JP2004151833A - Document verification method and device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method and a device for document verification which improve the effect of prevention of forgery and which are easily obtained. <P>SOLUTION: Conditions of the surface and the irregularity of thickness of paper used for an official document 12 (original) is detected by a condition detecting part 18 and condition information which indicates the condition of the detected paper is registered in a memory 20 in advance. The condition of the surface of the paper is observed by illuminating a non-image recording part of the document 12 and detecting its reflection, and the condition of the irregularity of thickness of the paper is observed by illuminating the non-image recording part of the document 12 and detecting transmitted light. As for the document 12 of a verification target, in the same way, the condition of the paper used for the document 12 of the verification target is detected by the condition detecting part 18, and the condition information obtained as the result of that is compared with the condition information of the original registered in advance by a verification processing part 22, and then it is determined whether the document 12 of the verification target is an original or not according to similarities between them. <P>COPYRIGHT: (C)2004,JPO

Description

【００８３】
このようにして求めた正規化相関Ｒの要素ｒ（ｍ，ｎ）は、その値が大きいほど類似度が高いことを示し、正規化相関Ｒの要素ｒ（ｍ，ｎ）の中の最大値ｒｍａｘが所定の閾値以上であれば、確認対象の書類１２は原本であると判定し、最大値ｒｍａｘが所定の閾値に満たなかった場合は、確認対象の書類１２は原本ではないと判定すればよい。
【００８４】
検出状態情報と登録状態情報との類似度は、上記のような相関値以外にも、従来公知の各種の類似性を評価するための評価値を用いることができる。例えば、検出状態情報及び登録状態情報の累積２乗誤差を求めることにより、類似度を評価することもできるし、登録状態情報と検出状態情報の各々を量子化・標本化して各々の特徴を抽出し、抽出した特徴をベクトルとして扱って、両者の距離を求めたり、ベクトルの成す角度から類似度を評価することもできる。また、検出状態情報と登録状態情報は画像データであるので、これを例えば２次元フーリエ変換により周波数領域へ変換し、フーリエ空間上で識別、照合しても良い。この場合、両者をフーリエ空間上で画像合成し、逆フーリエ変換することで相関強度画像を得て、そのピーク値から二つの両者の類似度を評価することができる。例えば、振幅のピークの大きさが予め決めておいた閾値以上であった場合に同一画像、即ち確認対象の書類は原本であると判断する。
【００８５】
なお、上記の比較判定時に用いる閾値については、状態情報の検出誤差を見込んで、所定の許容範囲を設けて設定することが好ましい。すなわち、原本保証をどの程度厳しくするかの利用者の要求に応じて、閾値の大きさを適宜選択すればよい。また、書類１２の種類によって許容範囲が異なる、すなわち、書類１２毎に閾値が異なる場合もあるので、登録状態情報の登録時（図７のステップ１０６）には、識別情報と共に、閾値についてもメモリ２０に記憶しておくとよい。
【００８６】
ここで、前述したように、各書類１２について４つの状態情報が取得されるので、上記の比較判定を確認対象の各書類１２について４回行うことができるので、具体的には、これら４回の判定結果が全て「原本」であった場合に、「原本」であるとの最終判定を下し、それ以外の場合には「非原本」と最終判定を下すようになっている。ただし、実際には、検出領域３６の汚れ等による状態情報の不一致を補うため、例えば、４回の判定のうち２回が「原本」であるとの結果を出せば最終的に「原本」であると判定するようにしてもよい。このように最終的な判定結果を下す基準は、原本保証をどの程度厳しくするかの利用者の要求に応じて定めればよい。
【００８７】
また、照合時に何らかの操作ミスや位置ずれなどの事故が発生する場合があるため、複数回の判定結果により最終的な判定を下すようにしたり、比較結果が原本となる印刷物であると判定しない場合に所定の回数まで再試行を認めるようにしてもよい。
【００８８】
なお、書類１２の表面（又は裏面）上に符号化して状態情報を表すデータが印刷してある場合には、画像読取部１４（又は専用の画像読取部）により当該書類１２の表面（又は裏面）からそのデータを読み取って、登録状態情報として用いればよいことは言うまでもない。
【００８９】
そして、最後に、ステップ１３６で、原本／非原本の判定結果を出力して、図８の確認処理は終了する。
【００９０】
このように、本実施の形態では、書類１２に用いられている紙表面状態や厚さムラ状態を表す状態情報を確認情報として用い、予め正規の書類１２（原本）の紙表面状態や厚さムラ状態を登録しておき、確認対象の書類１２から検出された紙表面状態や厚さムラ状態と比較することで、当該確認対象の書類１２の真偽（原本／非原本）を判定するようになっている。これにより、各書類１２の特徴を捕らえて原本であるか否かを確認することができる。
【００９１】
一般に、印刷物などの書類１２の偽造や改竄は、印刷情報である電子データを不正に入手し、またデータを改竄し、偽造、改竄されたデータはプリンタや複写機によって紙などの画像記録媒体上に印刷されるが、上記のように正規の書類１２（原本）の紙表面状態や厚さムラ状態を確認情報として用いることにより、表面状態や厚さムラ状態までも偽造することは不可能であるため、このような詐称を容易に見抜くことができる。
【００９２】
また、従来のように、書類１２に対して、原本であることを示す特殊印刷処理を施したり、特殊紙を用いる必要がないため、低コストで容易に実現可能である。
【００９３】
なお、上記では、書類１２に、当該書類１２に割当てられた識別情報が予め印刷などにより記録されている場合を説明したが、本発明はこれに限定されるものではない。例えば、書類１２の紙面全体或いは所定の部分に印刷或いは手書きされている文字をＯＣＲにより読取り、読取った文字数分のコードデータとして、このコードデータをある種の関数、例えばＭＤ５などのハッシュ関数を用いて求めたハッシュ値を識別情報として用いることも可能である。この場合、識別情報が唯一無二であることを保証することはできないが、経験上、同一のハッシュ値が異なった文字列から求められることは極めて稀である。
【００９４】
また、上記では、識別情報により対応する登録状態情報を選択し、検出状態情報と登録状態情報とを１対１で比較する場合を説明したが、メモリ２０の登録データ数が少ない場合には、識別情報を用いずに、検出状態情報を全登録状態情報と比較してもよい。この場合、識別情報の抽出が不要となるため、そのための画像読取部１４及び識別情報抽出部１６は省略可能である。
【００９５】
【発明の効果】
上記に示したように、本発明は、書類の表面状態や厚さムラ状態を利用して、書類が原本であるか否かを確認することで、偽造防止効果を向上させることができ、且つ特殊印刷や特殊紙が不要であるので容易に実現可能であるという優れた効果を有する。
【図面の簡単な説明】
【図１】本実施の形態に係わる書類確認装置の基本構成を示す構成図である。
【図２】書類の一例を示す図である。
【図３】画像読取装置の一例を示す斜視図である。
【図４】図３の画像読取装置の書類がセットされた状態を示す断面図である。
【図５】図３の画像読取装置のキャリッジ内の構成を示す斜視図である。
【図６】図３の画像読取装置の厚さムラ検出用にプラテンカバーに設けられた光源の一例を示す断面図である。
【図７】書類確認装置で実行される登録処理を示すフローチャートである。
【図８】書類確認装置で実行される確認処理を示すフローチャートである。
【符号の説明】
１０ 書類確認装置
１２ 書類
１４ 画像読取部
１６ 識別情報抽出部
１８ 状態検出部
２０ メモリ
２２ 確認処理部
３６ 検出領域
４０ 画像読取装置
４４ プラテンカバー
４６ プラテンガラス
４８ キャリッジ
４８Ａ 開口
５０ 光源
５２ ラインイメージセンサ
６０ 光源
６２ ラインイメージセンサ
６４ 反射板
６６ 光源
６８ ラインイメージセンサ[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a document checking method and device, and more particularly to a document checking method and device for checking the originality of a document.
