JP2005012247A - Image processing method and image processing apparatus, image processing program, and computer-readable recording medium recording this image processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing method, an image processing apparatus, an image processing program and a recording medium which can simplifying processing for performing image area separation, shortening the time required for image processing by just that much, reducing circuit scale and program scale for image processing, thereby realizing cost reduction. <P>SOLUTION: The image processing apparatus A, the image processing program Q and the recording medium 300 each substitute gradations of regions of a plurality of multi-value image information inputted from an image reading apparatus 2a into a predetermined formula for obtaining a variance and the variance is obtained, determine a region as a first image area (regions of a photograph dot image) if the obtained variance is less than a predetermined value, and determine a region as a second image area (region of a character image) if it is equal to or more than the predetermined value and separate the region. Image processing corresponding to each image area is performed on the basis of a result of determination of the first and second image areas, and the plurality of multi-value image information subjected to the image processing are converted into binary image information and are binarized. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

にそれぞれ代入して分散値を求め、得られた分散値が、濃度の濃い領域と淡い領域との階調性の集まりがばらつく傾向にある所定の値未満であれば当該多値画像情報の領域を前記第１像域（写真網点画像の領域）と判定する一方、濃度の濃い領域と淡い領域との階調性の集まりがかたよる傾向にある前記所定の値以上であれば当該多値画像情報の領域を前記第２像域（文字画像の領域）と判定して該第１及び第２の像域（写真網点画像及び文字画像の領域）を分離し、前記多値化された前記複数の多値画像情報について、前記第１及び第２の像域（写真網点画像及び文字画像の領域）の判定結果に基づいて、それぞれの像域に応じた画像処理を行えば、像域分離を行うための処理が簡素化され、全体の画像処理に要する時間が短くなるとともに、画像処理のためのハードウェアを構成する回路規模やソフトウェアを構成するプログラム規模が小さくなり、コストダウンにつながる、との知見を得た。
【００１８】
本発明はかかる知見に基づくものであり、前記課題を解決するため、次の画像処理方法並びに画像処理装置、画像処理プログラム及び記録媒体を提供する。すなわち、
（１）画像処理方法
原稿画像を複数の画素に分割して複数の画像情報を読み取り、該読み取った前記複数の画像情報をいずれも複数階調の多値画像情報に変換して多値化する画像読取装置から入力された前記複数の多値画像情報の領域のうち、連続的な階調性を有する第１の像域（さらに言えば、写真網点画像の領域）と該第１像域以外の第２の像域（さらに言えば、文字画像の領域）とを判定して該第１及び第２の像域（写真網点画像及び文字画像の領域）を分離する像域分離ステップであって、該複数の多値画像情報の階調を、分散値を求める所定の計算式にそれぞれ代入して分散値を求め、得られた分散値が、所定の値未満であれば当該多値画像情報の領域を前記第１像域（写真網点画像の領域）と判定する一方、前記所定の値以上であれば当該多値画像情報の領域を前記第２像域（文字画像の領域）と判定して該第１及び第２の像域（写真網点画像及び文字画像の領域）を分離する像域分離ステップと、
前記画像読取装置から入力された前記複数の多値画像情報について、前記像域分離ステップによる前記第１及び第２の像域（写真網点画像及び文字画像の領域）の判定結果に基づいて、それぞれの像域に応じた画像処理を行う多値画像処理ステップと、
前記多値画像処理ステップにて画像処理された前記複数の多値画像情報をいずれも二値画像情報に変換して二値化する二値化処理ステップと
を含むことを特徴とする画像処理方法。
【００１９】
（２）画像処理装置
原稿画像を複数の画素に分割して複数の画像情報を読み取り、該読み取った前記複数の画像情報をいずれも複数階調の多値画像情報に変換して多値化する画像読取装置から入力された前記複数の多値画像情報の領域のうち、連続的な階調性を有する第１の像域（さらに言えば、写真網点画像の領域）と該第１像域以外の第２の像域（さらに言えば、文字画像の領域）とを判定して該第１及び第２の像域（写真網点画像及び文字画像の領域）を分離する像域分離手段であって、該複数の多値画像情報の階調を、分散値を求める所定の計算式にそれぞれ代入して分散値を求め、得られた分散値が、所定の値未満であれば当該多値画像情報の領域を前記第１像域（写真網点画像の領域）と判定する一方、前記所定の値以上であれば当該多値画像情報の領域を前記第２像域（文字画像の領域）と判定して該第１及び第２の像域（写真網点画像及び文字画像の領域）を分離する像域分離手段と、
前記画像読取装置から入力された前記複数の多値画像情報について、前記像域分離手段による前記第１及び第２の像域（写真網点画像及び文字画像の領域）の判定結果に基づいて、それぞれの像域に応じた画像処理を行う多値画像処理手段と、
前記多値画像処理手段にて画像処理された前記複数の多値画像情報をいずれも二値画像情報に変換して二値化する二値化処理手段と
を備えることを特徴とする画像処理装置。
【００２０】
（３）画像処理プログラム
前記本発明に係る画像処理方法の各ステップを前記画像読取装置に接続されたコンピュータに実行させるための画像処理プログラム。
【００２１】
（４）記録媒体
前記本発明に係る画像処理プログラムを記録したコンピュータ読み取り可能な記録媒体。
【００２２】
本発明に係る画像処理方法並びに画像処理装置、画像処理プログラム及び記録媒体によると、前記画像読取装置から入力された前記複数の多値画像情報の領域のうち、連続的な階調性を有する第１の像域（写真網点画像の領域）と該第１像域以外の第２の像域（文字画像の領域）とを判定して該第１及び第２の像域（写真網点画像及び文字画像の領域）を分離するにあたり、該複数の多値画像情報の階調を、分散値を求める所定の計算式にそれぞれ代入して分散値を求め、得られた分散値が、所定の値未満であれば当該多値画像情報の領域を前記第１像域（写真網点画像の領域）と判定する一方、前記所定の値以上であれば当該多値画像情報の領域を前記第２像域（文字画像の領域）と判定して該第１及び第２の像域（写真網点画像及び文字画像の領域）を分離し、前記多値化された前記複数の多値画像情報について、前記第１及び第２の像域（写真網点画像及び文字画像の領域）の判定結果に基づいて、それぞれの像域に応じた画像処理を行うので、像域分離を行うための処理を簡素化でき、それだけ全体の画像処理に要する時間を短くできる。
【００２３】
さらに、本発明に係る画像処理方法及び画像処理装置によると、像域分離を行うための処理を簡素化できるので、画像処理のためのハードウェアを構成する回路規模やソフトウェアを構成するプログラム規模を小さくでき、それだけコストダウンを実現できる。
【００２４】
さらに、本発明に係る画像処理方法並びに画像処理プログラム及び記録媒体によると、像域分離を行うための処理を簡素化できるので、画像処理のためのソフトウェアを構成するプログラム規模を小さくでき、それだけコストダウンを実現できる。
【００２５】
本発明に係る画像処理方法並びに画像処理装置、画像処理プログラム及び記録媒体において、前記画像読取装置から入力された前記複数の多値画像情報の２次元座標を（ｘ，ｙ）とし、前記複数の多値画像情報（ｘ，ｙ）の階調をＰｉｘｅｌ（ｘ，ｙ）とすると、前記像域分離ステップ及び前記像域分離手段における前記分散値を求める所定の計算式Ｂｕｎｓａｎ（ｘ，ｙ）としては、
【数４】

を例示できる。この場合、前記像域分離ステップ及び前記像域分離手段では、前記計算式Ｂｕｎｓａｎ（ｘ，ｙ）による前記分散値が、前記所定の値未満であれば当該多値画像情報（ｘ，ｙ）の領域を前記第１像域（写真網点画像の領域）と判定する一方、前記所定の値以上であれば当該多値画像情報（ｘ，ｙ）の領域を前記第２像域（文字画像の領域）と判定して該第１及び第２の像域（写真網点画像及び文字画像の領域）を分離することができる。
【００２６】
前記式中のｍ及びｎの値は、これらが小さすぎると小さい範囲での階調性の分散を調べることになり、前記第１及び第２の像域（写真網点画像及び文字画像の領域）の判定が極端なものになってしまい易い。一方、大きすぎると、大きい範囲での階調性の分散を調べることになり、小さい範囲での前記第１及び第２の像域（写真網点画像及び文字画像の領域）の判定精度が低下し易い。従って、前記式中のｍ及びｎの値としては、それには限定されないが、いずれも３〜５程度を例示できる。また、前記ｍとｎをいずれも同じ値にしてもよい。さらに、前記ｍ及びｎは、所定の値に固定されていてもよいが、オペレータ等が適宜調整できるようにしてもよい。
【００２７】
本発明に係る画像形成方法並びに画像形成装置、画像形成プログラム及び記録媒体において、前記画像読取装置から入力される多値画像情報の階調を２５６階調とし、前記式中のｍ及びｎの値をいずれも５とすると、前記像域分離工程及び前記像域分離手段における前記所定の値としては、１１５０程度を例示できる。
【００２８】
本発明に係る画像処理方法並びに画像処理プログラム及び記録媒体において、前記像域分離ステップは、前記画像読取装置から入力された前記複数の多値画像情報に対応する複数の像域識別用画像情報であって、各対応する多値画像情報が前記第１像域（写真網点画像の領域）にあると判定したときは該第１像域（写真網点画像の領域）と識別できるように、且つ、前記第２像域（文字画像の領域）にあると判定したときは該第２像域（文字画像の領域）と識別できるように識別符号をそれぞれ備えた複数の像域識別用画像情報を予め形成しておく像域識別用画像情報形成ステップを含み、前記多値画像処理ステップでは、前記画像読取装置から入力された前記複数の多値画像情報について、前記像域分離ステップの前記像域識別用画像情報形成ステップにて予め形成された前記複数の像域識別用画像情報による前記第１及び第２の像域（写真網点画像及び文字画像の領域）の識別結果に基づいて、それぞれの像域に応じた画像処理を行ってもよい。
【００２９】
本発明に係る画像処理装置において、前記像域分離手段は、前記画像読取装置から入力された前記複数の多値画像情報に対応する複数の像域識別用画像情報であって、各対応する多値画像情報が前記第１像域（写真網点画像の領域）にあると判定したときは該第１像域（写真網点画像の領域）と識別できるように、且つ、前記第２像域（文字画像の領域）にあると判定したときは該第２像域（文字画像の領域）と識別できるように識別符号をそれぞれ備えた複数の像域識別用画像情報を予め形成しておく像域識別用画像情報形成手段を備え、前記多値画像処理手段は、前記画像読取装置から入力された前記複数の多値画像情報について、前記像域分離手段の前記像域識別用画像情報形成手段にて予め形成された前記複数の像域識別用画像情報による前記第１及び第２の像域（写真網点画像及び文字画像の領域）の識別結果に基づいて、それぞれの像域に応じた画像処理を行ってもよい。
【００３０】
この場合、前記像域識別用画像情報形成ステップ及び前記像域識別用画像情報形成手段の実行に先立って、前記画像読取装置から入力された前記各多値画像情報を補正処理する第１の補正処理ステップ及び第１の補正処理手段をさらに含んでいてもよい。この第１補正処理ステップ及び第１補正処理手段における前記補正処理としては、例えば、前記画像読取装置から入力された前記各多値画像情報について、調整可能にガンマ補正する従来周知の「ガンマ補正処理」、注目画素を中心とする画素領域で近傍画素の階調の平均（例えば３行×３列の画素領域で８近傍の階調の平均）をとる従来周知の「ぼかし処理」などを挙げることができる。また、前記第１補正処理ステップ及び前記第１補正処理手段において、これらの処理を二つ以上行ってもよい。
【００３１】
本発明に係る画像処理方法並びに画像処理プログラム及び記録媒体において、前記画像読取装置から入力された前記複数の多値画像情報に対応する複数の中間調画像用（さらに言えば、写真網点画像用）多値画像情報であって、各対応する多値画像情報に対してそれぞれ中間調画像用（写真網点画像用）処理を施して複数の中間調画像用（写真網点画像用）多値画像情報を予め形成しておく中間調画像用多値画像情報形成ステップと、前記画像読取装置から入力された前記複数の多値画像情報に対応する複数の非中間調画像用（さらに言えば、文字画像用）多値画像情報であって、各対応する多値画像情報に対してそれぞれ非中間調画像用（文字画像用）処理を施して複数の非中間調画像用（文字画像用）多値画像情報を予め形成しておく非中間調画像用多値画像情報形成ステップとをさらに含んでいてもよい。こうすることで、前記画像読取装置から入力された前記各多値画像情報について、原稿画像を読み込んだあと、いずれも中間調画像用（写真網点画像用）多値画像情報とすることも、いずれも非中間調画像用（文字画像用）多値画像情報とすることもできる。また、前記多値画像処理ステップにおいて、前記画像読取装置から入力された前記各多値画像情報について、前記像域分離ステップの前記像域識別用画像情報形成ステップにて予め形成された、対応する像域識別用画像情報により前記第１像域（写真網点画像の領域）であると識別されたときは前記中間調画像用多値画像情報形成ステップにて予め形成された、対応する中間調画像用（写真網点画像用）多値画像情報とする一方、前記第２像域（文字画像の領域）であると識別されたときは前記非中間調画像用多値画像情報形成ステップにて予め形成された、対応する非中間調画像用（文字画像用）多値画像情報としてもよい。
