JP4101526B2 - Image reading method, image reading apparatus, image reading system, image restoration method, and image reading program - Google Patents

Description

従来例（図１１）の場合には，まず，本式によりシェーデイング補正を実施し，正規化したあと，データの転送を行っていたため、レベル方向のビット数を削減できない。本発明では、受信側でシェーデイング補正するようにしたので、転送データにセンサ出力の絶対レベルを要求されない。このため、ビット数を削減し、相対レベルに変換しても、問題なく、シェーデイング補正できる。
【００４５】
この例では、保持部２４は、センサの各画素の入力特性として、各画素１律１ビット削減を保持する。又、この入力特性値と、図５のセンサの入力特性を、情報処理装置４０に、予め保持する。
【００４６】
このように、センサ入力特性に応じて画素毎にレベル方向のビット数を削減した画像生成する画像データ生成部と，画像データ生成部で生成された各画素のビット数に応じてデータのパッキングを行う画像データパッキング部を備えているため、空間解像度情報の欠落なく、センサの特性に合わせて、データ圧縮できる。
【００４７】
［他のデータ圧縮方法］
図６は、本発明の他の実施の形態のデータ圧縮方法の説明図である。ここでは，ラインセンサの各画素毎に，明特性Ｗ（ｎ）と暗特性Ｂ（ｎ）から削減ビット数を設定する。
【００４８】
図６には、２画素分のデータを例にしてあり、センサの１画素目、２画素目の入力特性が、暗特性のレベル値Ｂ（１）＝２、Ｂ（２）＝０で、明特性のレベル値Ｗ（１）＝１６，Ｗ（２）＝１４であるとする。
【００４９】
１画素目の有効ビット数Ｉ（１）は、１６−２＝１４＝４ビット、２画素目の有効ビット数Ｉ（２）は、１４−０＝１４＝４ビットである。１画素のビット数Ｎを「８」とすると、１画素目、２画素目は、各々４ビットのデータで済む。この場合に、画素レベルを、暗特性値を「０」にしたレベルに変換する。即ち、１画素目は、「０００００１１１」＝７−Ｂ（１）（＝２）＝５に変換し、２値表現で、「０００００１０１」に変換する。同様に、２画素目もレベルシフトするが、この場合に、Ｂ（２）＝０であるから、画素データは、変わらない。
【００５０】
そして、画素データの下位４ビット（有効ビット）を取り出し、１画素目と２画素目との４ビットを組み合わせ、パッキングデータを作成する。これを情報処理装置に転送する。
【００５１】
情報処理装置では、受信したパッキングデータを、各画素の有効ビット数から８ビットの画素データに復元し、更に、各画素の明特性値と暗特性値とから、逆レベルシフトし、各画素の８ビットデータを復元する。
【００５２】
このように、各画素毎に、明特性、暗特性に応じて、有効ビット数を設定し、レベルシフトすることにより、大幅なデータ圧縮が可能となる。同様に、各画素毎に明特性値のみから有効ビット数を、暗特性値のみから有効ビット数を設定することもできる。
【００５３】
［他の実施の形態］
図７は、本発明の他の実施の形態のイメージスキャナのブロック図、図８は、その処理フロー図、図９は、その復元処理フロー図である。
【００５４】
図７において、図１で示したものと同一のものは、同一の記号で示してあり、図１との相違は、図１の圧縮モードに加え、非圧縮モードを付加したものである。
【００５５】
即ち、画像読取部（ＰＣカード）２は、画像信号処理部２０、読取制御部、装置インターフェイス（Ｉ／Ｆ）部２８から構成されている。画像信号処理部２０は、スキャナ本体１の画像読取部１０からのアナログ画像信号を、デジタル化するものであり、Ａ／Ｄ変換回路により構成される。
【００５６】
読取制御部は、全体の読み取り制御と、情報処理装置４へのデータ転送を行う。読取制御部は、レベル変換画像データ生成部２２と、センサ入力特性保持部２４と、パッキングデータ生成部２６とを有する。センサ入力特性保持部２４は、図５に示すように、ラインセンサの各画素の入力特性（明特性、暗特性）で決定されたセンサ入力特性を保持する。
【００５７】
レベル変換画像データ生成部２４は、入力特性に応じて、画像読取部１０のラインセンサの各画素毎に信号レベル方向のビット数を削減した画像データを生成する。パッキングデータ生成部２６は、画像データ生成部２４で生成された各画素のデータをパッキングして、装置インタフェイス部２８に出力する。装置インタフェイス部２８は、情報処理装置４０へ画像データを転送する。
【００５８】
読取制御部は、画像読取部１０のラインセンサの起動（駆動）信号を、読み取り期間中は、一定周期で発生し、ラインセンサを駆動する。ラインセンサからライン信号（例えば、２３７０画素）は、画像信号処理部２０でＡ／Ｄ変換され、処理済みのラインデータは、画像データ生成部２２内部のラインバッファに保存する。例えば、Ａ／Ｄ変換は、アナログ画像信号を、８ビット、２５６レベルのデジタル値に変換する。
【００５９】
画像データ生成部２２は、スキャナ本体１の移動量検出部１２からの移動パルスを検出した場合，ラインバッファに保存してあるラインデータのレベル変換を行う。即ち、保持部２４の保持特性に従い、センサの明暗特性に応じてレベル削減を行う。
【００６０】
更に、この圧縮処理のあり・なしを切り替える手段として，装置Ｉ／Ｆ部２８の転送速度を検出するデータ転送速度検出部３０と、転送速度評価部３２を設ける。転送速度評価部３２は，センサ１０の駆動周期から求まるデータ発生速度と，実際のデータ転送速度を比較し，データ転送速度が遅い場合には，レベル変換画像データ生成部２２に必要なレベル数（センサ入力保持部２４の値）を伝達する。
【００６１】
ここで，データ転送速度の検出方法は，インタフェイスの種別に応じて、あらかじめ決められた速度を出力してもよいし，装置Ｉ／Ｆ部２８から単位時間に送られたデータから算出してもよい。
【００６２】
レベル変換画像データ生成部２２及びパッキングデータ生成部２６は、評価部３２からレベル数指示がない時は、前述のレベル変換処理、パッキング処理を行わず、画像データ生成部２２のラインバッファのラインデータをそのまま装置Ｉ／Ｆ部２８に出力する。
【００６３】
逆に、レベル変換画像データ生成部２２及びパッキングデータ生成部２６は、評価部３２からレベル数指示がある時は、前述のように、入力特性からレベル変換処理、パッキング処理を行い、ラインデータを装置Ｉ／Ｆ部２８に出力する。
【００６４】
図８により、読取制御部の処理を説明する。
【００６５】
（Ｓ１０）先ず、処理開始にあたり、情報処理装置４０からスキャナ２の読取制御部に、読込みライン数Ｌ（例えば、４）が設定される。
【００６６】
（Ｓ１１)設定ライン数Ｌと実際の読込みライン数を比較し、読込みライン数が、設定ライン数Ｌに到達している場合には、処理を終了する。
【００６７】
（Ｓ１２）スキャナ本体１の移動量検出部１２からエンコーダパルスが到来したかを検出する。エンコーダパルスが到来していない場合には、ステップＳ１１に戻る。
【００６８】
（Ｓ１３）エンコーダパルスが到来している場合には、ラインバッファに、画像処理部２０を介するラインデータをラッチする。
【００６９】
（Ｓ１４）読込みライン数を、「１」インクリメントする。
【００７０】
（Ｓ１５）Ｉ／Ｆ部２８の転送速度（クロック数）を検出し、ラインセンサの駆動周期から求めたデータ発生速度Ｓと比較する。転送速度がデータ発生速度より小さくない場合には、レベル削減、パッキング処理の必要がないため、ラインセンサのデータをＩ／Ｆ部２８に出力し、ステップＳ１１に戻る。