[0002]
[Prior art]
Securities, various titles, insurance policies, resident certificates, birth certificates, guarantees, paper documents such as passports, banknotes, confidential documents, etc., and documents such as ID cards, etc., must be original to prevent counterfeiting. It is necessary to take special measures that can be confirmed and cannot be forged.
[0003]
For this reason, conventionally, information indicating that the issuer is the original is added to the document by watermarking, magnetic printing, micro-characters, printing with special ink, or the like (special printing), or special paper which is not generally available. Forgery has been prevented by printing on the paper (see Patent Documents 1 to 7).
[0004]
In addition, there is a technique that, when a special printing is performed on a sheet and a copy of a paper document is made by a copying machine or the like, the background of the paper emerges, and it can be determined that the copy is a copy and not an original ( Patent Document 8).
[0005]
By the way, in a highly information-oriented society, it is assumed that contents included in numerous securities and certificates are digitized and printed and used. For this reason, it is expected that the general user will obtain information on securities and certificates through a network, print the information with a printer at home, and use the printed matter as securities and certificates. .
[0006]
[Patent Document 1]
JP-A-2002-083274
[Patent Document 2]
JP-A-2002-146254
[Patent Document 3]
JP-A-6-008678
[Patent Document 4]
JP-A-6-141135
[Patent Document 5]
JP-A-8-241451
[Patent Document 6]
JP-A-11-261806
[Patent Document 7]
JP 2000-222615 A
[Patent Document 8]
JP-A-10-071791
[0007]
[Problems to be solved by the invention]
However, conventional anti-counterfeit technology requires advanced technology for special printing and special paper must be generally unavailable, so it is not suitable for documents where general users are the issuers. There was a problem that there was no. In addition, the cost of special printing and special paper is high, and the issuers of securities and certificates other than general users are also required to reduce costs. Further, with the recent improvement of copy technology, there has been a problem that it is possible to copy securities and certificates which cannot be distinguished from the original by the conventional forgery prevention technology.
[0008]
SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and an object of the present invention is to provide a method and an apparatus for confirming a document which can improve the effect of preventing forgery and can be easily realized.
[0009]
[Means for Solving the Problems]
In order to achieve the above object, the invention according to claim 1 is a document confirmation method for confirming the originality of a document, and the surface state and thickness unevenness state of the document are determined in advance from the original document. Detecting at least one of the states, storing first state information indicating the detected state, and detecting at least one of the surface state and the thickness unevenness state of the document from the document to be confirmed, The second state information indicating the detected state is compared with the first state information stored in advance, and based on the comparison result, it is determined whether or not the document to be confirmed is the original. It is characterized by that.
[0010]
According to the first aspect of the present invention, at least one of the surface state and the thickness unevenness state is detected from the original document and stored as first state information in advance. The first state information may be stored in a storage device such as a memory, or may be stored in the original document itself by printing.
[0011]
When confirming whether or not the document is the original, a state is detected from the document to be confirmed in the same manner as the original, and the second characteristic information indicating the detected state is defined as the first characteristic information. Compare. In other words, by comparing the surface state and thickness unevenness state of the document to be checked with the surface state and thickness unevenness state of the original document registered in advance, the document to be checked can be compared based on the similarity between the two. Is determined.
[0012]
Since the surface state and the thickness unevenness state are the only ones that can be produced in the process of manufacturing a medium such as paper used for a document, the characteristics of this state are used to identify each document individually. And it is easy to confirm whether or not the original is the original. Therefore, the forgery prevention effect can be improved. Further, special printing, special paper, and the like are unnecessary as in the related art, so that the present invention can be easily realized.
[0013]
The document may be a printed matter on which an image is recorded by printing, or may be a handwritten document.
[0014]
In the document confirmation method described above, the surface state can be easily detected by irradiating the document with light and receiving reflected light from the document. In this case, it is preferable to irradiate the document with near-ultraviolet light. Further, the thickness unevenness state can be easily detected by irradiating the document with light and receiving transmitted light from the document. In this case, it is preferable to irradiate the document with near-infrared light.
[0015]
In the above-described document confirmation method, it is preferable that the state is detected from a predetermined non-image recording portion.
[0016]
Here, it is required to match the position (area) for detecting the state between the original document and the document to be confirmed, but it is difficult to completely match the two. For this reason, in the above-described document confirmation method, the area where one of the first state information and the second state information is detected is included in the area where the other is detected, and the one state information and the The similarity between the area in which one is detected and the information in the area of the same size in the other state information is obtained a plurality of times while moving the area of the same size in the other state information, and the maximum value of the similarity is calculated. Is greater than or equal to a predetermined value, the document to be confirmed may be determined to be the original.
[0017]
In the above-described document confirmation method, in order to improve the confirmation accuracy, a plurality of the first state information is obtained from the original document, and each of the plurality of the first state information is acquired. Acquiring a plurality of pieces of second state information from the document to be confirmed, comparing each of the plurality of pieces of second state information with a corresponding plurality of pieces of first state information, and performing the determination a plurality of times. It is preferable that a final determination is made as to whether or not the document to be confirmed is the original based on the results of the plurality of determinations.
[0018]
Also, in the above document confirmation method, identification information for identifying the document is obtained from the original document, and the first state information is associated with the identification information obtained from the original document. The identification information is acquired from the document to be confirmed, and the first state information and the first state information associated with the identification information that matches the identification information acquired from the document to be confirmed are stored. It is good to compare the status information of the second. The identification information in this case may be added to the original document in advance by, for example, character printing, barcode printing, or embedding in an image using an electronic watermark, or a predetermined information of at least a part of the document may be added. Characters recorded in the area are read, the read characters are converted into a hash value using a hash function, and the hash value can be used as identification information of the document.
[0019]
The above document confirmation method can be easily realized by a document confirmation device configured as follows. That is, as described in claim 10, a document checking device for checking the originality of a document, wherein at least one of a surface state and a thickness unevenness state of the document is detected from the document. Detection means; storage means for storing first state information indicating the state detected from the original document by the state detection means; and second means indicating the state detected from the document to be confirmed by the state detection means. Determining means for comparing the state information with the first state information stored by the storage means, and determining whether or not the document to be confirmed is the original based on the comparison result; It is characterized by having.