【００３２】
本発明に係る画像処理装置において、前記画像読取装置から入力された前記複数の多値画像情報に対応する複数の中間調画像用（さらに言えば、写真網点画像用）多値画像情報であって、各対応する多値画像情報に対してそれぞれ中間調画像用（写真網点画像用）処理を施して複数の中間調画像用（写真網点画像用）多値画像情報を予め形成しておく中間調画像用多値画像情報形成手段と、前記画像読取装置から入力された前記複数の多値画像情報に対応する複数の非中間調画像用（さらに言えば、文字画像用）多値画像情報であって、各対応する多値画像情報に対してそれぞれ非中間調画像用（文字画像用）処理を施して複数の非中間調画像用（文字画像用）多値画像情報を予め形成しておく非中間調画像用多値画像情報形成手段とをさらに備えていてもよい。こうすることで、前記画像読取装置から入力された前記各多値画像情報について、原稿画像を読み込んだあとに、いずれも中間調画像用（写真網点画像用）多値画像情報とすることも、いずれも非中間調画像用（文字画像用）多値画像情報とすることもできる。また、前記多値画像処理手段において、前記画像読取装置から入力された前記各多値画像情報について、前記像域分離手段の前記像域識別用画像情報形成手段にて予め形成された、対応する像域識別用画像情報により前記第１像域（写真網点画像の領域）であると識別されたときは前記中間調画像用多値画像情報形成手段にて予め形成された、対応する中間調画像用（写真網点画像用）多値画像情報とする一方、前記第２像域（文字画像の領域）であると識別されたときは前記非中間調画像用多値画像情報形成手段にて予め形成された、対応する非中間調画像用（文字画像用）多値画像情報としてもよい。
【００３３】
この場合、前記中間調画像用多値画像情報形成ステップ及び前記非中間調画像用多値画像情報形成ステップ並びに前記中間調画像用多値画像情報形成手段及び前記非中間調画像用多値画像情報形成手段の実行に先立って、前記画像読取装置から入力された前記各多値画像情報を補正処理する第２の補正処理ステップ及び第２の補正処理手段をさらに含んでいてもよい。この第２補正処理ステップ及び第２補正処理手段における前記補正処理としては、例えば、前記画像読取装置から入力された前記各多値画像情報について、所定の上限階調以上及び／又は所定の下限階調以下は原稿画像のベース濃度として処理する「ハイライト／シャドウ（ＨＬ／ＳＤ）処理」、調整可能にガンマ補正する従来周知の「ガンマ補正処理」などを挙げることができる。また、前記第２補正処理ステップ及び第２補正処理手段において、これらの処理を二つ以上行ってもよい。
【００３４】
一般的に、多値化された多値画像情報の領域のうち、例えば、比較的小さい文字画像や濃度が若干淡い中間調の文字画像については、二値画像情報に変換して単純に二値化すると、濃度（例えば、黒色）のきつい文字画像になったり、文字画像周縁部に生じることがあるノイズにより輪郭部分が強調された文字画像になったりする。
【００３５】
そこで、前記非中間調画像用多値画像情報形成ステップ及び前記非中間調画像用多値画像情報形成手段における前記非中間調画像用（文字画像用）処理では、前記画像読取装置から入力された前記複数の多値画像情報について、例えば、図１３に示すように、前記画像読取装置から入力された前記複数の多値画像情報のうち所定の階調範囲内（例えば、前記第１像域（写真網点画像の領域）Ｘ’を除く濃度の淡い側の網点画像領域近傍Ｚ１から濃度の濃い側の網点画像領域近傍Ｚ２の範囲内）にある各多値画像情報（図１３中斜線部）について、従来周知の誤差拡散法による誤差拡散処理を行うことができる。こうすることで、前記画像読取装置から入力された前記複数の多値画像情報の領域のうちの前記第２像域（文字画像の領域）であって、例えば、比較的小さい文字画像の領域や濃度が若干淡い中間調の文字画像の領域であっても、二値画像情報に変換して二値化したときに、中間調再現性よく、ノイズによる輪郭部分の強調が抑制された文字画像を得ることができる。
【００３６】
前記中間調画像用多値画像情報形成ステップ及び前記中間調画像用多値画像情報形成手段における前記中間調画像用（写真網点画像用）処理としては、モアレを除去するための、注目画素を中心とする画素領域で近傍画素の階調の平均（例えば３行×３列の画素領域で８近傍の階調の平均）をとる従来周知の「ぼかし処理」及び／又は注目画素を中心とする画素領域での階調の平均（例えば５行×５列、７行×７列、９行×９列等の画素領域での階調の平均）をとる「平均化処理」を含む「モアレ除去処理」、中間調を表現するための従来周知の「ディザ法」による「ディザ処理」などを例示できる。また、前記中間調画像用多値画像情報形成ステップ及び前記中間調画像用多値画像情報形成手段において、これらの処理を二つ以上行ってもよい。この場合、前記の「モアレ除去処理」は、前記の「ディザ処理」の前に行うことができる。
【００３７】
前記本発明に係る画像処理方法並びに画像処理装置、画像処理プログラム及び記録媒体において、カラー原稿画像について画像処理を行ってもよい。この場合、カラー原稿画像を、例えば、赤色（Ｒ）、緑色（Ｇ）、青色（Ｂ）に色分解するとともに、各色の原稿画像を複数の画素に分割して複数の画像情報を読み取り、該読み取った各色に対応する複数の画像情報について、いずれも複数階調の多値画像情報に変換して多値化するカラー画像読取装置から入力された前記各色に対応する複数の多値画像情報を前記したようなデジタル処理を行ったあと、二値画像情報に変換して二値化し、該二値化した二値画像情報を出力することができる。
【００３８】
【発明の実施の形態】
以下、本発明に係る実施の形態について図面を参照しながら説明する。
図１は本発明に係る画像処理方法を実施する画像処理装置の一例Ａを備えたスキャナ１の概略斜視図であり、図２は図１に示すスキャナ１における画像読取装置２ａの概略側面図である。また、図３は図１に示すスキャナ１がデジタル複写機１００に搭載されたときの画像処理装置Ａを中心とする概略ブロック図であり、図４は図１に示すスキャナ１の画像読取装置２ａがコンピュータ２００に接続されたときのコンピュータ２００を中心とする概略ブロック図である。
【００３９】
図１に示すスキャナ１は、図３に示すように、画像読取装置２ａ及び画像処理装置Ａを備えており、デジタル複写機１００に搭載されるか（図３参照）又はコンピュータ２００に接続される（図４参照）。画像読取装置２ａは、原稿画像を複数の画素に分割して複数の画像情報を読み取り、該読み取った複数の画像情報をいずれも複数階調の多値画像情報に変換して多値化するものである。スキャナ１がデジタル複写機１００に搭載される場合には、図３に示すように、前記画像読取装置２ａにて多値化した多値画像情報を、前記画像処理装置Ａにてデジタル処理したあと二値画像情報に変換して二値化し、該二値化した二値画像情報をデジタル複写機１００に出力する。また、スキャナ１がコンピュータ２００に接続される場合には、図４に示すように、前記画像読取装置２ａにて多値化した多値画像情報を、コンピュータ２００にてデジタル処理したあと二値画像情報に変換して二値化し、該二値化した二値画像情報を該コンピュータ２００に接続されたプリンタ２５２等の出力装置２５０に出力する。
【００４０】
また、カラー原稿画像について画像処理を行うときは、画像読取装置として、カラー原稿画像を、例えば、赤色（Ｒ）、緑色（Ｇ）、青色（Ｂ）に色分解するとともに、各色の原稿画像を複数の画素に分割して複数の画像情報を読み取り、該読み取った各色に対応する複数の画像情報について、いずれも複数階調の多値画像情報に変換して多値化するカラー画像読取装置を用いる。この場合、カラー原稿画像を色分解すること以外は白黒原稿画像を処理する場合と実質上同様であり、以下の説明では、カラー原稿画像を処理する場合の説明は省略し、白黒原稿画像を処理する場合について説明する。
【００４１】
前記のスキャナ１は、ここでは、デジタル複写機１００に搭載されるときは、図３に示すように、画像処理装置Ａの各手段Ｍ及びＡ１〜Ａ７にて画像処理を実行することができるとともに、コンピュータ２００に接続されるときは、図４に示すように、本発明に係る画像処理プログラムの一例Ｑを含むスキャナドライバＱ’を記録したコンピュータ読み取り可能な記録媒体３００にてコンピュータ２００に該スキャナドライバＱ’がインストールされたコンピュータ２００にて画像処理を実行することができるように構成されている。
【００４２】
スキャナ１は、図１に示すように、スキャナ本体２と自動原稿移送装置３とからなっている。スキャナ本体２は、前記の画像処理装置Ａ、画像読取装置２ａ及び操作部Ｓを備えている。このスキャナ１では、図２に示すように、自動原稿移送装置３にて送られてくる原稿Ｐ画像を画像読取装置２ａにて光学的に走査し、原稿Ｐ画像から到来する画像光を読み取る。原稿Ｐ画像の走査は、原稿Ｐが主走査方向（移送方法）及び該方向を横切る副走査方向に走査されることでなされ、この走査により到来する画像光が読み取られる。
【００４３】
前記自動原稿移送装置３は、図１に示すように、一側がスキャナ本体２に回動自在に支持されて前記画像読取装置２ａの図示を省略したガラス板に接離自在に当接するカバー５と、カバー５に連設された原稿Ｐの移送機構部６、及び複数枚の原稿Ｐ がセットされる原稿トレイ７とからなる。そして、カバー５の両端部には、図２に示す一対の搬送ローラ８がそれぞれ設けられ、両搬送ローラ８に搬送ベルト９が巻きかけられている。
【００４４】
前記移送機構部６には、図２に示すように、前記原稿トレイ７に連設されたガイド板１０が設けられるとともに、原稿Ｐの移送路Ｒが形成されている。また移送機構部６は、原稿トレイ７の下方に配置された給紙ローラ１１と、移送路Ｒの途中に配置された、給紙、排紙等を行うための移送ローラ１２とを備えている。ガイド板１０は、断面略Ｕ字形状に湾曲しており、この湾曲したガイド板１０の先端部が移送ローラ１２の周面に対向して配置されている。さらに、前記一対の搬送ローラ８のうち一方が移送路Ｒに臨設されている。これにより、原稿Ｐを原稿トレイ７から画像読取ユニット４に移送することができる。
【００４５】
画像読取装置２ａは、図２に示すように、搬送ローラ８の下方に配置されている画像読取ユニット４及びＡ−Ｄ変換部４ａ（図２では図示を省略、図３及び図４参照）を備えている。
【００４６】
画像読取ユニット４は、ここでは、図示を省略した発光素子、撮像素子及び結像レンズからなり、前記発光素子から自動原稿移送装置３にて送られてくる原稿画像に光を照射し、前記結像レンズにより、原稿Ｐ画像から反射してくる画像光を前記撮像素子に結像させて読み取ることができるものである。これにより、原稿Ｐ画像を複数の画素に分割して複数のアナログ画像情報として読み取ることができる。画像読取ユニット４はまた、図３及び図４に示すように、Ａ−Ｄ変換部４ａに接続されており、前記読み取った複数のアナログ画像情報をＡ−Ｄ変換部４ａに送ることができる。
【００４７】
Ａ−Ｄ変換部４ａは、画像読取ユニット４から送られてきた複数のアナログ画像情報を複数ビット（ここでは８ビット）で量子化し、デジタル多値画像情報に変換する。これにより、画像読取ユニット４にて読み取られた複数の画像情報をいずれも複数階調（ここでは０階調〜２５５階調の合計２５６階調）の多値画像情報に変換して多値化することができる。
【００４８】
Ａ−Ｄ変換部４ａはまた、図３及び図４に示すように、画像処理装置Ａと接続可能の接続部４１と、コンピュータと接続可能の接続部４２を備えている。これにより、接続部４１にて画像処理装置Ａに接続されるときは、読み取った画像情報を画像処理装置Ａに送ることができ（図３参照）、また、接続部４２にてコンピュータ２００に接続されるときは、読み取った画像情報をコンピュータ２００に送ることができる（図４参照）。
【００４９】
画像処理装置Ａは、画像読取装置２ａにて読み取った画像情報について、以下の画像処理を行うことができる。
【００５０】
この画像処理装置Ａは、図３に示すように、記憶手段Ｍ、第１補正処理手段Ａ１、第２補正処理手段Ａ２、像域識別用画像情報形成手段Ａ３’を備えた像域分離手段Ａ３、中間調画像用多値画像情報形成手段Ａ４、非中間調画像用多値画像情報形成手段Ａ５、多値画像処理手段Ａ６及び二値化処理手段Ａ７を備えている。
【００５１】
記憶手段Ｍは、画像読取装置２ａから入力される多値画像情報や、多値画像情報の演算処理途中のデータ、ＬＵＴ（ルックアップテーブル）に関する所定のデータ等を記憶する。そのため、ＲＡＭ及びＲＯＭ等によって構成される。
【００５２】
第１補正処理手段Ａ１は、像域識別用画像情報形成手段Ａ３’の実行に先立って、画像読取装置２ａから入力された各多値画像情報を補正処理するものであり、ここでは、装置２ａから入力された各多値画像情報について、調整可能にガンマ補正する従来周知の「ガンマ補正処理」と、注目画素を中心とする画素領域で近傍画素の階調の平均（ここでは３行×３列の画素領域で８近傍の階調の平均）をとる従来周知の「ぼかし処理」とを含むものである。
【００５３】
さらに説明すると、前記ガンマ補正処理では、ここでは２５６階調の多値画像情報の入力階調（入力輝度）に対して互いに異なる複数種類の出力階調（出力輝度）にそれぞれ変換できる複数種類の割当テーブルを予め記憶手段Ｍに記憶させておき、各多値画像情報の階調（輝度）について、この複数種類の割当テーブルのうち、後述するように、オペレータ等により選択される割当テーブルを参照することで階調（輝度）変換される。例えば、テーブルの６４番目を１２０と予め設定しておくと、入力階調が６４の画素の出力階調は前記のテーブル参照によって１２０に置き換えられる（図５（Ａ）参照）。
【００５４】
また、前記ぼかし処理では、多値画像情報の２次元座標を（ｘ，ｙ）とすると、ここでは注目画素（ｘ，ｙ）を中心とする３行×３列の画素領域で８近傍の階調の平均［｛ｆ（ｘ−１，ｙ−１）＋ｆ（ｘ，ｙ−１）＋ｆ（ｘ＋１，ｙ−１）＋ｆ（ｘ−１，ｙ）＋ｆ（ｘ＋１，ｙ）＋ｆ（ｘ−１，ｙ＋１）＋ｆ（ｘ，ｙ＋１）＋ｆ（ｘ＋１，ｙ＋１）｝／８］をとり（図５（Ｂ）の斜線部参照）、これを多値画像情報全体に対して行う。
【００５５】
第２補正処理手段Ａ２は、中間調画像用多値画像情報形成手段Ａ４及び非中間調画像用多値画像情報形成手段Ａ５の実行に先立って、画像読取装置２ａから入力された各多値画像情報を補正処理するものであり、ここでは、装置２ａから入力された各多値画像情報について、所定の上限階調以上及び所定の下限階調以下は原稿画像のベース濃度として処理する「ハイライト／シャドウ（ＨＬ／ＳＤ）処理」と、調整可能にガンマ補正する従来周知の「ガンマ補正処理」とを含むものである。
【００５６】
さらに説明すると、前記ＨＬ／ＳＤ処理では、ここでは、複数の多値画像情報の階調をＩｎとすると、
Ｏｕｔ＝（Ｉｎ−ＳＤ）×２５５／（ＨＬ−ＳＤ）
ここでは、デフォルト値：ＨＬは２４０、ＳＤは１０
の計算式にそれぞれ代入し、得られたＯｕｔ値をそれぞれ複数の多値画像情報の階調にする。これにより、所定の上限階調（ここでは２４０）以上及び所定の下限階調（ここでは１０）以下は原稿画像のベース濃度として処理される。また、前記ガンマ補正処理は、前記第１補正処理手段Ａ１のものと同様であり、ここでは説明を省略する。
【００５７】