【００７１】
（Ｓ１６）逆に、転送速度がデータ発生速度Ｓより小さい場合には、前述のレベル削減により、ビット数を少なくし、且つパッキング処理を行う。そして、このデータ及び圧縮フラグをＩ／Ｆ部２８に出力し、ステップＳ１１に戻る。
【００７２】
このようにして、１ライン分のバッファ容量で、空間解像度情報を欠落することなく、データ圧縮できる。しかも、センサの入力特性に応じて、レベル変換するため、レベル情報の欠落も防止できる。
【００７３】
更に、転送速度に応じて圧縮、非圧縮を制御するため、例えば、副走査速度に応じて、自動的に処理を変更できる。即ち、副走査速度が遅い場合には、圧縮しないため、全体の処理速度が向上し、副走査速度が速い場合には、圧縮するため、転送速度を向上できる。
【００７４】
図９は、本発明の一実施の形態の復元処理フロー図である。
【００７５】
（Ｓ２０）指定ライン数のライン画像データの読込み終了かを判定する。読込み終了なら復元処理を終了する。
【００７６】
（Ｓ２２）読込み終了でない場合には、転送されたラインデータを読み込む。
【００７７】
（Ｓ２４）図４又は図６に示したように、画像有効ビットIにより、ラインデータをアンパッキングし、逆レベル変換して、元のラインデータに復元する。そして、ステップＳ２０に戻る。
【００７８】
この復元処理は、データ圧縮された場合のものであり、図１及び図４、図６の実施の形態に適用できる。更に、図７、図８の実施の形態では、圧縮フラグを認識した場合に、この復元処理を実行する。
【００７９】
尚、イメージスキャナを手動操作するハンデイスキャナで説明したが、手動操作しないイメージスキャナにも適用できる。
【００８０】
（付記１）読取対象をラインセンサで走査して、読取対象の２次元画像データを生成する画像読取方法において、前記ラインセンサの各画素の出力ビット数を、前記ラインセンサの入力特性に応じて、削減するステップと、前記削減されたビット数の出力を、所定単位の転送データにパッキングするステップと、前記転送データを転送するステップとを有することを特徴とする画像読取方法。
【００８１】
（付記２）前記削減ステップは、前記ラインセンサの各画素の少なくとも明入力特性から、決定した削減ビット数により削減するステップからなることを特徴とする付記１の画像読取方法。
【００８２】
（付記３）前記削減ステップは、前記ラインセンサの各画素の明暗入力特性から各画素毎に決定した削減ビット数で削減するステップからなることを特徴とする付記１の画像読取方法。
【００８３】
（付記４）前記ラインセンサの駆動速度と前記データ転送速度を比較するステップと、前記駆動速度より前記データ転送速度が遅い場合には、前記ラインセンサの出力データを転送するステップと、前記駆動速度より前記データ転送速度が速い場合には、前記削減及びパッキングステップを実行するステップとを更に有することを特徴とする付記１の画像読取方法。
【００８４】
（付記５）読取対象をラインセンサで走査して、読取対象の２次元画像データを生成する画像読取装置において、前記ラインセンサの入力特性を保持する入力特性保持部と、前記ラインセンサの各画素の出力ビット数を前記保持された入力特性に応じて、削減し、且つ前記削減されたビット数の出力を、所定単位の転送データにパッキングして、転送する読取制御部とを有することを特徴とする画像読取装置。
【００８５】
（付記６）前記読取制御部は、前記ラインセンサの各画素の少なくとも明入力特性から、決定した削減ビット数により削減することを特徴とする付記５の画像読取装置。
【００８６】
（付記７）前記読取制御部は、前記ラインセンサの駆動速度と前記データ転送速度を比較し、前記駆動速度より前記データ転送速度が遅い場合には、前記ラインセンサの出力データを転送し、前記駆動速度より前記データ転送速度が速い場合には、前記削減及びパッキングを実行することを特徴とする付記５の画像読取装置。
【００８７】
（付記８）読取対象をラインセンサで走査して、前記読取対象の２次元画像データを生成する画像読取システムにおいて、前記ラインセンサの各画素の出力ビット数を前記ラインセンサの入力特性に応じて、削減し、且つ前記削減されたビット数の出力を、所定単位の転送データにパッキングして、転送するデータ圧縮部と、前記転送データを受け、各画素のデータを前記削減されたビット数に応じて、元のビット数のデータに復元するデータ復元部とを有することを特徴とする画像読取システム。
【００８８】
（付記９）読取対象をラインセンサで走査し、前記ラインセンサの各画素の出力ビット数を前記ラインセンサの入力特性に応じて、削減し、且つ前記削減されたビット数の出力を、所定単位の転送データにパッキングした圧縮データを復元する画像復元方法において、前記転送データを受け、各画素のデータに分割するステップと、前記削減されたビット数に応じて、元のビット数のデータに復元するステップとを有することを特徴とする画像復元方法。
【００８９】
（付記１０）ラインセンサで走査した読取対象の２次元画像データを生成するプログラムであって、前記ラインセンサの各画素の出力ビット数を、前記ラインセンサの入力特性に応じて、削減するプログラムデータと、前記削減されたビット数の出力を、所定単位の転送データにパッキングするプログラムデータとを有することを特徴とするプログラム。
【００９０】
【発明の効果】
以上説明してきたように、本発明は、センサ特性を利用した画素単位の画素レベル方向の情報圧縮により、空間解像度情報を保存することで、高解像度を維持しつつ、データ圧縮を実現できる。このため、２次元バーコード等の空間解像度情報の欠落を防止し、これらの解読率を向上できる読取が可能となる。又、ラインセンサの副走査速度に応じて、圧縮、非圧縮を自動切換えでき、高解像度を維持しつつ、処理速度を向上できる。
【図面の簡単な説明】
【図１】本発明の一実施の形態のイメージスキャナのブロック図である。
【図２】図１のイメージスキャナを使用したスキャナシステムの構成図である。
【図３】図１のイメージスキャナの外観図である。
【図４】図１のイメージスキャナの画像圧縮／復元方法の説明図である。
【図５】図４のためのラインセンサの入力特性図である。
【図６】本発明の他の実施の形態の画像圧縮／復元処理の説明図である。
【図７】本発明の他の実施の形態のイメージスキャナのブロック図である。
【図８】図７のデータ圧縮処理フロー図である。
【図９】図７のデータ復元処理フロー図である。
【図１０】従来技術の説明図である。
【図１１】従来技術のシェーデイング補正の説明図である。
【図１２】従来技術の画像圧縮処理の説明図である。
【図１３】従来技術の問題点の説明図である。
【符号の説明】
１ スキャナ本体
２ ＰＣカード（読取制御部）
１０ 画像読取部
１２ 移動量検出部
２０ 画像信号処理部
２２ レベル画像データ生成部
２４ センサ入力特性保持部
２６ パッキングデータ生成部