[0020]
In the above-described document confirmation device, the state detecting means may include a light source for irradiating the document with light, and light reflected or transmitted from the document by the light applied to the document by the light source. Light receiving means for receiving light. In this case, in order to efficiently detect the surface state, the light source irradiates near-ultraviolet light, and the light receiving unit receives the reflected light, and in order to efficiently detect the thickness unevenness state, Preferably, the light source emits near-infrared light. In addition, in order to reduce the influence of irradiation unevenness due to the light source, the state detection unit may further include a diffusion unit that diffuses light emitted to the document. Alternatively, an electroluminescent element may be used as the light source. Further, it is preferable that the state detecting means detects the state from a predetermined non-image recording portion of the document.
[0021]
Further, in the above-described document confirmation device, the determination unit is configured such that an area where one of the first state information and the second state information is detected is included in an area where the other is detected, The similarity between the one state information and the information in the area of the same size as the area in which the one is detected in the other state information is determined a plurality of times while moving the area of the same size in the other state information. Preferably, when the maximum value of the similarity is equal to or greater than a predetermined value, the document to be confirmed is preferably determined to be the original.
[0022]
Further, in the above-described document confirmation device, the state detecting means detects the state a plurality of times from the document, and the determining means determines the plurality of the second states obtained by the state detection a plurality of times. Comparing each of the information with a corresponding plurality of the first state information, performing the determination a plurality of times, and finally determining whether the document to be confirmed is the original based on the results of the plurality of determinations Is preferred.
[0023]
Further, the above-mentioned document confirmation device further comprises identification information acquisition means for acquiring identification information for identifying the document from the document, and the storage means stores the first state information in the identification information acquisition information. Means for storing the identification information obtained from the original document in association with the identification information, wherein the determination means associates the identification information with the identification information obtained from the document to be confirmed by the identification information obtaining means, The obtained first state information and the second state information may be compared. In this case, the identification information acquisition unit may read the identification information added in advance to the document by at least one of character printing, barcode printing, and embedding in an image by an electronic watermark, Characters recorded in at least a part of a predetermined area of the document may be read, the read characters may be converted into a hash value using a hash function, and the hash value may be used as identification information of the document.
[0024]
BEST MODE FOR CARRYING OUT THE INVENTION
Next, an example of an embodiment according to the present invention will be described in detail with reference to the drawings.
[0025]
FIG. 1 shows a basic configuration of a document confirmation apparatus to which the present invention is applied. A document checking device 10 shown in FIG. 1 extracts an identification information pre-recorded on the document 12 from a reading result by the image reading unit 14 as an image reading unit 14 for reading an image from the document 12 and an identification information acquiring unit. An identification information extracting unit 16, a state detecting unit 18 as state detecting means for detecting the state of the paper used in the document 12, and a memory for storing state information indicating the state detected by the state detecting unit 18 20 and a confirmation processing unit 22 for confirming the authenticity (original / non-original) of the document 12 based on the state of the document 12 as a discriminating means. Each of these units may be mounted on one device, or at least one may be configured as a physically different device and connected and used via connection means such as a cable.
[0026]
The document 12 may be a printed matter printed by a printing device such as a printer or a copier, or may be a handwritten document. Further, the document 12 is preliminarily recorded with unique identification information for identifying the document 12. Hereinafter, as an example, a case where an ID number serving as identification information is printed in advance at a predetermined position on an image recording surface of the document 12 will be described. The ID number is, for example, about 3 to 20 digits in decimal, and the ID number may be sequentially allocated to the paper to be used or may be randomly allocated. In any case, the same ID number should not be generated.
[0027]
For example, as shown in FIG. 2, a method for recording the identification information is to provide an entry area 30 for the identification information at a predetermined position on the image recording surface of the document 12, and to put a code indicating the identification information in the entry area 30. The barcode area 32 may be provided by printing, stamping, or handwriting, or the identification information may be converted into a barcode and printed in the barcode area 32. In addition, identification information may be embedded in an image such as the seal 34 by using a digital watermark technique (for example, refer to Japanese Patent Application No. 2002-244412). In particular, when the identification information is embedded by the digital watermarking technology, it is possible to hide from the user that the document 12 has a mechanism for confirming the originality of the document 12 by the document confirming device 10, which makes the user feel uncomfortable. In addition, there is an advantage that a forged document can be easily found and a forgery prevention effect can be obtained. Also, the same effect can be obtained by printing identification information on the document 12 using an image forming material such as ink or toner containing an infrared absorbing material or an ultraviolet absorbing material so that it cannot be visually recognized by humans.
[0028]
As the image reading unit 14, an image reading device (a so-called scanner) generally used as a peripheral device of a PC can be used. More specifically, the image reading unit 14 irradiates the document 12 with light and detects the reflected light. Then, the image recorded on the document 12 is read. In general, when an image is read from a document, when light is irradiated perpendicularly to the paper surface of the document, the direct reflected light causes the document to shine. Therefore, in the image reading unit 14, in order to prevent this shine, It is preferable to irradiate the light at an angle with respect to the paper surface (that is, an incident angle 入射 0). In the image reading section 14, image data representing the read image is obtained. This image data can be transmitted to an electronic information device such as a personal computer (PC) as necessary, and can be used as character information by being subjected to electronic filing of the document image or an OCR (Optical Character Recognition) algorithm. This image reading unit 14 is connected to the identification information extracting unit 16.
[0029]
The identification information extraction unit 16 extracts identification information from the image data acquired by the image reading unit 14. For example, as shown in FIG. 2, when a writing area 30 is provided at a predetermined position of the document 12 and a code indicating identification information is written in the writing area 30 by printing, stamping, or handwriting, an image By reading character information by OCR from a portion corresponding to the entry area 30 of the image data obtained by the reading unit 14, identification information can be extracted. When a barcode area 32 is provided at a predetermined position of the document 12 and a barcode indicating identification information is printed in the barcode area 32, the barcode of the image data acquired by the image reading unit 14 is displayed. By reading and decoding a barcode from a portion corresponding to the code area 32, identification information can be extracted. For example, when the identification information is embedded in the seal 34 of the document 12 by the digital watermarking technique, the identification data is analyzed by analyzing the data of the seal 34 of the image data acquired by the image reading unit 14. Information can be extracted.
[0030]
The state detection unit 18 detects at least one of the surface state and the thickness unevenness state of the paper used for the document 12. Examples of the surface state include irregularities on the surface of the paper, the thickness of the fiber, the quality of the fiber (scratch), the transparency, and the like, which are generated when the paper used for the document 12 is manufactured. Such surface state and thickness unevenness state are unique because they are formed due to the degree of fiber overlap when pulp is made in the paper manufacturing process and the concentration unevenness of the pulp suspension used for making. These states are not reproducible even with current copying technology.
[0031]
The surface state of the paper can be observed by irradiating the non-image recording portion of the document 12 with light and detecting the reflected light. In order to detect the surface state efficiently, light is radiated obliquely to the paper surface of the document 12 so that irregularities on the surface appear as shadows, and the light is irradiated in a direction substantially perpendicular to the paper surface of the document 12. It is preferable to detect the reflected light, and the irradiation light to the document 12 has a larger inclination (incident angle) than the irradiation light from the image reading unit 14. Note that dark field illumination has been conventionally used to detect dust and scratches attached to the surface, and the surface state of the paper may be detected under dark field illumination.