像域分離手段Ａ３は、画像読取装置２ａから入力された複数の多値画像情報の領域のうち、連続的な階調性を有する第１の像域（写真網点画像の領域）と該第１像域以外の第２の像域（文字画像の領域）とを判定して該第１及び第２の像域（写真網点画像及び文字画像の領域）を分離するものであり、複数の多値画像情報の階調を、分散値を求める所定の計算式にそれぞれ代入して分散値を求め、得られた分散値が、所定の値未満であれば当該多値画像情報の領域を第１像域（写真網点画像の領域）と判定する一方、前記所定の値以上であれば当該多値画像情報の領域を第２像域（文字画像の領域）と判定して該第１及び第２の像域（写真網点画像及び文字画像の領域）を分離するものである。
【００５８】
さらに説明すると、画像読取装置２ａから入力された複数の多値画像情報の２次元座標を（ｘ，ｙ）とし、複数の多値画像情報（ｘ，ｙ）の階調をＰｉｘｅｌ（ｘ，ｙ）とすると、像域分離手段Ａ３における前記分散値を求める所定の計算式Ｂｕｎｓａｎ（ｘ，ｙ）は、
【数５】

である。
【００５９】
そして、像域分離手段Ａ３は、前記計算式Ｂｕｎｓａｎ（ｘ，ｙ）による前記分散値が、前記所定の値未満であれば当該多値画像情報（ｘ，ｙ）の領域を第１像域（写真網点画像の領域）と判定する一方、前記所定の値以上であれば当該多値画像情報（ｘ，ｙ）の領域を第２像域（文字画像の領域）と判定して該第１及び第２の像域（写真網点画像及び文字画像の領域）を分離する。なお、ここでは、前記式中のｍ及びｎの値をいずれも５とし、前記所定の値を１１５０としている。
【００６０】
像域分離手段Ａ３はまた、画像読取装置２ａから入力された複数の多値画像情報に対応する複数の像域識別用画像情報であって、各対応する多値画像情報が第１像域（写真網点画像の領域）ＰＡ（図６（Ａ）参照）にあると判定したときは該第１像域（写真網点画像の領域）ＰＡと識別できる第１の識別符号（ここでは１２８）と、第２像域（文字画像の領域）ＣＡ（図６（Ａ）参照）にあると判定したときは該第２像域（文字画像の領域）ＣＡと識別できる第２の識別符号（ここでは０）とを備えた複数の像域識別用画像情報Ｄ（図６（Ｂ）参照）を予め形成しておく像域識別用画像情報形成手段Ａ３’を備えている。
【００６１】
中間調画像用多値画像情報形成手段Ａ４は、画像読取装置２ａから入力された複数の多値画像情報に対応する複数の中間調画像用（写真網点画像用）多値画像情報であって、各対応する多値画像情報に対してそれぞれ中間調画像用（写真網点画像用）処理を施して複数の中間調画像用（写真網点画像用）多値画像情報Ｂ（図７（Ａ）参照）を予め形成しておくものである。
【００６２】
この中間調画像用多値画像情報形成手段Ａ４における前記中間調画像用（写真網点画像用）処理は、モアレを除去するための、注目画素を中心とする画素領域で近傍画素の階調の平均（ここでは３行×３列の画素領域で８近傍の階調の平均）をとる従来周知の「ぼかし処理」及び注目画素を中心とする画素領域での階調の平均（ここでは５行×５列、７行×７列、９行×９列の画素領域のうちオペレータ等により選択される画素領域での階調の平均）をとる「平均化処理」からなる「モアレ除去処理」と、中間調を表現するための従来周知の「ディザ法」による「ディザ処理」とを含むものである。なお、前記の「モアレ除去処理」は、前記の「ディザ処理」の前に行う。
【００６３】
さらに説明すると、前記モアレ除去処理のうち、前記ぼかし処理は、前記第１補正処理手段Ａ１のものと同様であり、ここでは説明を省略する。また、前記平均化処理では、多値画像情報の２次元座標を（ｘ，ｙ）とすると、注目画素（ｘ，ｙ）を中心とする、５行×５列、７行×７列、９行×９列の画素領域のうち、後述するように、オペレータ等により選択される画素領域、例えば、５行×５列の画素領域での階調の平均［｛ｆ（ｘ−２，ｙ−２）＋ｆ（ｘ−１，ｙ−２）＋ｆ（ｘ，ｙ−２）＋・・・＋ｆ（ｘ，ｙ）＋・・・＋ｆ（ｘ，ｙ＋２）＋ｆ（ｘ＋１，ｙ＋２）＋ｆ（ｘ＋２，ｙ＋２）｝／２５］をとり（図８（Ａ）の斜線部参照）、これを多値画像情報全体に対して行う。
【００６４】
また、前記ディザ処理では、ここでは、４行×４列のマトリックスにおいて４×４＝１６通りの閾値を設定しておき、多値画像情報の４行×４列の画素領域で各画素の階調に対してマトリックスの閾値を当てはめていき、各画素が閾値より大きければ２５５、小さければ０に置き換え、これを多値画像情報全体に対して行う（図８（Ｂ）参照）。
【００６５】
非中間調画像用多値画像情報形成手段Ａ５は、画像読取装置２ａから入力された複数の多値画像情報に対応する複数の非中間調画像用（文字画像用）多値画像情報であって、各対応する多値画像情報に対してそれぞれ非中間調画像用（文字画像用）処理を施して複数の非中間調画像用（文字画像用）多値画像情報Ｃ（図７（Ｂ）参照）を予め形成しておくものである。
【００６６】
この非中間調画像用多値画像情報形成手段Ａ５における前記非中間調画像用（文字画像用）処理は、ここでは、画像読取装置２ａから入力された複数の多値画像情報について、従来周知の誤差拡散法による誤差拡散処理を行うものである。
【００６７】
さらに説明すると、前記誤差拡散処理では、各多値画像情報について、２次元座標を（ｘ，ｙ）とすると、ここでは注目画素（ｘ，ｙ）の階調ｆ（ｘ、ｙ）に対応する階調１２７を境にした固定閾値法による２階調変換後の階調の値（２５５又は０）をｇ（ｘ，ｙ）としたとき、Ｅｘ，ｙ＝ｆ（ｘ、ｙ）−ｇ（ｘ，ｙ）で生じる誤差を、次式に示すように、注目画素の右隣へ３／８、下へ３／８、右下へ１／４配分し、この処理を各画素に対して施す。
ｆｎｅｗ（ｘ，ｙ）＝ｇ（ｘ，ｙ）
ｆｎｅｗ（ｘ＋１，ｙ）＝ｆ（ｘ＋１，ｙ）＋（３／８）×Ｅｘ，ｙ
ｆｎｅｗ（ｘ，ｙ＋１）＝ｆ（ｘ，ｙ＋１）＋（３／８）×Ｅｘ，ｙ
ｆｎｅｗ（ｘ＋１，ｙ＋１）＝ｆ（ｘ＋１，ｙ＋１）＋（１／４）×Ｅｘ，ｙ
【００６８】
多値画像処理手段Ａ６は、画像読取装置２ａから入力された複数の多値画像情報について、像域分離手段Ａ３による第１及び第２の像域（写真網点画像及び文字画像の領域）の判定結果に基づいて、それぞれの像域に応じた画像処理を行うものであり、像域分離手段Ａ３の像域識別用画像情報形成手段Ａ３’にて予め形成された複数の像域識別用画像情報Ｄによる第１及び第２の像域（写真網点画像及び文字画像の領域）の識別結果に基づいて、それぞれの像域に応じた画像処理を行うものである。
【００６９】
さらに説明すると、多値画像処理手段Ａ６では、画像読取装置２ａから入力された各多値画像情報について、像域分離手段Ａ３の像域識別用画像情報形成手段Ａ３’にて予め形成された、対応する像域識別用画像情報Ｄ（図６（Ｂ）参照）により第１像域（写真網点画像の領域）ＰＡであると識別されたときは中間調画像用多値画像情報形成手段Ａ４にて予め形成された、対応する中間調画像用（写真網点画像用）多値画像情報Ｂとする一方（図７（Ａ）参照）、第２像域（文字画像の領域）ＣＡであると識別されたときは非中間調画像用多値画像情報形成手段Ａ５にて予め形成された、対応する非中間調画像用（文字画像用）多値画像情報Ｃとする（図７（Ｂ）参照）。
【００７０】
二値化処理手段Ａ７は、多値画像処理手段Ａ６にて画像処理された複数の多値画像情報をいずれも二値画像情報に変換して二値化するものである。
【００７１】
一方、図４に示すコンピュータ２００は、中央処理装置２１０、メインメモリ２２０、ファイル装置２３０、入力装置２４０及び出力装置２５０を備えている。入力装置２４０は、前記の画像処理プログラムＱを含むスキャナドライバＱ’を記録したコンピュータ読み取り可能な記録媒体３００からのプログラム情報を読み出すことができる読出部２４１と、マウスやキーボード等の入力部２４２とを備えている。ファイル装置２３０は、前記記録媒体３００から読み出されたプログラム情報を記録することができる。メインメモリ２２０は、ファイル装置２３０に記録されたプログラム情報を読込むことができ、プログラム作業領域を有している。出力装置２５０は、ここではディスプレイ等の画像表示装置２５１及びプリンタ等の印刷装置２５２を備えており、画像処理プログラムＱにより画像処理された情報を表示及び／又は印刷することができる。また、中央処理装置２１０は、メインメモリ２２０に読み込まれたスキャナドライバＱ’に基づいてスキャナ１全体を制御することができる。
【００７２】
前記画像処理プログラムＱは、コンピュータ２００を、前記したような手段、すなわち第１補正処理手段Ａ１、第２補正処理手段Ａ２、像域識別用画像情報形成手段Ａ３’を備えた像域分離手段Ａ３、中間調画像用多値画像情報形成手段Ａ４、非中間調画像用多値画像情報形成手段Ａ５、多値画像処理手段Ａ６及び二値化処理手段Ａ７として機能させるためのものである。これら手段の詳細については、既述した通りであり、ここでは説明を省略する。
【００７３】
以上説明した画像処理装置Ａ及び画像処理プログラムＱでは、次の（ａ）及び（ｂ）の調整のうちユーザーによって選択された調整によって調整された画像処理が実行される。
（ａ）ガンマ補正調整の選択
（ｂ）モアレ除去調整の選択
【００７４】
図９に画像処理装置Ａによって操作部Ｓ（図３参照）の表示画面上に表示された選択画面を示し、図１０に画像処理プログラムＱによってディスプレイ２５１（図４参照）上に表示された選択画面を示す。
【００７５】
画像処理装置Ａによる選択画面では、図９に示すように、ガンマ補正調整を選択するときは、（ａ）の「ガンマ補正調整」をキー操作により選択し（図９（Ａ）参照）、次に表示されるガンマ補正調整選択画面ａ（図９（Ｂ）参照）におけるガンマ補正調整レベル「Ａ」、「Ｂ」、「Ｃ」をキー入力することで、前記した割当テーブルの種類を変更することができる。モアレ除去調整を選択するときは、（ｂ）の「モアレ除去調整」をキー操作により選択し（図９（Ａ）参照）、次に表示されるモアレ除去調整選択画面ｂ（図９（Ｃ）参照）におけるモアレ除去調整レベル「弱」、「中」、「強」をキー入力することで、前記した「平均化処理」における画素領域を５行×５列：「弱」、７行×７列：「中」、９行×９列：「強」に変更することができる。
【００７６】
画像処理プログラムＱによる調整画面では、図１０に示すように、ガンマ補正調整を選択するときは、ガンマ補正調整選択画面ａにおけるガンマ補正調整レベル「Ａ」、「Ｂ」、「Ｃ」をマウス操作によるポインタでクリックすることで、前記した割当テーブルの種類を変更することができる。モアレ除去調整を選択するときは、モアレ除去調整選択画面ｂにおけるモアレ除去調整レベル「弱」、「中」、「強」をクリックすることで、前記した「平均化処理」における画素領域を５行×５列：「弱」、７行×７列：「中」、９行×９列：「強」に変更することができる。
【００７７】
図１１は画像処理装置Ａ及び画像処理プログラムＱによる処理の流れを示すフローチャートである。また、図１２（Ａ）は図１１に示すフローチャートにおけるモアレ除去処理のサブルーチンの流れを示すフローチャートであり、図１２（Ｂ）は図１１に示すフローチャートにおける像域分離処理のサブルーチンの流れを示すフローチャートである。
【００７８】
画像処理装置Ａ及び画像処理プログラムＱでは、図１１に示すように、画像読取装置２ａから入力された複数の多値画像情報は、まず、第１補正処理手段Ａ１による第１補正処理がなされるとともに（ステップＳａ）、第２補正処理手段Ａ２による第２補正処理がなされる（ステップＳｂ）。
【００７９】
すなわち、第１補正処理Ｓａでは、画像読取装置２ａから入力された各多値画像情報の階調について、まず、前記の選択画面ａにて選択された割当テーブルが参照されることで階調変換されて「ガンマ補正」がなされ（ステップＳ１）、次に、注目画素を中心とする３行×３列の画素領域で８近傍の階調の平均がとられて「ぼかし処理」がなされる（ステップＳ２）。
【００８０】
また、第２補正処理Ｓｂでは、画像読取装置２ａから入力された各多値画像情報の階調について、まず、上限階調（ここでは２４０）以上及び下限階調（ここでは１０）以下は原稿画像のベース濃度として処理されて「ハイライト／シャドウ処理」がなされ（ステップＳ３）、次に、前記の選択画面ａにて選択された割当テーブルが参照されることで階調変換されて「ガンマ補正」がなされる（ステップＳ４）。
【００８１】
次いで、第１補正処理Ｓａされた各多値画像情報は、像域識別用画像情報形成手段Ａ３’を備えた像域分離手段Ａ３による像域分離処理がなされる（ステップＳ５）。また、前記第２補正処理Ｓｂされた各多値画像情報は、中間調画像用多値画像情報形成手段Ａ４による中間調画像用多値画像情報形成処理がなされるとともに（ステップＳｃ）、非中間調画像用多値画像情報形成手段Ａ５による非中間調画像用多値画像情報形成処理がなされる（ステップＳｄ）。
【００８２】
すなわち、像域分離処理Ｓ５では、図１２（Ａ）に示すように、第１補正処理Ｓａされた各多値画像情報について、ｍ及びｎの値をいずれも５とした前記計算式Ｂｕｎｓａｎ（ｘ，ｙ）にそれぞれ代入して分散値Ｖを求め（ステップＳ５１）、得られた分散値が、所定の値（ここでは１１５０）未満であれば当該多値画像情報の領域を第１像域（写真網点画像の領域）ＰＡと判定し（ステップＳ５２）、該第１像域（写真網点画像の領域）ＰＡと識別できる第１の識別符号（ここでは１２８）を付すとともに（ステップＳ５３）、得られた分散値が、所定の値（ここでは１１５０）以上であれば当該多値画像情報の領域を第２像域（文字画像の領域）ＣＡと判定し（ステップＳ５２）、該第２像域（文字画像の領域）ＣＡと識別できる第２の識別符号（ここでは０）を付す（ステップＳ５４）ことで、複数の像域識別用画像情報Ｄ（図６（Ｂ）参照）が形成される。
【００８３】
また、図１１に示す中間調画像用多値画像情報形成処理Ｓｃでは、第２補正処理Ｓｂされた各多値画像情報について、まず、モアレ除去処理がなされたあと（ステップＳ６）、ディザ処理がなされる（ステップＳ７）。
【００８４】
すなわち、モアレ除去処理Ｓ６では、図１２（Ｂ）に示すように、第２補正処理Ｓｂされた各多値画像情報について、注目画素を中心とする３行×３列の画素領域で８近傍の階調の平均がとられて「ぼかし処理」がなされたあと（ステップＳ６１）、注目画素を中心とする、５行×５列、７行×７列、９行×９列の画素領域のうち、前記の選択画面ｂにて選択された画素領域での階調の平均がとられて「平均化処理」がなされる（ステップＳ６２）。
【００８５】
また、図１１に示すディザ処理Ｓ７では、モアレ除去処理Ｓ６された各多値画像情報について、４行×４列の画素領域で各画素の階調に対してマトリックスの閾値が当てはめられ、各画素が閾値より大きければ２５５、小さければ０に置き換えられる。
【００８６】
一方、非中間調画像用多値画像情報形成処理Ｓｄでは、第２補正処理Ｓｂされた各多値画像情報について、注目画素のＥｘ，ｙ＝ｆ（ｘ、ｙ）−ｇ（ｘ，ｙ）で生じる誤差が注目画素の右隣へ３／８、右下へ１／４、下へ３／８配分されて「誤差拡散処理」がなされる（ステップＳ８）。
【００８７】
さらに、多値画像処理手段Ａ６による多値画像処理がなされたあと（ステップＳ９）、二値化処理手段Ａ７による二値化処理がなされる（ステップＳ１０）。
【００８８】
すなわち、多値画像処理Ｓ９では、画像読取装置２ａから入力された複数の多値画像情報について、像域分離処理Ｓ５にて予め形成された、対応する像域識別用画像情報Ｄ（図６（Ｂ）参照）により識別符号が“１２８”の第１像域（写真網点画像の領域）であると識別されたときは中間調画像用多値画像情報形成処理Ｓｃにて予め形成された、対応する中間調画像用（写真網点画像用）多値画像情報Ｂ（図７（Ａ）参照）とされる一方、識別符号が“０”の第２像域（文字画像の領域）であると識別されたときは非中間調画像用多値画像情報形成処理Ｓｄにて予め形成された、対応する非中間調画像用（文字画像用）多値画像情報Ｃ（図７（Ｂ）参照）とされる。