２８ 装置Ｉ／Ｆ部[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an image reading method, an image reading apparatus, an image reading system, an image restoration method, and a program for reading a two-dimensional image with a scanner and compressing image data, and more particularly to a portable information device such as an electronic notebook, a personal computer, The present invention relates to an image reading method, an image reading apparatus, an image reading system, an image restoration method, and a program for reading an image from a document by connecting or integrating with a personal word processor or the like.
[0002]
[Prior art]
An image scanner that reads an image of a document and converts it into electronic data is widely used for digitizing the image. Image scanners can be broadly classified into three types: flatbed type, sheet feed type, and handy type. The flatbed and sheet feed type automatically reads at a speed that the device can control by setting the document in the device. On the other hand, the handy scanner manually scans an original and reads an image.
[0003]
Such an image scanner is usually provided with a line sensor, and reads a two-dimensional image in the main scanning direction, reads the line sensor one line at a time, and moves the line sensor in the sub-scanning direction to read the two-dimensional image. Output image data line by line.
[0004]
On the other hand, a processing device such as a personal computer joins the output image data for each line and restores the image. The data transfer speed for transferring the image data from the image scanner to the processing device is determined by the interface. For this reason, when the image reading speed is faster than the data transfer speed, it is difficult to transfer all the image data, and the image data is lost.
[0005]
For example, in a handy scanner, since manual sub-scanning is performed, if the sub-scanning speed is increased, missing images may occur in units of lines, and an image restored by the processing apparatus may shrink in the sub-scanning direction. In order to prevent this shrinkage, it is necessary to provide an image data holding unit (buffer) according to the data transfer rate, or to provide an interface with a high data transfer rate that matches the maximum image reading rate. .
[0006]
However, in recent years, since the resolution of the reading unit has been demanded, the required buffer capacity increases, and conversely, replacing with an interface having a high transfer rate causes an increase in the cost of the apparatus.
[0007]
For this reason, a handy scanner or the like detects the amount of movement of the scanner and transfers the amount of movement together with the image data, and the processing device copies the image data according to the amount of movement and interpolates to restore the restored image. Techniques for preventing shrinkage have been proposed (for example, JP 2000-349984 A).
[0008]
Further, it has been proposed to transfer data while reducing the amount of image data. For example, the present inventors have proposed a data compression method suitable for a manually operated handy scanner in Japanese Patent Application No. 2001-91084 (filed on Mar. 27, 2001).
[0009]
That is, two-dimensional image data is generated according to the movement amount (sub-scanning speed) detected by the movement amount detection means (encoder) of the handy scanner and the data capacity of the image data holding unit for holding the image data of the handy scanner. Proposed to do.