[0032]
Further, the unevenness of the thickness of the paper can be observed by irradiating the non-image recording portion of the document 12 with light and detecting the transmitted light. In this case, light may be emitted substantially perpendicularly to the paper surface of the document 12 to detect light transmitted substantially straight through the paper of the document 12 and transmitted.
[0033]
The portion (detection area) of the document 12 to which light is irradiated for state detection may be an image recording portion, but the reflected light and transmitted light from the image recording portion include image information. The recording part is preferred.
[0034]
Each of the identification information extraction unit 16 and the state detection unit 18 is connected to a memory 20, and the memory 20 stores, as first state information, state information indicating a state detected from the original document 12. The information is stored in association with the identification information extracted from the document 12. Note that, in the present embodiment, the state detection unit 18 has a function as a storage unit.
[0035]
Therefore, the state information of the original document 12 is stored in the memory 20 so as to be uniquely associated with the document 12. The present invention does not particularly limit the method of associating the state information with the identification information. For example, the method may be performed using a table indicating the correspondence between the state information and the identification information, or the data of the state information. The identification information may be used for at least a part of the name.
[0036]
Further, each of the identification information extraction unit 16 and the state detection unit 18 is also connected to the confirmation processing unit 22, and the confirmation processing unit 22 checks as the second state information whether or not the original is the original. State information indicating a state detected from the target document 12 and identification information extracted from the document 12 are input.
[0037]
The confirmation processing unit 22 is also connected to the memory 20 and can arbitrarily read the registration information in the memory 20. The confirmation processing unit 22 receives information input from the identification information extraction unit 16 and the state detection unit 18, reads out the state information stored in the memory 20 in association with the input identification information, and reads the input state. The information is compared with the read state information, and it is determined whether or not the document 12 is the original, that is, the authenticity of the document 12 according to the degree of similarity.
[0038]
Next, an example in which the state detection unit 18 is mounted on an image reading device (a so-called scanner) generally used as a peripheral device of a PC will be specifically described with reference to FIGS. That is, this image reading device is equipped with at least the image reading unit 14 and the state detecting unit 18 described above. As for the identification information extracting unit 16, the memory 20, and the confirmation processing unit 22, a microcomputer or a circuit having the function of each of these units may be mounted on the image reading device 40, or a PC to which the image reading device 40 is connected. May be constructed by executing a predetermined program.
[0039]
As shown in FIG. 3, the image reading apparatus 40 is configured such that a platen cover 44 is attached to a housing 42 of the main body so as to be openable and closable (see arrow A). A platen glass 46 is provided on the upper surface of the housing 42, and a carriage for reading an image from the document 12 while moving in a predetermined direction (arrow B direction) is provided in the housing 42 below the platen glass 46. 48 are provided. That is, in the image reading device 40, as shown in FIG. 4, the document 12 to be read is placed on the platen glass 46 so that the image recording surface contacts the platen glass 46, and the platen cover 44 is closed. The image is read while the document 12 is held between the platen glass 46 and the platen cover 44.
[0040]
As shown in FIGS. 4 and 5, the carriage 48 includes an image reading light source 50 and a line image sensor 52 that function as the image reading unit 14 inside its housing.
[0041]
The housing of the carriage 48 is formed of a light-shielding resin or metal material to block external light, and its inner surface is also black to prevent unnecessary light reflection (stray light). . On the upper surface of the housing of the carriage 48, the light source 50 irradiates the document 12 placed on the platen glass 46 with light, and reflects the irradiation light from the irradiation position P of the document 12 by the light source 50. In order to capture light, a rectangular opening 48A having a long axis in a direction orthogonal to the moving direction of the carriage (see arrow B) is provided. As described above, in order to prevent shine, it is preferable to irradiate the document 12 with light with an inclination to the paper surface. It is designed to be oblique with respect to.
[0042]
The light source 50 is a long tubular fluorescent lamp, and is suitable in terms of shape and power consumption to illuminate the document 12 in a straight line. In addition, as the light source 50, in addition to the fluorescent lamp, an LED array in which LEDs (light emitting diodes) are arranged in a line, an incandescent lamp having a linear filament, or the like can be used. A reflection plate 54 made of, for example, aluminum or stainless steel is provided around the light source 50 except for the direction in which the document 12 is irradiated with light (that is, the direction of the opening 48A). Accordingly, of the output light of the light source 50, light traveling in the direction of the opening 48A passes through the platen glass 46 through the opening 48A, and the document 12 placed on the platen glass 46, more specifically, the image recording surface The light is irradiated to the upper irradiation position P. Light traveling in a direction other than the direction of the opening 48A is reflected by the reflector 54, deflected in the direction of the opening 48A, transmitted through the platen glass 46 through the opening 48A, and placed on the platen glass 46. The document 12, more specifically, the irradiation position P on the image recording surface is irradiated. By providing the reflection plate 54 in this manner, the output light of the light source 50 can be efficiently emitted to the irradiation position P of the document 12.
[0043]
The outer surface of the reflection plate 54 is coated with black paint to prevent unnecessary reflection (stray light) in the casing of the carriage 48. When a resin-made reflecting plate 54 is used, the inner surface may be mirror-finished by plating or the like, and the outer surface may be painted black or the resin itself may be black and have light-shielding properties.
[0044]
Irradiation light output from the light source 50 and applied to the document 12 on the platen glass 46 is reflected by the document 12. Then, the reflected light reflected in the direction perpendicular to the document 12 passes through the platen glass 46 again, and is taken into the casing of the carriage 48 through the opening 48A.
[0045]
In the housing of the carriage 48, a plurality of mirrors 56 and lenses 58 are arranged in the traveling direction of the light taken into the housing. When the light emitted from the light source 50 and illuminated on the document 12 is reflected from the document 12 into the housing of the carriage 48, the light is reflected by the plurality of mirrors 56 and guided to the line image sensor 52. At the same time, an image is formed on the light receiving surface of the line image sensor 52 by the lens 58. That is, a line image at the irradiation position P of the document 12 is formed on the light receiving surface of the line image sensor 52.
[0046]
As the line image sensor 52, a line CCD sensor in which CCDs are arranged in a line can be used. The line image sensor 52 receives the light imaged on the light receiving surface and converts the light into an electric signal corresponding to the amount of received light. By converting this electric signal into a digital signal by a signal processing circuit in the casing of the carriage 48 (not shown), image data indicating a line image recorded at the irradiation position P of the document 12 can be obtained. That is, the line image sensor 52 can read a line image from the document 12, and by moving the carriage 48 in the direction of the arrow B, the reading position (line) of the image on the document 12 is moved. The entire image can be read.
[0047]
A light source 60 and a line image sensor 62 are provided inside the housing of the carriage 48 in order to detect the surface state of the paper used for the document 12. The light source 60 and the line image sensor 62 function as the state detection unit 18 when detecting a surface state.