【００８９】
そして、二値化処理Ｓ１０では、多値画像処理Ｓ９にて形成された複数の多値画像情報を閾値１２７とした固定閾値法により二値化して複数の二値画像情報に変換する。
【００９０】
以上説明した画像処理装置Ａ及び画像処理プログラムＱによると、画像読取装置２ａから入力され、第１補正処理Ｓａにて処理された複数の多値画像情報の領域のうち、連続的な階調性を有する第１の像域（写真網点画像の領域）と該第１像域以外の第２の像域（文字画像の領域）とを判定して該第１及び第２の像域（写真網点画像及び文字画像の領域）を分離するにあたり、該複数の多値画像情報の階調を、分散値を求める所定の計算式にそれぞれ代入して分散値を求め、得られた分散値が、所定の値未満であれば当該多値画像情報の領域を第１像域（写真網点画像の領域）と判定する一方、前記所定の値以上であれば当該多値画像情報の領域を第２像域（文字画像の領域）と判定して該第１及び第２の像域（写真網点画像及び文字画像の領域）を分離し、多値化された前記複数の多値画像情報について、第１及び第２の像域（写真網点画像及び文字画像の領域）の判定結果に基づいて、それぞれの像域に応じた画像処理を行うので、像域分離を行うための処理を簡素化でき、それだけ全体の画像処理に要する時間を短くでき、画像処理のためのハードウェアを構成する回路規模やソフトウェアを構成するプログラム規模を小さくでき、それだけコストダウンを実現できる。
【００９１】
また、非中間調画像用（文字画像用）処理Ｓｄでは、画像読取装置２ａから入力され、第２補正処理Ｓｂにて処理された複数の多値画像情報について、従来周知の誤差拡散法による誤差拡散処理を行うので、複数の多値画像情報の領域のうちの第２像域（文字画像の領域）であって、例えば、比較的小さい文字画像の領域や濃度が若干淡い中間調の文字画像の領域であっても、二値画像情報に変換して二値化したときに、中間調再現性よく、ノイズによる輪郭部分の強調が抑制された文字画像を得ることができる。
【００９２】
なお、画像処理装置Ａ及び画像処理プログラムＱは、中間調画像用（写真網点画像用）多値画像情報Ｂ及び非中間調画像用（文字画像用）多値画像情報Ｃを予め形成しておくので、画像読取装置２ａから入力された各多値画像情報について、原稿画像を読み込んだあとに、いずれも中間調画像用（写真網点画像用）多値画像情報とすることも、いずれも非中間調画像用（文字画像用）多値画像情報とすることもできる。
【００９３】
【発明の効果】
以上説明したように本発明によると、像域分離を行うための処理を簡素化でき、それだけ全体の画像処理に要する時間を短くできるとともに、画像処理のためのハードウェアを構成する回路規模やソフトウェアを構成するプログラム規模を小さくでき、ひいてはコストダウンを実現できる画像処理方法及び画像処理装置を提供することができる。
【００９４】
また、本発明によると、像域分離を行うための処理を簡素化でき、それだけ全体の画像処理に要する時間を短くできるとともに、画像処理のためのソフトウェアを構成するプログラム規模を小さくでき、ひいてはコストダウンを実現できる画像処理プログラム及び当該画像処理プログラムを記録したコンピュータ読み取り可能な記録媒体を提供することができる。
【図面の簡単な説明】
【図１】本発明に係る画像処理方法を実施する画像処理装置の一例を備えたスキャナの概略斜視図である。
【図２】図１に示すスキャナにおける画像読取装置の概略側面図である。
【図３】図１に示すスキャナがデジタル複写機に搭載されたときの画像処理装置を中心とする概略ブロック図である。
【図４】図１に示すスキャナの画像読取装置がコンピュータに接続されたときの該コンピュータを中心とする概略ブロック図である。
【図５】図（Ａ）は「ガンマ補正処理」を説明するための図であり、図（Ｂ）は「ぼかし処理」を説明するための図である。
【図６】図（Ａ）は原稿画像の一例を示し、図（Ｂ）は図（Ａ）に示す原稿画像に対応する像域識別用画像情報の一例を識別符号で示す概略図である。
【図７】図（Ａ）は図６（Ａ）に示す原稿画像に対応する中間調画像用（写真網点画像用）多値画像情報の一例を出力画像で示した概略図であり、図（Ｂ）は図６（Ａ）に示す原稿画像に対応する非中間調画像用（文字画像用）多値画像情報の一例を出力画像で示した概略図である。
【図８】図（Ａ）は「ぼかし処理」を説明するための図であり、図（Ｂ）は「ディザ処理」を説明するための図である。
【図９】画像処理装置によって操作部の表示画面上に表示された選択画面を示す図であり、図（Ａ）は調整選択メニューを、図（Ｂ）はガンマ補正調整選択画面を、図（Ｃ）はモアレ除去調整選択画面を示すものである。
【図１０】画像処理プログラムによってディスプレイ上に表示された選択画面を示す図である。
【図１１】画像処理装置及び画像処理プログラムによる処理の流れを示すフローチャートである。
【図１２】図（Ａ）は図１１に示すフローチャートにおけるモアレ除去処理のサブルーチンの流れを示すフローチャートであり、図（Ｂ）は図１１に示すフローチャートにおける像域分離処理のサブルーチンの流れを示すフローチャートである。
【図１３】第２像域（文字画像の領域）において誤差拡散処理を行う領域の一例を示す図である。
【図１４】原稿画像の種類と階調性との関係を概念的に示す図である。
【符号の説明】
Ａ１…第１補正処理手段 Ａ２…第２補正処理手段
Ａ３…像域分離手段 Ａ３’…像域識別用画像情報形成手段
Ａ４…中間調画像用多値画像情報形成手段
Ａ５…非中間調画像用多値画像情報形成手段 Ａ６…多値画像処理手段
Ａ７…二値化処理手段[0001]
BACKGROUND OF THE INVENTION
The present invention digitally processes multi-level image information of a plurality of gradations input from an image reading device, converts the binary image information into binary image information, binarizes the image, and outputs the binarized binary image information The present invention relates to a processing method, an image processing apparatus, an image processing program, and a recording medium.
[0002]
[Prior art]
2. Description of the Related Art Conventionally, image processing apparatuses have been widely used as image processing means such as scanners mounted on image forming apparatuses such as digital copying machines and scanners connected to computers and the like.
[0003]
In this type of image processing apparatus, usually, an original image is divided into a plurality of pixels, a plurality of image information is read, and the read image information is converted into multi-level image information of a plurality of gradations. The plurality of multi-valued image information input from the apparatus is digitally processed, converted into binary image information, binarized, and the binarized binary image information is output. When reading a color document image, the color document image is separated into, for example, red (R), green (G), and blue (B), and each color document image is divided into a plurality of pixels. Corresponding to the colors input from the color image reading apparatus that converts the plurality of pieces of image information corresponding to the read colors into multi-value image information of a plurality of gradations and multi-values them. After a plurality of pieces of multi-value image information are digitally processed, it is converted into binary image information and binarized, and the binarized binary image information is output.
[0004]
By the way, a document image generally has a so-called photographic image (hereinafter referred to as a photographic image) having continuous gradation like a silver salt photograph, a size of fine dots (dots), and a dot density. Three types of so-called halftone dot images (hereinafter referred to as halftone dot images) such as halftone dot printing expressing gradation by changing and so-called character images (hereinafter referred to as character images) such as characters and line drawings Can be roughly divided into
[0005]
In an image processing apparatus that performs image processing on these photographic images, halftone dot images, and character images, image processing corresponding to each document image is performed on a plurality of multi-valued multi-value image information. For example, in order to express halftones in a photographic image, image processing for photographic images such as “dither processing” using the conventionally known “dither method” or “error diffusion processing” using “error diffusion method” is performed, or halftone dots are used. In the image, in order to remove moire, a well-known conventional “takes an average of gradations of neighboring pixels in a pixel region centered on the target pixel (for example, an average of gradations of eight neighborhoods in a pixel region of 3 rows × 3 columns). “Blurring process” and / or average of gradation in pixel area centered on target pixel (for example, average of gradation in pixel area of 5 rows × 5 columns, 7 rows × 7 columns, 9 rows × 9 columns, etc.) The image processing for a halftone image including “averaging” is performed.
[0006]
However, when image processing is performed on a document image in which at least two of a photographic image, a halftone image, and a character image are mixed, for example, image processing for photographic images and halftone dots are performed in a photographic image area and a halftone image area. If the image processing for the image is not performed and converted to binary image information, the halftone of the photographic image is not expressed well, or moire occurs in the halftone image. Further, when image processing for photographic images or image processing for halftone images is performed on the character image area and converted to binary image information, the character image becomes blurred.