[0010]
This proposal will be described with reference to FIGS. As shown in FIG. 10, the line sensor of the handy scanner is driven by a start signal of a predetermined period, and generates an image signal for each line. This image signal is corrected by the shading correction of FIG. On the other hand, as shown in FIGS. 10 and 12, the movement amount of the handy scanner is detected from the encoder pulse of the encoder of the handy scanner.
[0011]
Then, as shown in FIG. 10 and FIG. 12, the data generation side transfers the number of lines that need to be interpolated and a set of line data, and the receiving side interpolates the line data according to the number of lines that need to be interpolated. Then, the two-dimensional image data without shrinking is restored. That is, line data is thinned out according to the amount of movement to improve the data transfer speed.
[0012]
Since this scanner uses an optical line sensor, the output of each element of the line sensor varies. In order to correct this, shading correction is performed. As shown in FIG. 11, the bright input characteristic (Wn) and dark input characteristic (Bn) of each pixel of the line sensor are measured, and the input data (Xn) of each pixel of the line sensor is adjusted to this characteristic. Corrected and normalized corrected data is generated.
[0013]
[Problems to be solved by the invention]
In the above-mentioned conventional proposal, in order to improve the data transfer speed, the data is compressed and transferred by omitting the spatial resolution information by thinning out the line data. Such a method is effective when a human visually confirms the read image data.
[0014]
However, with the recent demand for enlargement of the reading target of an image scanner, reading of an image that does not allow missing of spatial resolution information is required. For example, when a two-dimensional code printed with information black and white as shown in FIG. 13 is to be read, because the line data is thinned out, the spatial resolution information is missing. In some cases, the processing device cannot correctly decode the black and white points two-dimensionally. For this reason, it is necessary to improve to prevent a decrease in the decoding rate of a two-dimensional code or the like.
[0015]
Accordingly, an object of the present invention is to provide an image reading method, an image reading apparatus, an image reading system, an image restoration method, and a program for compressing line data while preserving the spatial resolution information of the image.
[0016]
Another object of the present invention is to provide an image reading method, an image reading apparatus, an image reading system, an image restoration method, and a program for compressing line data according to the input characteristics of the image scanner.
[0017]
Furthermore, another object of the present invention is to provide an image reading method, an image reading apparatus, an image reading system, an image restoration method, and a program thereof for compressing pixel level direction information and storing spatial resolution information. .
[0018]
[Means for Solving the Problems]
  In order to achieve this object, the present invention uses a line sensor as a reading target.ViceIn the image reading method of scanning and generating the two-dimensional image data to be read, the step of reducing the number of output bits of each pixel of the line sensor according to at least the bright input characteristic of each pixel of the line sensor Packing the output of the reduced number of bits into transfer data of a predetermined unit, transferring the transfer data, and driving speed of the line sensorData generation in the sub-scan directionComparing the speed and the data transfer speed of the transfer data;OccurrenceThe data transfer rate isfastAnd transferring the output data of the line sensor,OccurrenceThe data transfer rate isslowIn some cases, the method includes the reduction step and the packing step.
[0019]
  Further, the image reading apparatus of the present invention uses a line sensor as a reading target.ViceIn an image reading apparatus that scans and generates two-dimensional image data to be read, an input characteristic holding unit that holds at least a bright input characteristic of each pixel of the line sensor, and an output bit number of each pixel of the line sensor A read control unit that packs the output of the reduced number of bits into transfer data in a predetermined unit and transfers it according to at least the bright input characteristics of each held pixel. The reading control unit is the driving speed of the line sensor.Data generation in the sub-scan directionCompare the speed and the data transfer speed of the transfer data,OccurrenceThe data transfer rate isfastIn the case, the output data of the line sensor is transferred, and theOccurrenceThe data transfer rate isslowIf so, perform the reduction and the packing.
[0020]
  Further, the image reading system of the present invention uses a line sensor as a reading target.ViceIn the image reading system that scans and generates the two-dimensional image data to be read, the number of output bits of each pixel of the line sensor is reduced according to at least the bright input characteristics of each pixel of the line sensor, The reduced number of bits output is packed into transfer data in a predetermined unit and transferred, and the transfer data is received, and the data of each pixel is reduced according to the bright input characteristics. A data restoration unit that restores data of the original number of bits according to the number of bits, and the data transfer unit is a driving speed of the line sensor.Data generation in the sub-scan directionCompare the speed and the data transfer speed of the transfer data,OccurrenceThe data transfer rate isfastIn the case, the output data of the line sensor is transferred, and theOccurrenceThe data transfer rate isslowIn some cases, the reduction and the packing are performed, and the data restoration unit performs the restoration for the reduced and packed transfer data.
[0021]
  In addition, the image restoration method of the present invention uses a line sensor as a reading target.ViceScanning to reduce the number of output bits of each pixel of the line sensor according to at least the bright input characteristics of each pixel of the line sensor, and to output the reduced number of bits to a predetermined unit of transfer data To packingAdd a compression flagAn image restoration method for restoring the compressed data, the step of receiving the transfer data, recognizing a compression flag of the transfer data, and when the transfer flag is recognized as being compressed by the compression flag, Dividing the transfer data into data of each pixel, and restoring the data to the original number of bits according to the reduced number of bits in accordance with the bright input characteristics.
[0022]
  The program of the present invention is a line sensor.ViceGenerate scanned 2D image data to be scannedImage readingA program for reducing the number of output bits of each pixel of the line sensor in accordance with at least the bright input characteristic of each pixel of the line sensor and outputting the reduced number of bits in a predetermined unit. Program data to be packed into the transfer data and the drive speed of the line sensorData generation in the sub-scan directionCompare the speed and the data transfer speed of the transfer data,OccurrenceThe data transfer rate isfastIn the case, output the output data of the line sensor,OccurrenceThe data transfer rate isslowIn some cases, the data includes program data for executing the reduction and the packing.
[0023]
The present invention reduces the number of bits in the output level direction for each pixel according to the input characteristics of each pixel of the line sensor that converts the image information into electrical data, and compresses the information in the pixel level direction. As a result, a read image stored with spatial resolution information can be obtained.