[0048]
The light source 60 is installed at a position where a predetermined detection area for detecting state information on the document 12 can be irradiated. Specifically, in this example, a case where the document 12 shown in FIG. 2 is used will be described. In detail, the document 12 shown in FIG. 2 is a certificate, and the contents as an illumination book are printed on the image recording surface. Further, detection regions 36A and 36B (hereinafter, these are referred to as “detection areas”) for detecting the state of the paper used in the If no distinction is made, a detection area 36) is provided. The document 12 is placed on the platen glass 46 with the vertical direction coinciding with the moving direction of the carriage 48 (arrow B), and the platen cover 44 is closed so that the document 12 is set on the image reading device 40. Has become.
[0049]
In this example, for simplicity of description, the platen glass 46 and the document 12 have the same size, and the alignment of the document 12 is performed only by aligning the edge of the document 12 with the edge of the platen glass 46. It has become to be. Therefore, in this example, the light source 60 is provided at the center of the carriage 48 in the direction orthogonal to the moving direction (arrow B).
[0050]
Note that the position of the detection area 36 on the document 12 is not particularly limited as long as the background of the paper is exposed and printing or entry is not performed. However, it is necessary to detect the state of the paper at the same position at the time of registration and at the time of confirmation, and it is required that the position of the detection area 36 on the document 12 be easily specified. Therefore, in practice, it is appropriate to provide the detection areas 36 in the upper and lower margins of the document 12 whose position can be easily specified by the distance (for example, several centimeters) from the upper and lower ends of the paper as in this example. is there. In addition to the above, a predetermined mark 38 (see FIG. 2) such as a circular shape may be provided on the document 12 for specifying the position of the detection area 36. In this case, for example, an area within a radius of a predetermined dimension from the center of gravity of the mark 38 is set as the detection area 36, or two marks 38 are provided, and an area in a predetermined range between the two marks 38 is detected. The position of the document 12 can be specified if three or more regions are provided, and any position can be set as the detection region. Such a mark 38 can be provided on the document 12 by printing, or by forming holes and irregularities on the document 12. Further, the mark 38 may be printed on the document 12 using an image forming material such as an ink or a toner containing an infrared absorbing material or an ultraviolet absorbing material so that the mark 38 cannot be visually recognized by a human.
[0051]
Further, in consideration of the influence of the aberration of the lens 58 in the image data acquired by the line image sensor 62, the position of the detection area 36 is determined by the position of the document 12 corresponding to the direction orthogonal to the moving direction of the carriage 48 as in this example. The central part in the left-right direction is preferred. Also, the size of the detection area 36 is set to about 10 mm square in consideration of the influence of the aberration.
[0052]
In addition, the light source 60 may have a size capable of irradiating at least the length of the detection region 36 in a direction orthogonal to the moving direction of the carriage 48, and may be a single LED, a plurality of LEDs arranged on a plane or in a straight line. One LED, one EL (electroluminescence) element, or the like can be used. A reflector 64 similar to the reflector 54 of the light source 50 is provided around the light source 60 except for the direction in which the document 12 is irradiated with light (that is, the direction of the opening 48A). The document 12 placed on the platen glass, more specifically, to the irradiation position P on the image recording surface, is efficiently transmitted through the platen glass 46 through the opening 48A. The direction in which the light source 60 irradiates the document 12 with light is designed so that the inclination is further greater than the direction in which the light source 50 irradiates the document 12 with light.
[0053]
Light output from the light source 60 and applied to the document 12 on the platen glass 46 is reflected by the document 12. Then, the reflected light reflected in a direction substantially perpendicular to the document 12 again passes through the platen glass 46 and is taken into the housing of the carriage 48 through the opening 48A. The plurality of mirrors 56 and the lenses 58 are also used for the reflected light from the document 12 of the light emitted from the light source 60 and applied to the document 12, and the line image sensor 62 uses the lens 58 of the reflected light. It is arranged at the imaging position. That is, after the reflected light from the document 12 of the irradiation light radiated to the document 12 by the light source 60 is taken into the housing of the carriage 48, the reflected light is reflected by the plurality of mirrors 56 and guided to the line image sensor 62. At the same time, an image is formed on the light receiving surface of the line image sensor 62 by the lens 58.
[0054]
As the line image sensor 62, a line CCD sensor in which CCDs are arranged in a line can be used. The line image sensor 62 receives the light imaged on the light receiving surface and converts the light into an electric signal corresponding to the amount of received light. These electric signals are converted into digital signals by a signal processing circuit in the casing of the carriage 48 (not shown), and linear image data representing the surface state of the irradiation position P of the document 12 is obtained. By moving the carriage 48, planar image data in the detection area 36 can be obtained, and the planar image data is used as state information indicating a surface state. In this example, state information is acquired from the detection area 36 at a resolution of 400 DPI (Dot Per Inch), and image data of 128 × 128 pixels is acquired as state information.
[0055]
Here, the emission wavelength of the light source 60 for detecting the surface state will be described.
[0056]
The light emission wavelength of the light source 60 may be in a range from a near infrared region (about 950 nm) to a near ultraviolet region (about 370 nm). In particular, if the emission wavelength is within the visible region, the light source and the image sensor for detecting the surface state can be shared with the light source 50 and the line image sensor 52 for reading an image, and the light source 60 and the line image sensor 62 can be omitted. is there. When the light source and the image sensor are used in common for reading an image and for detecting a surface state in this manner, the dynamic range is extended to a lower density side than in normal image reading, and an image is read with more gradations. Accordingly, it is possible to acquire image data including not only information of an image printed or handwritten on the document 12 but also information of a surface state of a background portion of paper. If only a portion corresponding to the detection area 36 is cut out from the image data obtained in this way and is used as state information indicating the surface state, it is possible to read the image and detect the surface state by one scan.
[0057]
In general, it is known that the shorter the emission wavelength, the lower the penetration into the paper used for the document 12 and the more the reflection on the surface. Therefore, when the LED having the central emission wavelength in the blue region (about 470 nm) is used as the light source 60 within the visible region, the penetration of light into the paper used for the document 12 is reduced, and the surface state is improved. This is preferable because it can be well captured by the reflected light from the document 12. For detecting the surface state, it is more preferable to use an LED whose central emission wavelength is in the near ultraviolet region as the light source 60. However, in this case, since the sensitivity wavelength range of the line image sensor 52 does not extend, and the position of the image forming plane by the lens 58 shifts, the light source and the image sensor are shared for image reading and surface state detection. It is difficult to do. Therefore, it is necessary to provide the light source 60 and the line image sensor 62 for detecting the surface state separately from the light source 50 for reading the image and the line image sensor 52 as in this example.
[0058]
Further, the image reading device 40 detects the uneven thickness state at each position corresponding to the detection areas 36A and 36B when the document 12 is sandwiched between the platen cover 44 and the platen glass 46 and the platen cover 44. Light sources 66A and 66B (hereinafter, referred to as a light source 66 when these are not distinguished). Therefore, light is emitted from the back surface by the light source 66 to the detection area 36 of the document 12 placed on the platen glass 46. Further, a line image sensor 68 for detecting a thickness unevenness state is provided in the carriage 48. That is, the light source 66 and the line image sensor 68 function as the state detector 18 when detecting the thickness unevenness state.
[0059]
The light source 66 only needs to be large enough to irradiate the detection region 36, and a single LED, a plurality of LEDs arranged on a plane or in a straight line, an EL (electroluminescence) element, or the like can be used. .