[0007]
Therefore, when reading a document image in which at least two of a photographic image, a halftone image, and a character image are read, at least two of the photographic image, the halftone image, and the character image in the multivalued multivalue image information area are read. Various image processing apparatuses that determine one image area and separate the image areas and perform image processing according to the image area based on the determination result of the image area for the multi-valued multi-value image information are proposed. Has been.
[0008]
As such an image area separation method, for example, a method of performing image area separation by determining an image area by using pattern matching for information on a document image in which character images, photographic images, and the like are mixed ( For example, see Patent Document 1).
[0009]
[Patent Document 1]
JP-A-6-150059
[0010]
[Problems to be solved by the invention]
However, when the image processing is performed by using the technique of determining the image area by using pattern matching and performing image area separation on the information of the original image in which character images, photographic images, and the like are mixed as described above. The processing for image area separation is complicated, which increases the time required for the entire image processing, and the circuit scale constituting the hardware for image processing and the program scale constituting the software. It grows and costs increase.
[0011]
The present invention has been made in view of the above problems, and can simplify the processing for image area separation, shorten the time required for the entire image processing, and configure hardware for image processing. It is an object of the present invention to provide an image processing method and an image processing apparatus capable of reducing the circuit scale and the program scale constituting the software, and thus realizing cost reduction.
[0012]
In addition, the present invention can simplify the processing for image area separation, shorten the time required for the entire image processing, reduce the scale of the program constituting the software for image processing, and thus reduce the cost. It is an object of the present invention to provide an image processing program capable of realizing the above and a computer-readable recording medium on which the image processing program is recorded.
[0013]
[Means for Solving the Problems]
The present inventor has made extensive studies to solve the above problems, and has found the following.
[0014]
FIG. 14 is a diagram conceptually showing the relationship between the type of document image and gradation. In a plurality of multivalued image information in which a document image in which a photographic image, a halftone dot image, and a character image are mixed is read and the read image information is multivalued into a plurality of gradations, as shown in FIG. Of the multivalued multivalued image information area, the central area X can be regarded as a photographic image area, and the area Y slightly outside the central area X is a halftone image area and further outside. The region Z can be regarded as a character image region.
[0015]
When image processing is performed on a document image in which at least two of a photographic image, a halftone dot image, and a character image are mixed, the photographic image, the halftone dot image, and the character image are displayed in a plurality of multivalued multivalued image information areas. Even if all three image areas X, Y, and Z are not determined, for example, a photographic halftone image area X ′ including a halftone image area Y in the photographic image area X, and a character image area Z And the image processing for the halftone image such as the “blurring processing” is performed on the region X ′ of the photographic halftone image, and for the photographic image such as the “dither processing”. A good output image can be obtained by performing image processing.
[0016]
In addition, when image processing is performed on a document image in which at least two of a photographic image, a halftone dot image, and a character image are mixed, a plurality of multi-valued image information areas are determined by determining a plurality of image areas. Of the multi-valued image information within a predetermined range (for example, within the range of pixels of 11 rows × 11 columns), the dispersion value is obtained using a predetermined calculation formula for obtaining the dispersion, and the levels of the dark region and the light region are calculated. By shifting the tonal gathering in units of one pixel and looking at this as a whole for multi-value image information, in the photographic image, the tonal gathering varies between the dark and light areas, and the variance value is In the character image, the variance value tends to increase due to the collection of gradations between the dark area and the light area.
[0017]
From the above, among the multivalued multivalued image information areas, the first image area having continuous gradation (more specifically, the area of the photographic halftone image) and the first A second image area other than the image area (more specifically, a character image area) is determined to separate the first and second image areas (photo halftone dot image and character image area); In performing this image area separation, the gradation of the plurality of multi-valued image information is a predetermined calculation formula for obtaining a dispersion value, specifically,
[Equation 3]

If the obtained dispersion value is less than a predetermined value that tends to vary the collection of gradations between the dark area and the light area, the area of the multi-value image information is obtained. Is determined to be the first image area (photo halftone image area), and the multi-valued image is determined if it is equal to or greater than the predetermined value that tends to be a collection of gradations between the dark area and the light area. The area of information is determined to be the second image area (character image area), the first and second image areas (photo halftone image area and character image area) are separated, and the multivalued If image processing corresponding to each image area is performed based on the determination results of the first and second image areas (photo halftone image area and text image area) for a plurality of multi-value image information, The processing for performing the separation is simplified, and the time required for the entire image processing is shortened. , Program size included in a circuit scale and software to configure the hardware for image processing is reduced, leading to cost reduction, to obtain a knowledge that.
[0018]
The present invention is based on such knowledge, and provides the following image processing method, image processing apparatus, image processing program, and recording medium in order to solve the above problems. That is,
(1) Image processing method
A document image is divided into a plurality of pixels to read a plurality of pieces of image information, and the read pieces of the plurality of pieces of image information are all converted into a plurality of gradations of multi-value image information and inputted from an image reading device that performs multi-value conversion. Of the plurality of multivalued image information areas, a first image area having continuous gradation (more specifically, a photographic halftone image area) and a second image other than the first image area. An image area separation step of determining a region (more specifically, a character image region) and separating the first and second image regions (photo halftone image region and character image region), The gradation of the multi-valued image information is substituted into a predetermined calculation formula for obtaining a dispersion value to obtain a dispersion value. If the obtained dispersion value is less than a predetermined value, the region of the multi-valued image information is While determining as the first image area (area of the photographic halftone image), if it is greater than or equal to the predetermined value, the multi-value image And image area separation step of separating the regions of the multi-address the second image area of said first and second image area is determined (in the character image area) and (areas of the photographic halftone image and a character image),
For the plurality of multi-value image information input from the image reading device, based on the determination result of the first and second image areas (photo halftone image area and character image area) by the image area separation step, A multi-value image processing step for performing image processing according to each image area;
A binarization processing step of converting all of the plurality of multi-value image information subjected to image processing in the multi-value image processing step into binary image information and binarizing;
An image processing method comprising:
[0019]
(2) Image processing device
A document image is divided into a plurality of pixels to read a plurality of pieces of image information, and the read pieces of the plurality of pieces of image information are all converted into a plurality of gradations of multi-value image information and inputted from an image reading device that performs multi-value conversion. Of the plurality of multivalued image information areas, a first image area having continuous gradation (more specifically, a photographic halftone image area) and a second image other than the first image area. An image area separating means for determining a region (more specifically, a character image region) and separating the first and second image regions (photo halftone image region and character image region), The gradation of the multi-valued image information is substituted into a predetermined calculation formula for obtaining a dispersion value to obtain a dispersion value. If the obtained dispersion value is less than a predetermined value, the region of the multi-valued image information is While determining as the first image area (area of the photographic halftone image), if it is not less than the predetermined value, the multi-value image information An image area separating means for separating the region second image area of said first and second image area is determined (in the character image area) and (areas of the photographic halftone image and a character image),
For the plurality of multi-value image information input from the image reading device, based on the determination results of the first and second image areas (photo halftone dot image and character image area) by the image area separation unit, Multi-value image processing means for performing image processing according to each image area;
Binarization processing means for converting the plurality of multivalued image information processed by the multivalued image processing means into binary image information and binarizing them;
An image processing apparatus comprising:
[0020]
(3) Image processing program
An image processing program for causing a computer connected to the image reading apparatus to execute each step of the image processing method according to the present invention.
[0021]
(4) Recording medium
A computer-readable recording medium on which the image processing program according to the present invention is recorded.
[0022]
According to the image processing method, the image processing apparatus, the image processing program, and the recording medium according to the present invention, among the plurality of multi-value image information areas input from the image reading apparatus, there is a continuous gradation property. A first image area (photo halftone image area) and a second image area (character image area) other than the first image area are determined to determine the first and second image areas (photo halftone image). And the character image area), the gradations of the plurality of multi-value image information are respectively substituted into a predetermined calculation formula for determining a variance value to obtain a variance value, and the obtained variance value is a predetermined value. If it is less than the value, the multi-value image information area is determined as the first image area (photo halftone image area), and if it is greater than the predetermined value, the multi-value image information area is determined as the second image area. The first and second image areas (photo halftone images and characters) determined as image areas (character image areas) Image region), and for the multi-valued multi-value image information, based on the determination results of the first and second image regions (photo halftone image and character image regions), Since image processing corresponding to each image area is performed, the process for performing image area separation can be simplified, and the time required for the entire image processing can be shortened accordingly.
[0023]
Furthermore, according to the image processing method and the image processing apparatus according to the present invention, the processing for performing the image area separation can be simplified, so the circuit scale constituting the hardware for image processing and the program scale constituting the software can be reduced. The size can be reduced and the cost can be reduced accordingly.
[0024]
Furthermore, according to the image processing method, the image processing program, and the recording medium according to the present invention, the processing for performing the image area separation can be simplified, so that the scale of the program constituting the software for image processing can be reduced, and the cost can be reduced accordingly. Down can be realized.
[0025]
In the image processing method, the image processing apparatus, the image processing program, and the recording medium according to the present invention, the two-dimensional coordinates of the plurality of multivalued image information input from the image reading apparatus are (x, y), and the plurality of Assuming that the gradation of the multi-value image information (x, y) is Pixel (x, y), a predetermined calculation formula Bunsan (x, y) for obtaining the dispersion value in the image area separation step and the image area separation means. Is
[Expression 4]

Can be illustrated. In this case, in the image area separation step and the image area separation means, if the variance value according to the calculation formula Bunsan (x, y) is less than the predetermined value, the multi-value image information (x, y) While determining the region as the first image region (region of the photographic halftone image), if the region is greater than the predetermined value, the region of the multi-value image information (x, y) is determined as the second image region (character image). The first and second image areas (photo halftone dot image and character image areas) can be separated.
[0026]
If the values of m and n in the above equation are too small, the dispersion of gradation in a small range will be examined, and the first and second image areas (photo halftone image and character image areas) will be examined. ) Is likely to be extreme. On the other hand, if it is too large, the dispersion of gradation in a large range will be examined, and the determination accuracy of the first and second image areas (photo halftone image area and character image area) in a small area will decrease. Easy to do. Accordingly, the values of m and n in the above formula are not limited thereto, but can be exemplified by about 3 to 5. Further, both m and n may be set to the same value. Furthermore, m and n may be fixed to predetermined values, but may be adjusted as appropriate by an operator or the like.
[0027]
In the image forming method, the image forming apparatus, the image forming program, and the recording medium according to the present invention, the gradation of the multi-value image information input from the image reading apparatus is 256 gradations, and the values of m and n in the above formula As for all, the predetermined value in the image area separation step and the image area separation means is about 1150.
[0028]
In the image processing method, the image processing program, and the recording medium according to the present invention, the image area separation step includes a plurality of image area identifying image information corresponding to the plurality of multi-value image information input from the image reading apparatus. When it is determined that each corresponding multi-value image information is in the first image area (photo halftone image area), the first image area (photo halftone image area) can be identified. And when it determines with it being in the said 2nd image area (area | region of a character image), the several image area identification image information each provided with the identification code so that it can identify with this 2nd image area (area | region of a character image) An image area identifying image information forming step in which the image in the image area separating step is applied to the plurality of multi-value image information input from the image reading device. Image information for area identification Based on the identification results of the first and second image areas (photo halftone image area and character image area) based on the plurality of image area identification image information formed in advance in the forming step, The corresponding image processing may be performed.
[0029]
In the image processing apparatus according to the present invention, the image area separating means includes a plurality of image area identifying image information corresponding to the plurality of multi-value image information input from the image reading apparatus, When it is determined that the value image information is in the first image area (photo halftone image area), the second image area can be distinguished from the first image area (photo halftone image area). An image in which a plurality of image area identification image information each having an identification code so as to be distinguished from the second image area (character image area) when it is determined to be in (character image area) Image information forming means for area identification, wherein the multi-value image processing means is configured to perform image area identification image information forming means of the image area separating means for the plurality of multi-value image information input from the image reading device. The plurality of image area identifying image information formed in advance at According based on the identification result of said first and second image area (area of the photographic halftone image and a character image), may be carried out image processing in accordance with each of the image area.
[0030]
In this case, prior to the execution of the image area identifying image information forming step and the image area identifying image information forming means, a first correction for correcting each of the multivalued image information input from the image reading device. Processing steps and first correction processing means may be further included. As the correction processing in the first correction processing step and the first correction processing means, for example, a conventionally known “gamma correction processing for performing gamma correction on each of the multi-value image information input from the image reading device in an adjustable manner. ”, A conventionally known“ blurring process ”that averages the gradations of neighboring pixels in the pixel area centered on the target pixel (for example, the average of gradations in the vicinity of 8 in the pixel area of 3 rows × 3 columns) Can do. Two or more of these processes may be performed in the first correction processing step and the first correction processing means.
[0031]
In the image processing method, the image processing program, and the recording medium according to the present invention, a plurality of halftone images corresponding to the plurality of multivalued image information input from the image reading device (more specifically, for a photographic halftone image) ) Multi-value image information, each of the corresponding multi-value image information is subjected to processing for halftone images (for photographic halftone images), and a plurality of halftone images (for photographic halftone images). A half-tone image multi-value image information forming step for forming image information in advance, and a plurality of non-half-tone images corresponding to the plurality of multi-value image information input from the image reading device (more specifically, Multi-value image information (for character images), each of the corresponding multi-value image information is subjected to processing for non-halftone images (for character images) and a plurality of non-halftone images (for character images) Value image information is pre-formed Macho may further include a multi-value image information forming step image. In this way, for each of the multi-value image information input from the image reading device, after reading the document image, both can be used as half-tone image (photo halftone image) multi-value image information. Any of them may be non-halftone image (character image) multi-value image information. Further, in the multi-value image processing step, the multi-value image information input from the image reading device is correspondingly formed in advance in the image area identification image information forming step of the image area separation step. When the image area identification image information identifies the first image area (photo halftone image area), the corresponding halftone image formed in advance in the halftone image multi-value image information forming step. While the multi-value image information for images (for photo halftone images) is identified as the second image area (character image area), in the multi-value image information forming step for non-halftone images Corresponding non-halftone image (character image) multi-value image information may be formed in advance.