[0024]
Furthermore, in the present invention, it is preferable that the reduction step includes a step of reducing the number of bits determined based on at least a bright input characteristic of each pixel of the line sensor. Thereby, the number of bits can be reduced by effectively using the input characteristics of the line sensor.
[0025]
Furthermore, in the present invention, it is preferable that the reduction step further includes a step of reducing the number of bits determined for each pixel from the light / dark input characteristics of each pixel of the line sensor, thereby further reducing the number of bits. , The compression effect can be improved.
[0026]
In the present invention, preferably, the step of comparing the driving speed of the line sensor with the data transfer speed, and the step of transferring the output data of the line sensor when the data transfer speed is slower than the driving speed. And a step of executing the reduction and packing step when the data transfer speed is faster than the driving speed. Thus, compression and non-compression can be automatically switched according to the sub-scanning speed of the line sensor, and the processing speed can be improved.
[0027]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, the present invention will be described with reference to embodiments. However, the present invention is not limited thereto, and other forms are possible.
[0028]
[Scanner system]
1 is a block diagram of an image scanner according to an embodiment of the present invention, FIG. 2 is a block diagram of a scanner system according to an embodiment of the present invention, and FIG. 3 is a block diagram of the image scanner of FIG. 4 is an explanatory diagram of the data compression operation, and FIG. 5 is an explanatory diagram of the input characteristics of the sensor.
[0029]
As shown in FIG. 3, the image scanner optically reads an image of a document 5 (see FIG. 2) and converts it into an electrical signal. The scanner main body 1, a cable 16, a connector 18, a PC ( A personal computer) card (reading control unit) 2.
[0030]
The scanner body 1 includes an image reading unit 10, a movement amount detection unit 12, and an operation (start / interrupt) instruction unit 14. The image reading unit 10 includes a line sensor, a light source (LED or cold cathode tube), and a lens, and performs image reading in close contact with the reading medium 5. The line sensor is composed of a CCD having 2370 pixels, for example.
[0031]
In response to a start instruction from the operation instruction unit (button) 14, the light source is turned on during reading, and the line sensor is started to be read by an activation signal from the reading control unit. The line sensor converts the amount of reflected light into an electrical signal at each position (pixel) on the sensor and outputs it as an analog signal.
[0032]
The movement amount detection unit 12 includes a rotation roller and an encoder, and the rotation roller contacts a document to be read. As the scanner body 1 moves, the rotating roller rotates, and the encoder converts the amount of movement on the reading medium into a pulse signal and outputs it.
[0033]
The scanner body 1 is connected to the PC card 2 via a cable 16 and a connector 18. As illustrated in FIG. 2, the information processing apparatus 4 includes a data processing unit 40, a keyboard 44, and a display 42. For example, it is composed of a notebook computer or a desktop computer. The PC card 2 is inserted into the PC card slot of the processing unit 40, and the scanner is connected to the information processing apparatus 4.
[0034]
As shown in FIG. 1, the image reading unit (PC card) 2 includes an image signal processing unit 20, a reading control unit, and a device interface (I / F) unit 28. The image signal processing unit 20 digitizes an analog image signal from the image reading unit 10 of the scanner main body 1 and includes an A / D conversion circuit.
[0035]
The reading control unit performs overall reading control and data transfer to the information processing apparatus 4. The reading control unit includes a level conversion image data generation unit 22, a sensor input characteristic holding unit 24, and a packing data generation unit 26. As shown in FIG. 5, the sensor input characteristic holding unit 24 holds the sensor input characteristic determined by the input characteristic (bright characteristic, dark characteristic) of each pixel of the line sensor.
[0036]
The level-converted image data generation unit 24 generates image data in which the number of bits in the signal level direction is reduced for each pixel of the line sensor of the image reading unit 10 according to the input characteristics. The packing data generation unit 26 packs the data of each pixel generated by the image data generation unit 24 and outputs it to the device interface unit 28. The device interface unit 28 transfers image data to the information processing device 40.
[0037]
The reading control unit generates a line sensor activation (drive) signal of the image reading unit 10 at a constant period during the reading period, and drives the line sensor. A line signal (for example, 2370 pixels) from the line sensor is A / D converted by the image signal processing unit 20, and the processed line data is stored in a line buffer inside the image data generation unit 22. For example, the A / D conversion converts an analog image signal into an 8-bit, 256-level digital value.
[0038]
When the image data generation unit 22 detects a movement pulse from the movement amount detection unit 12 of the scanner body 1, the image data generation unit 22 performs level conversion of the line data stored in the line buffer. That is, the level is reduced according to the light / dark characteristics of the sensor according to the holding characteristics of the holding unit 24.
[0039]
The data compression operation by this level reduction will be described with reference to FIG. Referring to the sensor input characteristic of FIG. 5, the dark characteristic level value is “60”, and the bright characteristic level value is “160” to “240”. Therefore, the number of signal bits is “8” (256 levels). Even if the number of effective bits is set to 7 bits and the LSB is deleted, the level expression is not affected.
[0040]
For this reason, one bit of the effective bit of the transfer data is deleted according to the input characteristics of the sensor. In the example of FIG. 4, data for four lines is shown, and the number of pixel effective bits I (n) = 7 bits.
[0041]
In FIG. 4, the line data is data of 2400 pixels of each pixel N (= 8) bits. For the 8-bit data of each pixel, the LSB is uniformly deleted and converted to 7-bit data. That is, the intermediate image data is 7-bit 2400 pixels. The level-converted image data generation unit 22 shifts the data of each pixel by the pixel effective bit number I (n) and converts the level.
[0042]
The data of each pixel subjected to level conversion is converted into data of 2 pixels by 1 bit (= 8 bits) and 6 bits by the packing data generation unit 26 by the number of effective bits I (n) = 7 bits. Packing. The packed data is transferred from the device I / F unit 28 to the information processing device 40.