[0060]
When an LED is used as the light source 66, as shown in FIG. 6, in order to reduce the influence of the light emitting shape of the LED, a diffusion plate is used as a diffusion means in the light output direction of the LED 70 (the irradiation direction to the document 12). It is preferable that the light source 74 be disposed so that the output light of the LED 70 irradiates the document 12 substantially uniformly (that is, the light emitting surface is made uniform). In the example of FIG. 6, the light output direction of the LED 70 on which the diffusion plate 74 is disposed is set to the front, and the rear and sides of the LED 70 are case-processed so that the light is diffused and reflected inside. Covered with As the surface processing of the case 76, for example, the surface may be uniformly roughened with a material such as aluminum to eliminate the luster, or a process such as application of metal particles may be performed.
[0061]
Although FIG. 6 shows an example in which a single LED is used as the light source 66, the same applies when a plurality of LEDs are used. When an incandescent light bulb or a discharge tube is used as the light source 66 instead of the LED, the light emitting surface may be made uniform by a mechanism as shown in FIG. In the case where EL is used as the light source 66, a mechanism as shown in FIG. 6 is unnecessary because it has a uniform light emitting surface.
[0062]
Irradiation light output from the light source 66 and applied to the document 12 passes through the document 12 and passes through the platen glass 46. This transmitted light is taken into the casing of the carriage 48 through the opening 48A when the opening 48A is located below the light source 66 due to the movement of the carriage 48.
[0063]
The plurality of mirrors 56 and the lenses 58 are also used for transmitting the light emitted from the light source 66 to the document 12 and irradiating the document 12, and the line image sensor 68 uses the lens 58 of the transmitted light. Are arranged at the image forming position. That is, the transmitted light from the document 12 of the irradiation light applied to the document 12 by the light source 66 is taken into the housing of the carriage 48, reflected by the plurality of mirrors 56, and guided to the line image sensor 68. At the same time, an image is formed on the light receiving surface of the line image sensor 68 by the lens 58.
[0064]
As the line image sensor 68, a line CCD sensor in which CCDs are arranged in a line can be used. The line image sensor 68 receives the light imaged on the light receiving surface and converts the light into an electric signal corresponding to the amount of received light. These electric signals are converted into digital signals by a signal processing circuit in the casing of the carriage 48 (not shown), and linear image data representing the thickness unevenness state at the irradiation position P of the document 12 is obtained. By moving the carriage 48, planar image data in the detection area 36 can be obtained, and the planar image data is used as state information indicating a thickness unevenness state. In this example, since the state information is acquired at a resolution of 400 DPI (Dot Per Inch), even in the case of thickness unevenness, image data of 128 × 128 pixels is acquired as the state information.
[0065]
Here, the emission wavelength of the light source 66 for detecting the thickness unevenness state will be described.
[0066]
The light emission wavelength of the light source 66 may be in the near-infrared region (about 950 nm) to the near-ultraviolet region (about 370 nm) similarly to the light source 60. The image sensor for detecting the state can be shared with the line image sensor 52 for reading an image, and the line image sensor 68 can be omitted. However, when the image sensor is used in common for image reading and for detecting the uneven thickness state, it is necessary to perform two scans for the image reading and for detecting the uneven thickness state. At times, the dynamic range is expanded to a lower density side than at the time of image reading.
[0067]
In general, it is known that the longer the emission wavelength, the less the reflection on the surface of the paper used for the document 12 and the more the penetration into the paper. Therefore, when the LED having the central emission wavelength in the red region (about 700 nm) is used as the light source 66 in the visible region, the penetration of light into the paper used for the document 12 increases, and the state of the pressure unevenness is increased. Is better captured by the transmitted light from the document 12. In order to detect the thickness unevenness state, it is more preferable to use an LED whose central emission wavelength is in the near infrared region as the light source 66. However, in this case, since the sensitivity wavelength range of the line image sensor 52 does not extend, and the position of the image forming surface by the lens 58 shifts, the image sensor is shared for reading an image and detecting the thickness unevenness state. It is difficult to do. Therefore, it is necessary to always provide the line image sensor 68 for detecting the thickness unevenness state separately from the line image sensor 52 as in this example.
[0068]
As described above, by mounting the state detection unit 18 on a conventionally known image reading device (scanner), the document reading device 10 can be configured by combining the image reading device 40 with an information processing device such as a PC. It is not necessary to newly manufacture hardware, and it can be realized at low cost. In the above description, an example is shown in which both the surface state and the thickness unevenness state of the paper used for the document 12 can be detected, but only one of them may be used. Further, although an example in which two detection areas 36 are provided in the document 12 has been described, the present invention does not particularly limit the number of the detection areas 36. If it is possible to detect both the surface state and the thickness unevenness state, increase the number of detection areas, or increase the size of the detection area, the accuracy of checking the original can be improved, but the device Since the manufacturing cost increases, appropriate settings are made in relation to the cost.
[0069]
Next, the operation of the present embodiment will be described. In the following, a case where the above-described image reading device 40 is used will be specifically described.
[0070]
In the document checking device 10 according to the present embodiment, in order to determine the authenticity of the document 12, it is necessary to register the status information of the document 12 (original) in advance. FIG. 7 shows a registration process of the state information executed in the document confirmation apparatus 10 for this purpose. FIG. 7 shows an example of the registration process, and the order of each step may be appropriately changed.
[0071]
This registration process is executed after the document 12 to be registered is set in the document confirmation device 10. Specifically, the operator makes the original document 12 the vertical direction of the document 12 coincide with the movement direction of the carriage 48, and aligns the edge of the document 12 with the edge of the platen glass 46 so that the platen glass 46 The document 12 is set by placing it on the top and closing the platen cover 44.
[0072]
In the registration process, as shown in FIG. 7, first, in step 100, the image reading unit 14 reads an image from the set original document 12, and acquires image data indicating the image. Specifically, the carriage 48 is moved in the direction of arrow A from the upper end to the lower end of the document 12 while irradiating the document 12 with the light source 50 and receiving the reflected light with the line image sensor 52 to read the line image. Then, an image recorded on the document 12 is read, and image data for one sheet (page) of the document 12 is obtained. Subsequently, in step 102, the identification information is extracted from the image data acquired in step 100 by the identification information extracting unit 16, specifically, for example, from the entry area 30, the bar code area 32, or the seal 34 described above. Extract.
[0073]
In the next step 104, the state of the paper used for the document 12 is detected. Specifically, the carriage 48 is moved to each of the positions corresponding to the two detection areas 36 provided on the document 12, the document 12 is irradiated by the light source 60, and the reflected light is received by the line image sensor 62. , The state information indicating the surface state of the paper in each detection area 36 is obtained. Similarly, the document 12 is illuminated by the light source 66, the transmitted light is received by the line image sensor 68, and the state information indicating the unevenness of the paper thickness in each detection area 36 is obtained. That is, in this example, for one document 12, a total of four pieces of state information are obtained for two places of state information representing the surface state and two places of state information representing the thickness unevenness state.