[0032]
In the image processing apparatus according to the present invention, a plurality of half-value image information (more specifically, for a photographic halftone image) corresponding to the plurality of multi-value image information input from the image reading device. Then, each corresponding multi-value image information is subjected to halftone image (photo halftone image) processing to form a plurality of halftone image (photo halftone image) multi-value image information in advance. A half-tone image multi-value image information forming unit, and a plurality of non-half-tone image (more specifically, character image) multi-value images corresponding to the plurality of multi-value image information input from the image reading device A plurality of non-halftone image (character image) processing is performed on each corresponding multi-value image information to form a plurality of non-halftone image (character image) multi-value image information in advance. Multi-value image information forming means for non-halftone images It may be provided. In this way, each of the multi-value image information input from the image reading device can be used as half-tone image (photo halftone image) multi-value image information after the original image is read. Any of them can be non-halftone image (character image) multi-value image information. Further, in the multi-value image processing means, the multi-value image information input from the image reading device is correspondingly formed in advance by the image area identifying image information forming means of the image area separating means. When the image area identification image information identifies the first image area (photo halftone image area), the corresponding halftone image formed in advance by the halftone image multi-value image information forming means is used. On the other hand, when the multi-value image information for the image (photo halftone image) is identified as the second image area (character image area), the multi-value image information forming means for the non-halftone image is used. Corresponding non-halftone image (character image) multi-value image information may be formed in advance.
[0033]
In this case, the half-tone image multi-value image information forming step, the non-half-tone image multi-value image information forming step, the half-tone image multi-value image information forming means, and the non-half-tone image multi-value image information. Prior to the execution of the forming means, the image forming apparatus may further include a second correction processing step and a second correction processing means for correcting each of the multi-value image information input from the image reading apparatus. As the correction processing in the second correction processing step and the second correction processing means, for example, for each of the multi-value image information input from the image reading apparatus, a predetermined upper limit gradation and / or a predetermined lower limit Below the key, there are “highlight / shadow (HL / SD) processing” that is processed as the base density of the original image, and conventionally known “gamma correction processing” that adjustably adjusts gamma. Two or more of these processes may be performed in the second correction processing step and the second correction processing means.
[0034]
In general, among the multi-valued multi-valued image information area, for example, a relatively small character image or a halftone character image with a slightly lighter density is converted into binary image information and simply binary. If it is converted into a character image, the character image may be a character image with a strong density (for example, black), or may be a character image in which the outline portion is emphasized by noise that may occur at the periphery of the character image.
[0035]
Accordingly, in the non-halftone image multi-value image information forming step and the non-half-tone image multi-value image information forming means in the non-half-tone image (character image) processing, the non-halftone image multi-value image information forming unit As for the plurality of multi-value image information, for example, as shown in FIG. 13, within the predetermined gradation range (for example, the first image area ( Each half-value image information (within the hatched line in FIG. 13) in the vicinity of the dark dot halftone dot image area Z1 excluding X ′ to the dark dot halftone dot image area vicinity Z2) Part) can be subjected to error diffusion processing by a conventionally known error diffusion method. By doing so, the second image area (character image area) of the plurality of multi-value image information areas input from the image reading device, for example, a relatively small character image area, Even if it is a halftone character image area with a slightly lighter density, when converted to binary image information and binarized, a character image in which enhancement of the contour portion due to noise is suppressed with good halftone reproducibility is obtained. Obtainable.
[0036]
In the halftone image multi-value image information forming step and the half-tone image multi-value image information forming means, the halftone image (photo halftone image) processing includes a pixel of interest for removing moire. Centering on a conventionally known “blurring process” and / or focusing on a pixel of interest that takes the average of the gradation of neighboring pixels in the central pixel area (for example, the average of the gradation of eight neighboring areas in the pixel area of 3 rows × 3 columns) “Moire removal” including “averaging” that takes the average of the gradation in the pixel area (for example, the average of the gradation in the pixel area of 5 rows × 5 columns, 7 rows × 7 columns, 9 rows × 9 columns, etc.) “Process” ”,“ Dither processing ”by the conventionally known“ Dither method ”for expressing halftone, and the like can be exemplified. In the halftone image multi-value image information forming step and the halftone image multi-value image information forming means, two or more of these processes may be performed. In this case, the “moire removal process” can be performed before the “dither process”.
[0037]
In the image processing method, the image processing apparatus, the image processing program, and the recording medium according to the present invention, the color original image may be processed. In this case, the color document image is color-separated into, for example, red (R), green (G), and blue (B), and each color document image is divided into a plurality of pixels to read a plurality of pieces of image information. A plurality of pieces of image information corresponding to the respective colors read from the color image reading apparatus that converts the plurality of pieces of image information corresponding to the respective colors into multi-value image information having a plurality of gradations and multi-values them. After performing the digital processing as described above, it can be converted into binary image information and binarized, and the binarized binary image information can be output.
[0038]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments according to the present invention will be described below with reference to the drawings.
FIG. 1 is a schematic perspective view of a scanner 1 provided with an example A of an image processing apparatus for performing an image processing method according to the present invention. FIG. 2 is a schematic side view of an image reading apparatus 2a in the scanner 1 shown in FIG. is there. 3 is a schematic block diagram centering on the image processing apparatus A when the scanner 1 shown in FIG. 1 is mounted on the digital copying machine 100, and FIG. 4 is an image reading apparatus 2a of the scanner 1 shown in FIG. 2 is a schematic block diagram centering on the computer 200 when is connected to the computer 200. FIG.
[0039]
As shown in FIG. 3, the scanner 1 shown in FIG. 1 includes an image reading device 2 a and an image processing device A, and is mounted on the digital copying machine 100 (see FIG. 3) or connected to a computer 200. (See FIG. 4). The image reading device 2a reads a plurality of pieces of image information by dividing a document image into a plurality of pixels, and converts all of the read pieces of image information into multi-value image information of a plurality of gradations to be multi-valued. It is. When the scanner 1 is mounted on the digital copying machine 100, as shown in FIG. 3, after the multi-value image information multi-valued by the image reading device 2a is digitally processed by the image processing device A, The image data is converted into binary image information and binarized, and the binarized binary image information is output to the digital copying machine 100. When the scanner 1 is connected to the computer 200, as shown in FIG. 4, the binary image is obtained by digitally processing the multi-value image information multi-valued by the image reading device 2a by the computer 200. The information is converted into information and binarized, and the binarized binary image information is output to an output device 250 such as a printer 252 connected to the computer 200.
[0040]
When image processing is performed on a color original image, the color original image is color-separated into, for example, red (R), green (G), and blue (B) as an image reading device, and the original image of each color is displayed. A color image reading apparatus that reads a plurality of pieces of image information by dividing into a plurality of pixels, and converts a plurality of pieces of image information corresponding to the read colors into a plurality of levels of multi-value image information. Use. In this case, the processing is substantially the same as the case of processing a black and white original image except that the color original image is color-separated. The case where it does is demonstrated.
[0041]
Here, when the scanner 1 is mounted on the digital copying machine 100, as shown in FIG. 3, each means M and A1 to A7 of the image processing apparatus A can execute image processing. When connected to the computer 200, as shown in FIG. 4, the scanner is connected to the computer 200 by a computer-readable recording medium 300 on which a scanner driver Q ′ including an example Q of the image processing program according to the present invention is recorded. The computer 200 in which the driver Q ′ is installed is configured to be able to execute image processing.
[0042]
As shown in FIG. 1, the scanner 1 includes a scanner body 2 and an automatic document feeder 3. The scanner body 2 includes the image processing apparatus A, the image reading apparatus 2a, and the operation unit S. As shown in FIG. 2, the scanner 1 optically scans a document P image sent by an automatic document feeder 3 by an image reading device 2a, and reads image light coming from the document P image. The scanning of the original P image is performed by scanning the original P in the main scanning direction (transfer method) and the sub-scanning direction crossing the direction, and the image light coming by this scanning is read.
[0043]
As shown in FIG. 1, the automatic document feeder 3 includes a cover 5 that is rotatably supported on one side by a scanner main body 2 and abuts on and away from a glass plate (not shown) of the image reading device 2a. , The transfer mechanism 6 of the document P provided continuously with the cover 5, and a plurality of documents P And a document tray 7 on which is set. A pair of conveying rollers 8 shown in FIG. 2 is provided at both ends of the cover 5, and a conveying belt 9 is wound around both conveying rollers 8.
[0044]
As shown in FIG. 2, the transport mechanism 6 is provided with a guide plate 10 connected to the document tray 7 and a transport path R for the document P. The transfer mechanism unit 6 includes a paper feed roller 11 disposed below the document tray 7 and a transport roller 12 disposed in the middle of the transport path R for performing paper feed, paper discharge, and the like. . The guide plate 10 is curved in a substantially U-shaped cross section, and the tip portion of the curved guide plate 10 is disposed to face the peripheral surface of the transfer roller 12. Furthermore, one of the pair of transport rollers 8 is provided on the transfer path R. Thereby, the document P can be transferred from the document tray 7 to the image reading unit 4.
[0045]
As shown in FIG. 2, the image reading device 2a includes an image reading unit 4 and an A-D conversion unit 4a (not shown in FIG. 2, see FIGS. 3 and 4) disposed below the conveying roller 8. I have.
[0046]
Here, the image reading unit 4 includes a light emitting element, an imaging element, and an imaging lens (not shown). The image reading unit 4 irradiates light on a document image sent from the light emitting element by the automatic document feeder 3 and performs the above-described connection. With the image lens, the image light reflected from the original P image can be imaged on the image sensor and read. Thereby, the original P image can be divided into a plurality of pixels and read as a plurality of analog image information. As shown in FIGS. 3 and 4, the image reading unit 4 is connected to an A / D conversion unit 4a, and can send the read analog image information to the A / D conversion unit 4a.
[0047]
The A-D conversion unit 4a quantizes a plurality of analog image information sent from the image reading unit 4 with a plurality of bits (here, 8 bits) and converts the information into digital multi-value image information. As a result, the plurality of pieces of image information read by the image reading unit 4 are converted into multi-valued image information of multiple gradations (here, a total of 256 gradations from 0 gradation to 255 gradations) to be multi-valued. can do.
[0048]
As shown in FIGS. 3 and 4, the AD conversion unit 4 a also includes a connection unit 41 that can be connected to the image processing apparatus A and a connection unit 42 that can be connected to a computer. Thus, when the connection unit 41 is connected to the image processing apparatus A, the read image information can be sent to the image processing apparatus A (see FIG. 3), and the connection unit 42 is connected to the computer 200. When it is done, the read image information can be sent to the computer 200 (see FIG. 4).
[0049]
The image processing apparatus A can perform the following image processing on the image information read by the image reading apparatus 2a.
[0050]
As shown in FIG. 3, the image processing apparatus A includes an image area separating means A3 including a storage means M, a first correction processing means A1, a second correction processing means A2, and an image area identifying image information forming means A3 ′. A halftone image multi-value image information forming means A4, a non-halftone image multi-value image information forming means A5, a multi-value image processing means A6, and a binarization processing means A7.
[0051]
The storage unit M stores multi-value image information input from the image reading device 2a, data in the middle of arithmetic processing of multi-value image information, predetermined data related to an LUT (look-up table), and the like. Therefore, it is configured by a RAM, a ROM, and the like.
[0052]
Prior to the execution of the image area identifying image information forming means A3 ′, the first correction processing means A1 corrects each multivalued image information input from the image reading apparatus 2a. Here, the apparatus 2a The conventional well-known “gamma correction process” for gamma correction in an adjustable manner for each multi-valued image information input from, and the average of the gradations of neighboring pixels in the pixel area centered on the target pixel (here, 3 rows × 3 And a conventionally well-known “blurring process” that takes an average of the gradations in the vicinity of 8 in the pixel region of the column.
[0053]
More specifically, in the gamma correction process, a plurality of types of output gradations (output luminances) that can be converted into different types of output gradations (output luminances) with respect to the input gradations (input luminances) of 256-level multi-value image information here. An allocation table is stored in the storage means M in advance, and the gradation (brightness) of each multi-value image information is referred to an allocation table selected by an operator or the like among the plurality of types of allocation tables as will be described later. Thus, gradation (luminance) conversion is performed. For example, if the 64th of the table is set to 120 in advance, the output gradation of the pixel whose input gradation is 64 is replaced with 120 by referring to the table (see FIG. 5A).
[0054]
Further, in the blurring process, assuming that the two-dimensional coordinates of the multi-value image information is (x, y), here, in the pixel region of 3 rows × 3 columns centered on the pixel of interest (x, y), there are 8 neighboring floors. Key average [{f (x-1, y-1) + f (x, y-1) + f (x + 1, y-1) + f (x-1, y) + f (x + 1, y) + f (x-1) , Y + 1) + f (x, y + 1) + f (x + 1, y + 1)} / 8] (see the shaded area in FIG. 5B), and this is performed on the entire multi-value image information.
[0055]
Prior to the execution of the half-tone image multi-value image information forming means A4 and the non-half-tone image multi-value image information forming means A5, the second correction processing means A2 receives each multi-value image input from the image reading device 2a. In this example, each multi-value image information input from the apparatus 2a is processed as a base density of the original image for a predetermined upper limit gradation and a predetermined lower limit gradation. / Shadow (HL / SD) processing "and conventionally known" gamma correction processing "for performing gamma correction in an adjustable manner.
[0056]
More specifically, in the HL / SD processing, here, if the gradation of a plurality of multi-value image information is In,
Out = (In-SD) × 255 / (HL-SD)
Here, the default values are 240 for HL and 10 for SD.
Each of the obtained Out values is used as a gradation of a plurality of pieces of multi-value image information. Thus, a predetermined upper limit gradation (240 here) and a predetermined lower limit gradation (here 10) or less are processed as the base density of the original image. The gamma correction processing is the same as that of the first correction processing means A1, and description thereof is omitted here.