[0043]
The information processing apparatus 40 depacks the received data of one line by 7 bits, uniformly adds LSB (“1”), and restores the pixel data to M (= 8) bits. Although the packing data amount is schematically shown here, 7.25 / 8 data compression can be realized for the actual sensor characteristics of FIG. Further, the shading correction shown in FIG. That is, assuming that the input value Xn of each pixel of the line sensor, the bright input characteristic Wn, the dark input characteristic Bn, and the output value Yn normalized by the light / dark characteristic of the sensor, these relationships can be expressed by the following equations. The graph of FIG. 5 shows Wn and Bn in this equation.
[0044]

In the case of the conventional example (FIG. 11), the number of bits in the level direction cannot be reduced because shading correction is first performed according to this formula and data is transferred after normalization. In the present invention, since the shading correction is performed on the receiving side, the absolute level of the sensor output is not required for the transfer data. Therefore, even if the number of bits is reduced and converted to a relative level, the shading correction can be performed without any problem.
[0045]
In this example, the holding unit 24 holds a 1-bit reduction for each pixel as the input characteristic of each pixel of the sensor. In addition, the input characteristic value and the input characteristic of the sensor of FIG.
[0046]
In this way, the image data generation unit that generates an image in which the number of bits in the level direction is reduced for each pixel according to the sensor input characteristics, and the data packing according to the number of bits of each pixel generated by the image data generation unit. Since the image data packing unit is provided, data compression can be performed in accordance with the characteristics of the sensor without loss of spatial resolution information.
[0047]
[Other data compression methods]
FIG. 6 is an explanatory diagram of a data compression method according to another embodiment of the present invention. Here, the number of bits to be reduced is set for each pixel of the line sensor from the light characteristic W (n) and the dark characteristic B (n).
[0048]
In FIG. 6, data for two pixels is taken as an example, and the input characteristics of the first and second pixels of the sensor are dark characteristic level values B (1) = 2, B (2) = 0, It is assumed that the level value W (1) = 16 and W (2) = 14 of the bright characteristic.
[0049]
The effective bit number I (1) of the first pixel is 16-2 = 14 = 4 bits, and the effective bit number I (2) of the second pixel is 14-0 = 14 = 4 bits. If the bit number N of one pixel is “8”, the first pixel and the second pixel each need only 4 bits of data. In this case, the pixel level is converted to a level where the dark characteristic value is “0”. That is, the first pixel is converted to “00000111” = 7−B (1) (= 2) = 5 and converted to “00000101” in a binary expression. Similarly, the level of the second pixel is also shifted. In this case, since B (2) = 0, the pixel data does not change.
[0050]
Then, the lower 4 bits (effective bits) of the pixel data are extracted and the 4 bits of the first pixel and the second pixel are combined to create packing data. This is transferred to the information processing apparatus.
[0051]
In the information processing apparatus, the received packing data is restored to 8-bit pixel data from the number of effective bits of each pixel, and is further subjected to a reverse level shift from the light characteristic value and the dark characteristic value of each pixel, Restore 8-bit data.
[0052]
Thus, for each pixel, the effective bit number is set according to the bright characteristic and the dark characteristic, and the level shift is performed, so that significant data compression is possible. Similarly, for each pixel, the effective bit number can be set only from the bright characteristic value, and the effective bit number can be set only from the dark characteristic value.
[0053]
[Other embodiments]
FIG. 7 is a block diagram of an image scanner according to another embodiment of the present invention, FIG. 8 is a processing flowchart thereof, and FIG. 9 is a restoration processing flowchart thereof.
[0054]
7, the same components as those shown in FIG. 1 are indicated by the same symbols, and the difference from FIG. 1 is that a non-compression mode is added to the compression mode of FIG.
[0055]
That is, the image reading unit (PC card) 2 includes an image signal processing unit 20, a reading control unit, and a device interface (I / F) unit 28. The image signal processing unit 20 digitizes an analog image signal from the image reading unit 10 of the scanner main body 1 and includes an A / D conversion circuit.
[0056]
The reading control unit performs overall reading control and data transfer to the information processing apparatus 4. The reading control unit includes a level conversion image data generation unit 22, a sensor input characteristic holding unit 24, and a packing data generation unit 26. As shown in FIG. 5, the sensor input characteristic holding unit 24 holds the sensor input characteristic determined by the input characteristic (bright characteristic, dark characteristic) of each pixel of the line sensor.
[0057]
The level-converted image data generation unit 24 generates image data in which the number of bits in the signal level direction is reduced for each pixel of the line sensor of the image reading unit 10 according to the input characteristics. The packing data generation unit 26 packs the data of each pixel generated by the image data generation unit 24 and outputs it to the device interface unit 28. The device interface unit 28 transfers image data to the information processing device 40.
[0058]
The reading control unit generates a line sensor activation (drive) signal of the image reading unit 10 at a constant period during the reading period, and drives the line sensor. A line signal (for example, 2370 pixels) from the line sensor is A / D converted by the image signal processing unit 20, and the processed line data is stored in a line buffer inside the image data generation unit 22. For example, the A / D conversion converts an analog image signal into an 8-bit, 256-level digital value.
[0059]
When the image data generation unit 22 detects a movement pulse from the movement amount detection unit 12 of the scanner body 1, the image data generation unit 22 performs level conversion of the line data stored in the line buffer. That is, the level is reduced according to the light / dark characteristics of the sensor according to the holding characteristics of the holding unit 24.
[0060]
Further, a data transfer rate detection unit 30 for detecting the transfer rate of the device I / F unit 28 and a transfer rate evaluation unit 32 are provided as means for switching the presence / absence of the compression process. The transfer rate evaluation unit 32 compares the data generation rate obtained from the driving cycle of the sensor 10 with the actual data transfer rate. If the data transfer rate is low, the level conversion image data generation unit 22 requires the number of levels ( The value of the sensor input holding unit 24) is transmitted.