[0074]
In the next step 106, the state information (four state information in this example) acquired in step 104 is registered in the memory 20 in association with the identification information extracted in step 102, and the registration processing in FIG. Will be terminated.
[0075]
In the above description, the case where the state information of the original document 12 is stored (registered) in the memory 20 has been described as an example, but the present invention is not limited to this. That is, the state information is stored in the original document 12 itself, for example, by encoding and printing data representing the state information on the front side (or the back side) of the document 12, so that the document 12 and the state information are combined. It is also possible to register them in association with each other. In this case, since the document 12 as well as the memory 20 functions as the identification information, the identification information extracting unit 16 can be omitted.
[0076]
Next, a case where the authenticity of the document 12 is confirmed will be described. FIG. 8 shows a confirmation process executed by the document confirmation device 10 for this purpose. FIG. 8 shows an example of the confirmation processing, and the order of each step may be appropriately changed.
[0077]
This confirmation process is executed after the document 12 to be confirmed is set in the document confirmation device 10 in the same manner as the original document 12 to be registered.
[0078]
In the confirmation process, as shown in FIG. 8, in step 120, the image reading unit 14 reads an image from the set document 12 to be confirmed, acquires image data indicating the image, and in step 122, The identification information is extracted from the image data to obtain the identification information of the document 12 to be checked, and in step 124, the state of the paper used in the document 12 to be checked is detected, 12 state information is obtained. Note that the processing in steps 120 to 124 is the same as the processing in steps 100 to 104 in the registration processing shown in FIG.
[0079]
In the next step 126, the confirmation processing unit 22 reads from the memory 20 state information registered in correspondence with the identification information acquired in step 122 (hereinafter, referred to as “registered state information”). In step 128, the status information of the document 12 to be confirmed acquired in step 124 (hereinafter, referred to as “detection status information”) is compared with the read registration status information. As a result of the comparison, if the similarity between the two is equal to or greater than the predetermined threshold, the process proceeds from the next step 130 to step 132, where the document 12 to be checked is determined to be the “original” (genuine). In this case, the process proceeds to step 134, where it is determined that the document 12 to be confirmed is a copied or copied “non-original” (fake).
[0080]
Here, it is impossible to detect the state of the paper from exactly the same position of the document 12 at the time of registration processing and at the time of confirmation processing, depending on how the paper 12 is placed. For this reason, either one of the detection state information and the registration state information is cut out, and the area (cutout area) where the state information after one cutout is detected is included in the area where the other state information is detected, The degree of similarity between the state information after one cut-out and the information of the area of the same size as the cut-out area in the other state information, while moving the position of the area of the same size as the cut-out area in the other state information The document to be confirmed may be determined to be the original when the similarity is obtained a plurality of times and the obtained maximum value of the similarity is equal to or greater than a predetermined value.
[0081]
Hereinafter, as an example, a case where a correlation value is obtained as a similarity will be specifically described. The central part of either the state information acquired at the time of registration (registered state information) or the state information acquired at the time of confirmation (detection state information) is cut out to make it smaller than the size of the original state information, and this cut-out is performed. A normalized correlation value is obtained between the state information and the other state information. As described above, the detection state information and the registration state information are a set of detection values of 128 × 128 pixels. For example, a central portion of the registration state information of 64 × 64 pixels is cut out, and its element is set to x as state information X (I, j) (i, j = 0, 1,... 63) is a two-dimensional data array. On the other hand, the detection state information is kept as it is, and the state information Y is a two-dimensional data array whose elements are y (k, l) (k, l = 0, 1,... 127). Between these x (i, j) and y (k, l), the element r (m, n) (m, n = 0,1,..., 64) of the normalized correlation R is calculated by the following equation 1. Ask.
[0082]
(Equation 1)

[0083]
The element r (m, n) of the normalized correlation R obtained in this way indicates that the larger the value is, the higher the similarity is, and the maximum value among the elements r (m, n) of the normalized correlation R is If rmax is equal to or greater than a predetermined threshold, it is determined that the document 12 to be confirmed is the original, and if the maximum value rmax is less than the predetermined threshold, it is determined that the document 12 to be confirmed is not the original. Good.
[0084]
As the similarity between the detection state information and the registration state information, an evaluation value for evaluating various similarities known in the art other than the above-described correlation value can be used. For example, the similarity can be evaluated by calculating the cumulative square error between the detection state information and the registration state information, and the registration state information and the detection state information are quantized and sampled to extract each feature. Then, by treating the extracted features as vectors, the distance between them can be obtained, or the similarity can be evaluated from the angle formed by the vectors. In addition, since the detection state information and the registration state information are image data, the detection state information and the registration state information may be converted into a frequency domain by, for example, two-dimensional Fourier transform, and may be identified and collated in Fourier space. In this case, the two are image-combined on the Fourier space, and a Fourier transform is performed to obtain a correlation strength image, and the similarity between the two can be evaluated from the peak value. For example, when the magnitude of the amplitude peak is equal to or greater than a predetermined threshold, the same image, that is, the document to be confirmed is determined to be the original.
[0085]
It is preferable that the threshold value used in the above-described comparison determination is set with a predetermined allowable range in consideration of the detection error of the state information. That is, the size of the threshold may be appropriately selected according to the user's request for how strict the original document guarantee is. In addition, since the allowable range differs depending on the type of the document 12, that is, the threshold value may be different for each document 12, when the registration state information is registered (step 106 in FIG. 7), the threshold value is stored in the memory together with the identification information. 20.
[0086]
Here, as described above, since four pieces of state information are acquired for each document 12, the above-described comparison determination can be performed four times for each document 12 to be confirmed. Are all "original", the final determination is "original". Otherwise, the final determination is "non-original". However, in actuality, in order to compensate for inconsistency in state information due to contamination of the detection area 36, for example, if a result indicating that two of the four determinations are "original" is given, the "original" is finally used. It may be determined that there is. The criterion for making the final determination result in this manner may be determined in accordance with a user's request for how strict the original guarantee should be.
[0087]
In addition, since an accident such as an operation error or misalignment may occur at the time of collation, a final judgment is made based on a plurality of judgment results, or when the comparison result is not judged to be the original printed matter. The retry may be permitted up to a predetermined number of times.
[0088]
If the encoded data representing the status information is printed on the front side (or the back side) of the document 12, the image reading unit 14 (or a dedicated image reading unit) prints the front side (or the back side) of the document 12. It goes without saying that the data may be read from the data and used as the registration status information.
[0089]
Then, finally, in step 136, the judgment result of the original / non-original is output, and the confirmation processing of FIG. 8 ends.
[0090]
As described above, in the present embodiment, the state information indicating the paper surface state and the thickness unevenness state used for the document 12 is used as the confirmation information, and the paper surface state and the thickness of the regular document 12 (original) are previously determined. By registering the unevenness state and comparing it with the paper surface state and thickness unevenness state detected from the document 12 to be checked, the authenticity (original / non-original) of the document 12 to be checked is determined. It has become. As a result, the characteristics of each document 12 can be captured and whether or not the document 12 is the original can be confirmed.