[0057]
The image area separating unit A3 includes a first image area (photo halftone image area) having continuous gradation among the plurality of multi-value image information areas input from the image reading device 2a and the first image area. A second image area (character image area) other than one image area is determined and the first and second image areas (photo halftone image and character image areas) are separated; The gradation of the multi-value image information is substituted into a predetermined calculation formula for obtaining the variance value to obtain the variance value. If the obtained variance value is less than the predetermined value, the multi-value image information area While determining as one image area (photo halftone image area), if it is equal to or greater than the predetermined value, the multi-value image information area is determined as the second image area (character image area). The second image area (photo halftone image area and character image area) is separated.
[0058]
More specifically, the two-dimensional coordinates of a plurality of multivalued image information input from the image reading device 2a are set to (x, y), and the gradations of the plurality of multivalued image information (x, y) are set to Pixel (x, y). ), A predetermined calculation formula Bunsan (x, y) for obtaining the dispersion value in the image area separation unit A3 is:
[Equation 5]

It is.
[0059]
Then, the image area separating means A3 determines the area of the multi-value image information (x, y) as the first image area (x, y) if the variance value according to the calculation formula Bunsan (x, y) is less than the predetermined value. On the other hand, if it is greater than or equal to the predetermined value, the region of the multi-value image information (x, y) is determined as the second image region (character image region) and the first And a second image area (photo halftone image area and text image area). Here, the values of m and n in the above formula are both 5 and the predetermined value is 1150.
[0060]
The image area separating means A3 is also a plurality of image area identifying image information corresponding to the plurality of multi-value image information input from the image reading device 2a, and each corresponding multi-value image information is stored in the first image area ( A first identification code (128 in this case) that can be identified from the first image area (photo halftone image area) PA when it is determined that it is in the area (photo halftone image area) PA (see FIG. 6A). And a second identification code (here, it can be identified from the second image area (character image area) CA when it is determined that it is in the second image area (character image area) CA (see FIG. 6A)). Image area identifying image information forming means A3 ′ for previously forming a plurality of image area identifying image information D (see FIG. 6B).
[0061]
The half-tone image multi-value image information forming means A4 is a plurality of half-tone image (photo halftone image) multi-value image information corresponding to the plurality of multi-value image information input from the image reading device 2a. Each of the corresponding multi-value image information is subjected to halftone image (photo halftone image) processing to obtain a plurality of halftone image (photo halftone image) multivalue image information B (FIG. 7A). )) Is previously formed.
[0062]
In the halftone image multi-value image information forming means A4, the halftone image (photo halftone image) processing is performed in the pixel area centered on the target pixel to remove the moiré. Conventionally known “blur processing” that takes an average (here, an average of 8 neighboring gradations in a pixel area of 3 rows × 3 columns) and an average of gradations in a pixel area centered on the target pixel (here 5 rows) “Moire removal process” consisting of “averaging process” that takes the average of the gradation in the pixel area selected by the operator or the like among the pixel areas of × 5 columns, 7 rows × 7 columns, and 9 rows × 9 columns. And “dither processing” by the conventionally known “dither method” for expressing halftones. The “moire removal process” is performed before the “dither process”.
[0063]
More specifically, in the moire removal process, the blurring process is the same as that of the first correction processing unit A1, and the description thereof is omitted here. In the averaging process, if the two-dimensional coordinates of the multi-valued image information is (x, y), 5 rows × 5 columns, 7 rows × 7 columns, 9 centered on the target pixel (x, y). As will be described later, an average of gradations in a pixel region selected by an operator, for example, a pixel region of 5 rows × 5 columns [{f (x−2, y− 2) + f (x-1, y-2) + f (x, y-2) +... + F (x, y) +... + F (x, y + 2) + f (x + 1, y + 2) + f (x + 2, y + 2)} / 25] (see the shaded area in FIG. 8A), and this is performed on the entire multi-value image information.
[0064]
In the dither processing, here, 4 × 4 = 16 threshold values are set in a matrix of 4 rows × 4 columns, and each pixel level is set in a pixel region of 4 rows × 4 columns of the multi-value image information. A matrix threshold value is applied to the key, and if each pixel is larger than the threshold value, it is replaced with 255, and if it is smaller, it is replaced with 0, and this is performed for the entire multi-value image information (see FIG. 8B).
[0065]
The non-halftone image multi-value image information forming means A5 is a plurality of non-half-tone image (character image) multi-value image information corresponding to the plurality of multi-value image information input from the image reading device 2a. Each of the corresponding multi-value image information is subjected to a non-halftone image (character image) process to obtain a plurality of non-halftone image (character image) multi-value image information C (see FIG. 7B). ) Is formed in advance.
[0066]
In the non-halftone image multi-value image information forming unit A5, the non-halftone image (character image) processing is performed here for a plurality of multi-value image information input from the image reading device 2a. An error diffusion process by an error diffusion method is performed.
[0067]
More specifically, in the error diffusion process, if the two-dimensional coordinate is (x, y) for each multivalued image information, here, it corresponds to the gradation f (x, y) of the target pixel (x, y). Ex, y = f (x, y) −g (), where g (x, y) is a gradation value (255 or 0) after two gradation conversion by the fixed threshold method with gradation 127 as a boundary. As shown in the following equation, the error generated in x, y) is distributed to the right of the target pixel 3/8, to the bottom 3/8, and to the bottom right 1/4, and this processing is performed on each pixel. .
fnew (x, y) = g (x, y)
fnew (x + 1, y) = f (x + 1, y) + (3/8) × Ex, y
fnew (x, y + 1) = f (x, y + 1) + (3/8) × Ex, y
fnew (x + 1, y + 1) = f (x + 1, y + 1) + (1/4) × Ex, y
[0068]
The multi-value image processing means A6 uses the first and second image areas (photo halftone image and character image areas) by the image area separation means A3 for a plurality of pieces of multi-value image information input from the image reading device 2a. Based on the determination result, image processing corresponding to each image area is performed, and a plurality of image area identifying images formed in advance by the image area identifying image information forming means A3 ′ of the image area separating means A3. Based on the identification results of the first and second image areas (photo halftone image area and character image area) based on the information D, image processing corresponding to each image area is performed.
[0069]
More specifically, in the multi-value image processing means A6, each multi-value image information input from the image reading device 2a is formed in advance by the image area identifying image information forming means A3 ′ of the image area separating means A3. When the corresponding image area identifying image information D (see FIG. 6B) is identified as the first image area (photo halftone image area) PA, halftone image multi-value image information forming means A4 The second image area (character image area) CA is formed as the corresponding halftone image (photo halftone image) multi-value image information B (see FIG. 7A), which is formed in advance. Is determined as the corresponding non-halftone image (character image) multi-value image information C previously formed by the non-halftone image multi-value image information forming means A5 (FIG. 7B). reference).
[0070]
The binarization processing means A7 converts all of the plurality of multivalued image information processed by the multivalued image processing means A6 into binary image information and binarizes it.
[0071]
On the other hand, the computer 200 shown in FIG. 4 includes a central processing unit 210, a main memory 220, a file device 230, an input device 240, and an output device 250. The input device 240 includes a reading unit 241 that can read program information from a computer-readable recording medium 300 that records the scanner driver Q ′ including the image processing program Q, and an input unit 242 such as a mouse or a keyboard. It has. The file device 230 can record program information read from the recording medium 300. The main memory 220 can read program information recorded in the file device 230 and has a program work area. Here, the output device 250 includes an image display device 251 such as a display and a printing device 252 such as a printer, and can display and / or print information image-processed by the image processing program Q. Further, the central processing unit 210 can control the entire scanner 1 based on the scanner driver Q ′ read into the main memory 220.
[0072]
The image processing program Q converts the computer 200 into image area separation means A3 including the above-described means, that is, first correction processing means A1, second correction processing means A2, and image area identifying image information forming means A3 ′. The multi-value image information forming unit A4 for halftone images, the multi-value image information forming unit A5 for non-halftone images, the multi-value image processing unit A6, and the binarization processing unit A7. Details of these means are as described above, and a description thereof is omitted here.
[0073]
In the image processing apparatus A and the image processing program Q described above, the image processing adjusted by the adjustment selected by the user among the following adjustments (a) and (b) is executed.
(A) Selection of gamma correction adjustment
(B) Selection of moire removal adjustment
[0074]
FIG. 9 shows a selection screen displayed on the display screen of the operation unit S (see FIG. 3) by the image processing apparatus A, and FIG. 10 shows a selection displayed on the display 251 (see FIG. 4) by the image processing program Q. Show the screen.
[0075]
On the selection screen by the image processing apparatus A, as shown in FIG. 9, when selecting gamma correction adjustment, “gamma correction adjustment” in (a) is selected by key operation (see FIG. 9A), and By inputting the gamma correction adjustment levels “A”, “B”, and “C” on the gamma correction adjustment selection screen a (see FIG. 9B) displayed on the screen, the type of the allocation table is changed. be able to. When the moire removal adjustment is selected, “moire removal adjustment” in (b) is selected by key operation (see FIG. 9A), and the next moire removal adjustment selection screen b (FIG. 9C) is displayed. The pixel area in the above-mentioned “averaging process” is set to 5 rows × 5 columns: “weak”, 7 rows × 7 by inputting the moire removal adjustment levels “weak”, “medium”, and “strong” Column: “Medium”, 9 rows × 9 columns: “Strong” can be changed.
[0076]
In the adjustment screen by the image processing program Q, as shown in FIG. 10, when gamma correction adjustment is selected, the gamma correction adjustment levels “A”, “B”, and “C” on the gamma correction adjustment selection screen a are operated by the mouse. The type of the allocation table can be changed by clicking with the pointer. When selecting moire removal adjustment, click the moire removal adjustment levels “weak”, “medium”, and “strong” on the moire removal adjustment selection screen b, and the five pixel areas in the above “averaging process” are displayed. × 5 columns: “weak”, 7 rows × 7 columns: “medium”, 9 rows × 9 columns: “strong”.
[0077]
FIG. 11 is a flowchart showing the flow of processing by the image processing apparatus A and the image processing program Q. 12A is a flowchart showing the flow of the moire removal processing subroutine in the flowchart shown in FIG. 11, and FIG. 12B is a flowchart showing the flow of the image area separation processing subroutine in the flowchart shown in FIG. It is.
[0078]
In the image processing apparatus A and the image processing program Q, as shown in FIG. 11, a plurality of multi-value image information input from the image reading apparatus 2a is first subjected to a first correction process by the first correction processing means A1. At the same time (step Sa), the second correction processing by the second correction processing means A2 is performed (step Sb).
[0079]
That is, in the first correction processing Sa, the gradation conversion is performed by referring to the allocation table selected on the selection screen a for the gradation of each multi-value image information input from the image reading device 2a. Then, “gamma correction” is performed (step S 1), and next, the gradation of 8 neighborhoods is averaged in the 3 × 3 pixel area centered on the pixel of interest, and “blurring processing” is performed ( Step S2).
[0080]
Further, in the second correction process Sb, with regard to the gradation of each multi-value image information input from the image reading device 2a, first, an upper limit gradation (240 here) and a lower limit gradation (here 10) or less are originals. The image is processed as the base density of the image to perform “highlight / shadow processing” (step S 3). Next, gradation conversion is performed by referring to the allocation table selected on the selection screen a, and “gamma” is selected. Correction "is made (step S4).
[0081]
Next, each multivalued image information subjected to the first correction processing Sa is subjected to image area separation processing by the image area separation means A3 including the image area identification image information forming means A3 '(step S5). The multi-value image information subjected to the second correction process Sb is subjected to half-tone image multi-value image information forming processing by the multi-value image information forming means A4 for half-tone images (step Sc) and non-intermediate. Multi-value image information forming processing for non-halftone images is performed by the multi-value image information forming means A5 for toned images (step Sd).
[0082]
That is, in the image area separation process S5, as shown in FIG. 12A, for each multivalued image information subjected to the first correction process Sa, the calculation formula Bunsan (x , Y) to obtain the variance value V (step S51), and if the obtained variance value is less than a predetermined value (here, 1150), the multi-value image information area is designated as the first image area ( It is determined that it is a photographic halftone image area (PA) (step S52), and a first identification code (128 here) that can be distinguished from the first image area (photo halftone image area) PA is attached (step S53). If the obtained dispersion value is equal to or larger than a predetermined value (here, 1150), the multi-value image information area is determined as the second image area (character image area) CA (step S52), and the second A second area that can be identified from the image area (character image area) CA. (In this case 0) another code that the subjecting (step S54), a plurality of image region discrimination image information D (see FIG. 6 (B)) is formed.
[0083]
Further, in the half-tone image multi-value image information forming process Sc shown in FIG. 11, the moire removal process is first performed on each multi-value image information subjected to the second correction process Sb (step S6), and then the dither process is performed. This is done (step S7).
[0084]
That is, in the moire removal process S6, as shown in FIG. 12B, for each multivalued image information subjected to the second correction process Sb, there are 8 neighboring pixel regions in a 3 × 3 pixel area centered on the target pixel. After the gradation is averaged and the “blurring process” is performed (step S61), the pixel region of 5 rows × 5 columns, 7 rows × 7 columns, and 9 rows × 9 columns centered on the target pixel is selected. Then, the gradation in the pixel area selected on the selection screen b is averaged, and the “averaging process” is performed (step S62).
[0085]
Further, in the dither processing S7 shown in FIG. 11, for each multi-value image information subjected to the moire removal processing S6, a matrix threshold is applied to the gradation of each pixel in the pixel region of 4 rows × 4 columns, and each pixel Is larger than the threshold, it is replaced with 255, and smaller is replaced with 0.
[0086]
On the other hand, in the multi-value image information forming process Sd for non-halftone images, for each multi-value image information subjected to the second correction process Sb, Ex, y = f (x, y) −g (x, y) of the target pixel. The error generated in step 3 is distributed 3/8 to the right of the pixel of interest, 1/4 to the lower right, and 3/8 to the lower, and “error diffusion processing” is performed (step S8).