[0061]
Here, the data transfer rate detection method may output a predetermined rate according to the type of interface, or may be calculated from data sent from the device I / F unit 28 per unit time. Also good.
[0062]
When there is no level number instruction from the evaluation unit 32, the level conversion image data generation unit 22 and the packing data generation unit 26 do not perform the above-described level conversion processing and packing processing, and the line data in the line buffer of the image data generation unit 22 Is output to the device I / F unit 28 as it is.
[0063]
On the contrary, the level conversion image data generation unit 22 and the packing data generation unit 26 perform the level conversion process and the packing process from the input characteristics as described above when the level number instruction is given from the evaluation unit 32, and the line data is converted. The data is output to the device I / F unit 28.
[0064]
The processing of the reading control unit will be described with reference to FIG.
[0065]
(S10) First, at the start of processing, the number L of read lines (for example, 4) is set from the information processing apparatus 40 to the read control unit of the scanner 2.
[0066]
(S11) The set line number L is compared with the actual read line number. If the read line number has reached the set line number L, the process is terminated.
[0067]
(S12) It is detected whether an encoder pulse has arrived from the movement amount detection unit 12 of the scanner body 1. If the encoder pulse has not arrived, the process returns to step S11.
[0068]
(S13) When the encoder pulse has arrived, the line data via the image processing unit 20 is latched in the line buffer.
[0069]
(S14) The read line number is incremented by “1”.
[0070]
(S15) The transfer speed (number of clocks) of the I / F unit 28 is detected and compared with the data generation speed S obtained from the drive cycle of the line sensor. If the transfer rate is not lower than the data generation rate, there is no need for level reduction and packing processing, so the line sensor data is output to the I / F unit 28 and the process returns to step S11.
[0071]
(S16) Conversely, when the transfer rate is lower than the data generation rate S, the number of bits is reduced and the packing process is performed by the above-described level reduction. The data and the compression flag are output to the I / F unit 28, and the process returns to step S11.
[0072]
In this way, data compression can be performed with a buffer capacity for one line without losing spatial resolution information. In addition, since level conversion is performed according to the input characteristics of the sensor, it is possible to prevent missing level information.
[0073]
Furthermore, since compression and non-compression are controlled according to the transfer speed, for example, the processing can be automatically changed according to the sub-scanning speed. That is, since the compression is not performed when the sub-scanning speed is low, the entire processing speed is improved, and when the sub-scanning speed is high, the compression is performed, so that the transfer speed can be improved.
[0074]
FIG. 9 is a restoration process flow diagram of one embodiment of the present invention.
[0075]
(S20) It is determined whether or not the reading of the line image data of the designated number of lines is completed. If reading is complete, the restoration process ends.
[0076]
(S22) If the reading is not completed, the transferred line data is read.
[0077]
(S24) As shown in FIG. 4 or FIG. 6, the line data is unpacked by the image effective bit I, subjected to reverse level conversion, and restored to the original line data. Then, the process returns to step S20.
[0078]
  This restoration process is performed when data is compressed, and can be applied to the embodiments shown in FIGS. 1, 4, and 6. Furthermore, FIG., FIG.In this embodiment, when the compression flag is recognized, this decompression process is executed.
[0079]
The handy scanner that manually operates the image scanner has been described, but the present invention can also be applied to an image scanner that is not manually operated.
[0080]
(Supplementary note 1) In an image reading method of scanning a reading target with a line sensor to generate two-dimensional image data to be read, the number of output bits of each pixel of the line sensor is set according to the input characteristics of the line sensor. An image reading method comprising: a step of reducing, a step of packing the output of the reduced number of bits into transfer data of a predetermined unit, and a step of transferring the transfer data.
[0081]
(Supplementary note 2) The image reading method according to supplementary note 1, wherein the reduction step includes a step of reducing at least a bright input characteristic of each pixel of the line sensor by a determined number of reduction bits.
[0082]
(Supplementary note 3) The image reading method according to supplementary note 1, wherein the reduction step includes a step of reducing by a reduction bit number determined for each pixel from a light-dark input characteristic of each pixel of the line sensor.
[0083]
(Appendix 4) A step of comparing the drive speed of the line sensor and the data transfer speed, a step of transferring output data of the line sensor when the data transfer speed is slower than the drive speed, and the drive speed The image reading method according to claim 1, further comprising a step of executing the reduction and packing steps when the data transfer rate is higher.
[0084]
(Supplementary Note 5) In an image reading apparatus that scans a reading target with a line sensor and generates two-dimensional image data of the reading target, an input characteristic holding unit that holds input characteristics of the line sensor, and each pixel of the line sensor A read control unit that reduces the number of output bits in accordance with the held input characteristics, and packs and transfers the output of the reduced number of bits into transfer data of a predetermined unit. An image reading apparatus.
[0085]
(Supplementary note 6) The image reading apparatus according to supplementary note 5, wherein the reading control unit reduces the number of bits determined based on at least a bright input characteristic of each pixel of the line sensor.
[0086]
(Appendix 7) The reading control unit compares the driving speed of the line sensor with the data transfer speed, and when the data transfer speed is slower than the driving speed, transfers the output data of the line sensor, The image reading apparatus according to appendix 5, wherein the reduction and packing are executed when the data transfer speed is faster than a driving speed.
[0087]
(Supplementary Note 8) In an image reading system that scans a reading target with a line sensor and generates two-dimensional image data of the reading target, the number of output bits of each pixel of the line sensor is determined according to the input characteristics of the line sensor. And reducing and reducing the output of the reduced number of bits into a predetermined unit of transfer data and transferring the data, and receiving the transfer data and converting the data of each pixel into the reduced number of bits Accordingly, an image reading system having a data restoration unit that restores data having the original number of bits.