[0091]
In general, forgery or falsification of the document 12, such as a printed matter, illegally obtains electronic data, which is print information, and falsifies the data. The falsified or falsified data is stored on an image recording medium such as paper by a printer or a copying machine. However, as described above, it is impossible to forge even the surface state and the uneven thickness state by using the paper surface state and the uneven thickness state of the regular document 12 (original) as the confirmation information. Therefore, such spoofing can be easily detected.
[0092]
Further, unlike the related art, there is no need to perform a special printing process on the document 12 to indicate that it is an original or use a special paper, so that it can be easily realized at low cost.
[0093]
In the above description, the case where the identification information assigned to the document 12 is recorded in advance by printing or the like on the document 12 has been described, but the present invention is not limited to this. For example, characters printed or handwritten on the entire paper surface or a predetermined portion of the document 12 are read by OCR, and the code data is used as code data for the number of read characters by using a certain function, for example, a hash function such as MD5. It is also possible to use the hash value obtained as described above as identification information. In this case, it is not possible to guarantee that the identification information is unique, but from experience, it is extremely rare that the same hash value is obtained from different character strings.
[0094]
Further, in the above description, the case where the corresponding registration state information is selected based on the identification information and the detection state information and the registration state information are compared on a one-to-one basis has been described. The detection state information may be compared with all the registered state information without using the identification information. In this case, it is not necessary to extract the identification information. Therefore, the image reading unit 14 and the identification information extraction unit 16 can be omitted.
[0095]
【The invention's effect】
As described above, the present invention can improve the forgery prevention effect by checking whether or not the document is the original by using the surface state and thickness unevenness state of the document, and Since special printing and special paper are not required, there is an excellent effect that it can be easily realized.
[Brief description of the drawings]
FIG. 1 is a configuration diagram showing a basic configuration of a document confirmation device according to an embodiment.
FIG. 2 is a diagram illustrating an example of a document.
FIG. 3 is a perspective view illustrating an example of an image reading apparatus.
FIG. 4 is a cross-sectional view illustrating a state where a document is set in the image reading apparatus of FIG. 3;
FIG. 5 is a perspective view showing a configuration inside a carriage of the image reading apparatus of FIG. 3;
FIG. 6 is a cross-sectional view illustrating an example of a light source provided on a platen cover for detecting thickness unevenness of the image reading apparatus of FIG. 3;
FIG. 7 is a flowchart illustrating a registration process performed by the document confirmation device.
FIG. 8 is a flowchart illustrating a confirmation process performed by the document confirmation device.
[Explanation of symbols]
10 Document confirmation device
12 documents
14 Image reading unit
16 Identification information extraction unit
18 Status detector
20 memories
22 Confirmation processing section
36 detection area
40 Image reading device
44 Platen cover
46 Platen glass
48 carriage
48A opening
50 light sources
52 line image sensor
60 light source
62 line image sensor
64 reflector
66 light source
68 line image sensor

Claims (19)

A document confirmation method for confirming the originality of a document,
In advance, detecting at least one of the surface state and the thickness unevenness state of the document from the original document,
First state information indicating the detected state is stored,
Detecting at least one of the surface state and the thickness unevenness state of the document from the document to be checked,
Comparing the second state information indicating the detected state with the first state information stored in advance,
Based on the comparison result, determine whether the document to be confirmed is the original document,
Document confirmation method characterized by the following. Irradiating the document with light and detecting the surface state by receiving reflected light from the document;
2. The document confirmation method according to claim 1, wherein: Irradiating the document with near-ultraviolet light,
3. The document confirmation method according to claim 2, wherein: Irradiating the document with light and detecting the surface state by receiving transmitted light from the document;
2. The document confirmation method according to claim 1, wherein: Irradiating the document with near-infrared light,
The document confirmation method according to claim 4, wherein: Detecting the state from a predetermined non-image recording portion,
The document confirmation method according to any one of claims 1 to 5, wherein: An area where one of the first state information and the second state information is detected is included in an area where the other is detected,
The similarity between the one state information and the information in the area of the same size as the area in which the one is detected in the other state information is determined a plurality of times while moving the area of the same size in the other state information. Asked,
If the maximum value of the similarity is equal to or greater than a predetermined value, it is determined that the document to be confirmed is the original document.
The document confirmation method according to any one of claims 1 to 6, wherein: Acquiring a plurality of the first state information from the original document;
Acquiring a plurality of the second state information from the document to be confirmed, in correspondence with each of the plurality of the first state information;
Comparing each of the plurality of second state information with a corresponding plurality of the first state information, performing the determination a plurality of times,
Based on the determination results of the plurality of times, a final determination whether the document to be confirmed is the original document,
The document confirmation method according to any one of claims 1 to 6, wherein: From the original document, obtain identification information for identifying the document,
The first state information is stored in association with the identification information acquired from the original document,
Acquiring the identification information from the document to be confirmed,
Comparing the first state information and the second state information associated with identification information that matches the identification information acquired from the document to be confirmed,
9. The document confirmation method according to claim 1, wherein: A document checking device for checking the originality of a document,
State detecting means for detecting at least one of a surface state and a thickness unevenness state in the document from the document;
Storage means for storing first state information indicating the state detected from the original document by the state detection means;
The second state information indicating the state detected from the document to be confirmed by the state detection unit is compared with the first state information stored by the storage unit, and based on the comparison result, Determining means for determining whether the document to be confirmed is the original,
A document confirmation device comprising: The state detection means,
A light source for irradiating the document with light,
Light receiving means for receiving reflected light or transmitted light from the document of light emitted to the document by the light source,
The document confirmation device according to claim 10, further comprising: The light source emits near-ultraviolet light,
The light receiving means receives the reflected light,
The document confirmation device according to claim 11, wherein: The light source emits near-infrared light,
The light receiving means receives the transmitted light,
The document confirmation device according to claim 11, wherein: The state detection unit further includes a diffusion unit that diffuses light applied to the document,
The document confirmation device according to claim 13, wherein The light source is an electroluminescent element,
The document confirmation device according to claim 13, wherein The state detecting means detects the state from a predetermined non-image recording portion of the document,
The document confirmation device according to claim 10, wherein The determination means is formed such that an area where one of the first state information and the second state information is detected is included in an area where the other is detected,
The similarity between the one state information and the information in the area of the same size as the area in which the one is detected in the other state information is determined a plurality of times while moving the area of the same size in the other state information. Asked,
If the maximum value of the similarity is equal to or greater than a predetermined value, it is determined that the document to be confirmed is the original document.
The document confirmation device according to claim 10, wherein The state detecting means detects the state a plurality of times from the document,
The determination unit compares each of the plurality of second state information obtained by the plurality of state detections with a corresponding plurality of the first state information, performs the determination a plurality of times, and performs the determination a plurality of times. Based on the determination result of the final determination whether or not the document to be confirmed is the original
The document confirmation device according to any one of claims 10 to 17, wherein: From the document, further comprising identification information acquisition means for acquiring identification information for identifying the document,
The storage unit stores the first state information in association with the identification information acquired from the original document by the identification information acquisition unit,
The determination unit compares the first state information and the second state information associated with identification information that matches the identification information acquired from the document to be confirmed by the identification information acquisition unit.
19. The document confirmation device according to claim 10, wherein

Images