ye _j And target value Y ₀ ~ Y ₇ The specific input gradation level Df that matches the actual reproduction characteristics ₀ '~ Df ₇ ', De ₀ '~ De ₇ 'Is calculated (# 32). Then, the obtained specific input gradation level Df ₀ '~ Df ₇ ', De ₀ '~ De ₇ 'Is the new specific input gradation level Df ₀ ~ Df ₇ , De ₀ ~ De ₇ Is stored in the gradation level memory 302 (# 33).
[0060]
FIG. 18 is a flowchart of the print processing routine.
This routine is started in response to a print request for a document image or an external input image from the image reader unit 20.
[0061]
When printing the first half of the image data in the main scanning direction, one specific input gradation level Df is read from the gradation level memory 302. ₀ ~ Df ₇ Are read out (# 51, # 52). When printing the second half of the image data in the main scanning direction, the other specific input gradation level De is read from the gradation level memory 302. ₀ ~ De ₇ Are read out (# 51, # 53).
[0062]
The thresholds th0-6 are set in accordance with the read specific input gradation level, the contents of the error detection table 530 are rewritten (# 54), and the value of the threshold setting register 521 is replaced. Then, multilevel error diffusion processing (# 56) for controlling the gradation reproduction circuit 50 and printing control (# 57) for controlling the electrophotographic process are executed.
[0063]
According to the above-described embodiment, the density reproduction characteristics can be flexibly adjusted by rewriting the contents of the lower thresholds th0 to th6 and the error detection table 530. That is, as described above, the density reproduction characteristic can be brought close to the linear characteristic based on the measurement result of the AIDC pattern, and for example, the gradation reproduction of the highlight portion (bright portion) of the document is given priority over other portions. Adjustment is also possible. The threshold interval on the highlight side is narrowed.
[0064]
In the above-described embodiment, the AIDC pattern 90 may be created at three or more locations, the density gradient in the main scanning direction may be measured in more detail, and the result reflected in the gradation conversion setting.
[0065]
In the gradation reproduction circuit 50, the number of gradations N of the input image data D20 is 256, and the number of gradations M of the output image data D50 is 8. However, the bit of the image data is within a range satisfying the relationship of N> M. The numbers n and m can be arbitrarily selected.
[0066]
【The invention's effect】
Claim 1 Thru Claim 3 According to the invention, it is possible to make the gradation reproducibility constant with higher accuracy and to realize a print output with stable quality.
[Brief description of the drawings]
FIG. 1 is a diagram showing an overall configuration of a copying machine according to the present invention.
FIG. 2 is a block diagram of an image processing circuit of an image reader unit.
FIG. 3 is a block diagram of a main part of an MTF correction circuit.
4 is a diagram illustrating an example of an output waveform of each unit in FIG. 3;
FIG. 5 is a diagram illustrating an air frequency correction function.
FIG. 6 is a graph showing the relationship between primary differential data and edge strength.
FIG. 7 is a schematic diagram of a signal processing system of a printer unit.
FIG. 8 is a block diagram of a gradation reproduction circuit.
FIG. 9 is a block diagram of a lowering circuit.
FIG. 10 is a block diagram of a γ correction / PWM control circuit.
FIG. 11 is a waveform diagram showing an example of PWM control.
FIG. 12 is a schematic diagram showing an example of the arrangement of AIDC patterns.
FIG. 13 is a diagram for explaining a procedure for changing settings of gradation conversion;
FIG. 14 is a flowchart of an image stabilization processing routine.
FIG. 15 is a flowchart of an AIDC pattern creation subroutine.
FIG. 16 is a flowchart of a reproduction characteristic measurement subroutine.
FIG. 17 is a flowchart of a gradation level setting subroutine.
FIG. 18 is a flowchart of a print processing routine.
[Explanation of symbols]
300 Data processing system (image processing device)
D20 Image data (input image data)
D520 Image data (output image data)
Df, De Specific input gradation level (specific gradation level)
301 CPU (control means)

Claims (3)

Applying the error diffusion method to input image data with a gradation number N of 2 ⁿ (n is an integer of 3 or more), output image data with a gradation number M of 2 ^m (m is an integer of 2 or more smaller than n) An image processing device for converting to
The correspondence relationship between the N gradation levels of the input image data and the M gradation levels of the output image data is variable,
The correspondence relationship of the gradation levels is set according to the measurement data of the print output density corresponding to the specific gradation level that does not cause an error in the conversion of the gradation level of the input image data to the output image data. Having control means,
The control means is based on the measurement data corresponding to each of the plurality of specific gradation levels, and an estimated floor corresponding to a print output density target value predetermined for the gradation level of the output image data. A tone level is calculated, and the tone level of the input image data closest to the calculated estimated tone level is set as a new specific tone level. The image processing apparatus according to claim 1, wherein the estimated gradation level is calculated by a numerical operation with a bit number larger than the bit number m of the output image data. Applying the error diffusion method to input image data with a gradation number N of 2 ⁿ (n is an integer of 3 or more), output image data with a gradation number M of 2 ^m (m is an integer of 2 or more smaller than n) An image processing method for converting to
The correspondence relationship between the N gradation levels of the input image data and the M gradation levels of the output image data is variable,
Based on the print output density measurement data corresponding to each of a plurality of specific gradation levels that do not cause an error in the conversion of the input image data to the output image data, the output image data is previously stored. An estimated gradation level corresponding to the target value of the print output density determined for the gradation level is calculated, and the gradation level of the input image data closest to the calculated estimated gradation level is set as a new specific gradation. Set as level,
A method for image processing, wherein a correspondence relationship between the gradation levels is set according to the new specific gradation level.

Images