[0088]
(Supplementary Note 9) A scanning target is scanned with a line sensor, the number of output bits of each pixel of the line sensor is reduced according to the input characteristics of the line sensor, and the output of the reduced number of bits is a predetermined unit. In the image restoration method for restoring the compressed data packed in the transferred data, the step of receiving the transferred data and dividing it into the data of each pixel, and restoring the original number of bits according to the reduced number of bits An image restoration method comprising the steps of:
[0089]
(Additional remark 10) The program which produces | generates the two-dimensional image data of the reading object scanned with the line sensor, Comprising: The program data which reduces the output bit number of each pixel of the said line sensor according to the input characteristic of the said line sensor And program data for packing the output of the reduced number of bits into transfer data of a predetermined unit.
[0090]
【The invention's effect】
  As described above, according to the present invention, data compression can be realized while maintaining high resolution by storing spatial resolution information by information compression in the pixel level direction in pixel units using sensor characteristics. For this reason, it is possible to prevent reading of spatial resolution information such as a two-dimensional bar code and to improve the decoding rate.Further, compression and non-compression can be automatically switched according to the sub-scanning speed of the line sensor, and the processing speed can be improved while maintaining high resolution.
[Brief description of the drawings]
FIG. 1 is a block diagram of an image scanner according to an embodiment of the present invention.
2 is a configuration diagram of a scanner system using the image scanner of FIG. 1. FIG.
3 is an external view of the image scanner of FIG. 1. FIG.
4 is an explanatory diagram of an image compression / decompression method of the image scanner of FIG. 1. FIG.
FIG. 5 is an input characteristic diagram of the line sensor for FIG. 4;
FIG. 6 is an explanatory diagram of image compression / decompression processing according to another embodiment of this invention.
FIG. 7 is a block diagram of an image scanner according to another embodiment of the present invention.
8 is a flowchart of the data compression process of FIG.
FIG. 9 is a flowchart of data restoration processing in FIG. 7;
FIG. 10 is an explanatory diagram of the prior art.
FIG. 11 is an explanatory diagram of conventional shading correction.
FIG. 12 is an explanatory diagram of a conventional image compression process.
FIG. 13 is an explanatory diagram of problems in the prior art.
[Explanation of symbols]
1 Scanner body
2 PC card (reading control unit)
10 Image reading unit
12 Movement amount detector
20 Image signal processor
22 level image data generator
24 Sensor input characteristics holding unit
26 Packing data generator
28 Device I / F section

Claims (6)

In an image reading method for sub- scanning a reading target with a line sensor and generating two-dimensional image data of the reading target,
Reducing the number of output bits of each pixel of the line sensor according to at least the bright input characteristics of each pixel of the line sensor;
Packing the reduced number of bits of output into a predetermined unit of transfer data; and
Transferring the transfer data;
Comparing the data generation speed in the sub-scanning direction, which is the driving speed of the line sensor, and the data transfer speed of the transfer data;
When the data transfer rate is faster than the generation rate, transferring the output data of the line sensor;
An image reading method comprising: the step of reducing and the step of packing when the data transfer rate is slower than the generation rate. In an image reading apparatus that sub- scans a reading target with a line sensor and generates two-dimensional image data of the reading target,
An input characteristic holding unit that holds at least a bright input characteristic of each pixel of the line sensor;
The number of output bits of each pixel of the line sensor is reduced according to at least the bright input characteristics of each of the held pixels, and the output of the reduced number of bits is packed into transfer data in a predetermined unit. A reading control unit for transferring,
The reading control unit compares the data generation speed in the sub-scanning direction, which is the driving speed of the line sensor, with the data transfer speed of the transfer data, and when the data transfer speed is faster than the generation speed, the line sensor The output data is transferred, and when the data transfer speed is slower than the generation speed, the reduction and the packing are executed. In an image reading system that sub- scans a reading target with a line sensor and generates the two-dimensional image data of the reading target,
The number of output bits of each pixel of the line sensor is reduced according to at least the bright input characteristics of each pixel of the line sensor, and the output of the reduced number of bits is packed into transfer data in a predetermined unit. A data transfer unit to transfer,
A data restoration unit that receives the transfer data and restores data of each pixel to data of the original number of bits in accordance with the bright input characteristics, according to the reduced number of bits;
The data transfer unit compares a data generation speed in the sub-scanning direction, which is a driving speed of the line sensor, with a data transfer speed of the transfer data, and when the data transfer speed is higher than the generation speed, the line sensor When the data transfer rate is slower than the generation rate, the reduction and the packing are performed.
The image restoration system, wherein the data restoration unit performs the restoration on the reduced and packed transfer data. The scanning target is sub- scanned by a line sensor, the number of output bits of each pixel of the line sensor is reduced according to at least the bright input characteristics of each pixel of the line sensor, and the output of the reduced number of bits Is restored to compressed data with a compression flag added to a predetermined unit of transfer data,
Receiving the transfer data and recognizing a compression flag of the transfer data;
Dividing the transfer data into data of each pixel when the transfer flag recognizes that the transfer data is compressed by the compression flag;
An image restoration method comprising: restoring the original bit number data according to the reduced number of bits according to the bright input characteristic. An image reading program for generating two-dimensional image data to be read sub- scanned by a line sensor,
Program data for reducing the number of output bits of each pixel of the line sensor according to at least the bright input characteristics of each pixel of the line sensor;
Program data for packing the reduced number of bits of output into a predetermined unit of transfer data; and
The data generation speed in the sub-scanning direction, which is the driving speed of the line sensor , is compared with the data transfer speed of the transfer data. If the data transfer speed is faster than the generation speed, output data of the line sensor is output. An image reading program comprising: program data for executing the reduction and the packing when the data transfer rate is slower than the generation rate. The image reading apparatus according to claim 2, wherein the reading control unit transfers the transfer data and the compression flag when the reduction and the packing are executed.

Images