[0087]
Further, after multi-value image processing is performed by the multi-value image processing means A6 (step S9), binarization processing by the binarization processing means A7 is performed (step S10).
[0088]
That is, in the multi-value image processing S9, the corresponding image area identifying image information D (FIG. 6 (FIG. 6)) formed in advance in the image area separation processing S5 for a plurality of multi-value image information input from the image reading device 2a. When the identification code is identified as the first image area (area of the photographic halftone image) having an identification code of “128” by (see B)), the halftone image multi-value image information forming process Sc was previously formed. The corresponding halftone image (photo halftone image) multi-value image information B (see FIG. 7A) is the second image area (character image area) with the identification code “0”. Corresponding non-halftone image multi-value image information forming process Sd, the corresponding non-half-tone image (character image) multi-value image information C (see FIG. 7B). It is said.
[0089]
In the binarization process S10, binarization is performed by a fixed threshold method using a plurality of multi-value image information formed in the multi-value image process S9 as a threshold value 127, and converted into a plurality of binary image information.
[0090]
According to the image processing apparatus A and the image processing program Q described above, continuous tone characteristics among a plurality of multi-value image information areas input from the image reading apparatus 2a and processed in the first correction process Sa. A first image area (photo halftone image area) and a second image area (character image area) other than the first image area, and the first and second image areas (photographs) In order to separate the halftone image and the character image area), the gradation of the plurality of multi-value image information is substituted into a predetermined calculation formula for obtaining a dispersion value, and a dispersion value is obtained. If the value is less than the predetermined value, the multi-value image information area is determined as the first image area (photo halftone image area). It is determined that there are two image areas (character image areas), and the first and second image areas (photo halftone image and character image Each of the plurality of multi-valued image information separated into multi-values based on the determination results of the first and second image regions (photo halftone image and character image regions). Therefore, the processing for image area separation can be simplified, the time required for the entire image processing can be shortened, and the circuit scale and software constituting the hardware for image processing can be configured. The scale of the program to be performed can be reduced and the cost can be reduced accordingly.
[0091]
Further, in the non-halftone image (character image) process Sd, an error by a conventionally known error diffusion method is applied to a plurality of multi-value image information input from the image reading device 2a and processed in the second correction process Sb. Since the diffusion process is performed, the second image area (character image area) of the plurality of multi-value image information areas, for example, a relatively small character image area or a halftone character image with a slightly light density Even in this region, when converted into binary image information and binarized, it is possible to obtain a character image with good halftone reproducibility and suppressed emphasis of the contour portion due to noise.
[0092]
The image processing apparatus A and the image processing program Q preliminarily form multi-value image information B for halftone images (for photographic halftone images) and multi-value image information C for non-halftone images (for character images). Therefore, each of the multivalued image information input from the image reading device 2a may be used as halftone image (photo halftone image) multivalued image information after the original image is read. Multi-value image information for non-halftone images (for character images) can also be used.
[0093]
【The invention's effect】
As described above, according to the present invention, the processing for image area separation can be simplified, the time required for the entire image processing can be shortened, and the circuit scale and software constituting the hardware for image processing can be reduced. Therefore, it is possible to provide an image processing method and an image processing apparatus that can reduce the scale of the program that constitutes the above, and that can realize cost reduction.
[0094]
Further, according to the present invention, the processing for image area separation can be simplified, the time required for the entire image processing can be shortened, the program scale constituting the software for image processing can be reduced, and the cost can be reduced. An image processing program capable of realizing down and a computer-readable recording medium on which the image processing program is recorded can be provided.
[Brief description of the drawings]
FIG. 1 is a schematic perspective view of a scanner provided with an example of an image processing apparatus that performs an image processing method according to the present invention.
2 is a schematic side view of an image reading apparatus in the scanner shown in FIG.
3 is a schematic block diagram centering on an image processing apparatus when the scanner shown in FIG. 1 is mounted on a digital copying machine;
4 is a schematic block diagram centering on the computer when the image reading apparatus of the scanner shown in FIG. 1 is connected to the computer. FIG.
FIG. 5A is a diagram for explaining “gamma correction processing”, and FIG. 5B is a diagram for explaining “blurring processing”;
6A is a schematic diagram illustrating an example of an original image, and FIG. 6B is a schematic diagram illustrating an example of image area identifying image information corresponding to the original image illustrated in FIG.
7A is a schematic diagram showing an example of multi-value image information for halftone images (for photographic halftone images) corresponding to the original image shown in FIG. 6A as an output image. FIG. 7B is a schematic diagram showing an example of non-halftone image (character image) multi-value image information corresponding to the document image shown in FIG.
FIG. 8A is a diagram for explaining “blur processing”, and FIG. 8B is a diagram for explaining “dither processing”;
9A and 9B are diagrams showing a selection screen displayed on the display screen of the operation unit by the image processing apparatus. FIG. 9A shows an adjustment selection menu, FIG. 9B shows a gamma correction adjustment selection screen, and FIG. C) shows a moire removal adjustment selection screen.
FIG. 10 is a diagram showing a selection screen displayed on the display by the image processing program.
FIG. 11 is a flowchart illustrating a flow of processing by the image processing apparatus and the image processing program.
12A is a flowchart showing the flow of a moiré removal processing subroutine in the flowchart shown in FIG. 11, and FIG. 12B is a flowchart showing the flow of the image area separation processing subroutine in the flowchart shown in FIG. It is.
FIG. 13 is a diagram illustrating an example of a region where error diffusion processing is performed in a second image region (character image region).
FIG. 14 is a diagram conceptually showing the relationship between the type of document image and gradation.
[Explanation of symbols]
A1: First correction processing means A2: Second correction processing means
A3: Image area separating means A3 ': Image area identifying image information forming means
A4: Multi-value image information forming means for halftone image
A5: Non-halftone image multi-value image information forming means A6: Multi-value image processing means
A7: Binarization processing means

Claims (18)

A document image is divided into a plurality of pixels to read a plurality of pieces of image information, and the read pieces of the plurality of pieces of image information are all converted into a plurality of gradations of multi-value image information and inputted from an image reading device that performs multi-value conversion. Of the plurality of multi-value image information areas, a first image area having continuous gradation and a second image area other than the first image area are determined to determine the first and second image areas. An image area separation step for separating the image areas of the plurality of multi-value image information, substituting the gradations of the plurality of pieces of multi-value image information into predetermined calculation formulas for obtaining a dispersion value, obtaining a dispersion value, and obtaining the dispersion value If it is less than a predetermined value, the multi-valued image information area is determined as the first image area, and if it is equal to or greater than the predetermined value, the multi-valued image information area is determined as the second image area. An image area separating step for separating the first and second image areas;
The plurality of multi-value image information input from the image reading device is subjected to image processing corresponding to each image area based on the determination result of the first and second image areas by the image area separation step. A multi-value image processing step;
An image processing method comprising: a binarization processing step of converting all of the plurality of multi-value image information subjected to image processing in the multi-value image processing step into binary image information to binarize the information. . The two-dimensional coordinates of the plurality of multi-value image information input from the image reading device is (x, y), and the gradation of the plurality of multi-value image information (x, y) is Pixel (x, y). Then, a predetermined calculation formula Bunsan (x, y) for obtaining the dispersion value in the image area separation step is

And
In the image area separation step, if the variance value according to the calculation formula Bunsan (x, y) is less than the predetermined value, the area of the multi-value image information (x, y) is determined as the first image area. On the other hand, if the value is equal to or greater than the predetermined value, the region of the multi-value image information (x, y) is determined as the second image region, and the first and second image regions are separated. Image processing method. The gradation of the multi-value image information input from the image reading device is 256 gradations, the values of m and n in the formula are both 5, and the predetermined value in the image area separation step is 1150. The image processing method according to claim 2. The image area separation step includes a plurality of image area identification image information corresponding to the plurality of multi-value image information input from the image reading device, wherein each corresponding multi-value image information is the first image area. A plurality of identification codes each provided with an identification code so that it can be distinguished from the first image area when it is determined to be in the second image area, and so that it can be distinguished from the second image area when it is determined that the second image area is determined. An image area identifying image information forming step for forming image area identifying image information in advance,
In the multi-value image processing step, the plurality of images formed in advance in the image area identifying image information forming step of the image area separation step with respect to the plurality of multi-value image information input from the image reading device. The image processing method according to any one of claims 1 to 3, wherein image processing corresponding to each image area is performed based on the identification result of the first and second image areas based on the image information for area identification. The image processing method according to claim 4, further comprising a correction processing step of correcting each of the multi-value image information input from the image reading device prior to the image area identification image information forming step. The image processing method according to claim 5, wherein the correction process in the correction process step includes a gamma correction process in which the multi-value image information input from the image reading device is gamma-corrected so as to be adjustable. A plurality of half-tone image multi-value image information corresponding to the plurality of multi-value image information input from the image reading device, each of the corresponding multi-value image information being subjected to half-tone image processing. A half-tone image multi-value image information forming step for previously forming a plurality of half-tone image multi-value image information;
A plurality of non-halftone image multi-value image information corresponding to the plurality of multi-value image information input from the image reading device, wherein each corresponding multi-value image information is processed for each non-half-tone image And a non-halftone image multi-value image information forming step of previously forming a plurality of non-halftone image multi-value image information,
In the multi-value image processing step, for each multi-value image information input from the image reading device, a corresponding image area formed in advance in the image area identifying image information forming step of the image area separation step. When it is identified as the first image area by the image information for identification, the multi-value image information for halftone image formed in advance in the multi-value image information formation step for halftone image, The multi-value image information for the non-halftone image that has been formed in advance in the multi-value image information forming step for the non-halftone image when the second image area is identified. An image processing method according to any one of the above. The non-halftone image processing in the non-halftone image multi-value image information forming step performs error diffusion processing by an error diffusion method for the plurality of multi-value image information input from the image reading device. 8. The image processing method according to 7. A document image is divided into a plurality of pixels to read a plurality of pieces of image information, and the read pieces of the plurality of pieces of image information are all converted into a plurality of gradations of multi-value image information and inputted from an image reading device that performs multi-value conversion. Of the plurality of multi-value image information areas, a first image area having continuous gradation and a second image area other than the first image area are determined to determine the first and second image areas. An image area separating means for separating the image areas of the plurality of multi-valued image information, substituting the gradations of the plurality of multi-value image information into a predetermined calculation formula for obtaining the dispersion value, obtaining the dispersion value, and obtaining the dispersion value If it is less than a predetermined value, the multi-valued image information area is determined as the first image area, and if it is equal to or greater than the predetermined value, the multi-valued image information area is determined as the second image area. Image area separating means for separating the first and second image areas;
The plurality of multi-value image information input from the image reading device is subjected to image processing corresponding to each image area based on the determination result of the first and second image areas by the image area separation unit. Multi-value image processing means;
An image processing apparatus comprising: binarization processing means for converting all of the plurality of multivalued image information processed by the multilevel image processing means into binary image information and binarizing the information. . The two-dimensional coordinates of the plurality of multi-value image information input from the image reading device is (x, y), and the gradation of the plurality of multi-value image information (x, y) is Pixel (x, y). Then, a predetermined calculation formula Bunsan (x, y) for obtaining the dispersion value in the image area separating means is

And
The image area separation means determines that the area of the multi-value image information (x, y) is the first image area if the variance value according to the calculation formula Bunsan (x, y) is less than the predetermined value. On the other hand, if it is not less than the predetermined value, the region of the multi-valued image information (x, y) is determined as the second image region, and the first and second image regions are separated. Image processing device. The gradation of the multi-value image information input from the image reading device is 256 gradations, the values of m and n in the equation are both 5, and the predetermined value in the image area separation unit is 1150. The image processing apparatus according to claim 10. The image area separating means is a plurality of image area identifying image information corresponding to the plurality of multi-value image information input from the image reading device, and each corresponding multi-value image information is stored in the first image area. A plurality of identification codes each provided with an identification code so that it can be distinguished from the first image area when it is determined to be in the second image area, and so that it can be distinguished from the second image area when it is determined that the second image area is determined. Image area identifying image information forming means for previously forming image area identifying image information is provided,
The multi-value image processing means is configured to obtain the plurality of images formed in advance by the image area identifying image information forming means of the image area separating means for the plurality of multi-value image information input from the image reading device. The image processing apparatus according to claim 9, wherein image processing corresponding to each image area is performed based on the identification result of the first and second image areas based on area identification image information. The image processing apparatus according to claim 12, further comprising a correction processing unit that corrects each of the multi-valued image information input from the image reading device prior to execution of the image area identifying image information forming unit. The image processing apparatus according to claim 13, wherein the correction process in the correction processing unit is a process including a gamma correction process that adjustably gamma-corrects each of the multi-value image information input from the image reading apparatus. A plurality of half-tone image multi-value image information corresponding to the plurality of multi-value image information input from the image reading device, each of the corresponding multi-value image information being subjected to half-tone image processing. A half-tone image multi-value image information forming means for previously forming a plurality of half-tone image multi-value image information;
A plurality of non-halftone image multi-value image information corresponding to the plurality of multi-value image information input from the image reading device, wherein each corresponding multi-value image information is processed for each non-half-tone image And non-halftone image multi-value image information forming means for previously forming a plurality of non-half-tone image multi-value image information,
The multi-value image processing means corresponds to each of the multi-value image information input from the image reading device, and the corresponding image area previously formed by the image area identifying image information forming means of the image area separating means. On the other hand, when the image area for identification is identified as the first image area, the multi-value image information for halftone image formed in advance by the half-tone image multi-value image information forming unit is used. The multi-value image information for the non-halftone image that is formed in advance by the multi-value image information forming unit for the non-halftone image when it is identified as the second image area. An image processing apparatus according to any one of the above. The non-halftone image processing in the non-halftone image multi-value image information forming unit performs error diffusion processing by an error diffusion method on the plurality of multi-value image information input from the image reading device. 15. The image processing apparatus according to 15. An image processing program for causing a computer connected to the image reading apparatus to execute each step of the image processing method according to claim 1. A computer-readable recording medium on which the image processing program according to claim 17 is recorded.

Images