TWM626684U - Document proofreading device - Google Patents

Abstract

M

N)個框選的文字是唯一的之第一文字框配對到M個框選的文字是唯一的之第二文字框，並取得M個第一文字框與M個第二文字框的四個角點及中點位置的座標，且根據M個第一文字框的該等座標和M個第二文字框的該等座標的對應關係計算一座標轉換矩陣，並利用座標轉換矩陣將該等第一文字框投影到該待驗文件影像檔案中，再比對原始文件檔案之被該第一文字框框選的文字與待驗文件影像檔案之被該第一文字框框選的文字。 A document proofreading device, which obtains characters framed by a plurality of first character frames in an original document file and characters framed by a plurality of second character frames in a document image file to be checked, and converts M(2

M

N) The first text box where the selected text is unique is paired to the second text box where the text selected in the M box is unique, and the four corner points of the M first text boxes and the M second text boxes are obtained. and the coordinates of the midpoint position, and calculate a coordinate transformation matrix according to the correspondence between the coordinates of the M first text boxes and the coordinates of the M second text boxes, and use the coordinate transformation matrix to project the first text boxes. In the image file of the document to be checked, the text framed by the first text frame in the original document file is compared with the character framed by the first character frame in the image file of the document to be checked.

Description

document proofing device

The new model relates to a proofreading device, in particular to a document proofreading device for comparing the differences between two documents.

In banking, insurance and other industries, documents signed by customers such as contracts and consents often require manual verification and confirmation, which often consumes a lot of time. To find out the differences between the original document file and the image file generated by scanning or shooting, in addition to manual proofreading, an existing method is to first use a deep learning model trained by artificial intelligence to detect the text in the image file and then A text image selected by the text box is generated, and then the text image selected by the text box is recognized, and then the recognized whole text is compared with the original document file. This approach can indeed get results quickly, but the images generated after scanning or shooting the original document file often have displacement, rotation, noise, and handwritten text/signature, etc., which is easy to make the deep learning model unable to detect and frame the text Selecting some text images will affect the accuracy of text recognition.

Therefore, the purpose of the present invention is to provide a document proofreading device, which can accurately find out the text in the original document file and its generated text after scanning or photographing. The correspondence between the text in the image file to accurately compare the differences between the two documents.

2且N為正整數)個第一文字框所框選的文字是唯一的時，根據該N個第一文字框配對到該待驗文件影像檔案中框選唯一出現的文字的M(2

M

N且M為正整數)個第二文字框，並取得配對的M個第一文字框與M個第二文字框的四個角點及其中點位置的座標；該轉換矩陣產生模組根據配對的M個第一文字框的該等座標以及M個第二文字框的該等座標之間的對應關係，計算一座標轉換矩陣；該文字框投影模組根據該座標轉換矩陣，將該原始文件檔案中的該等第一文字框投影到該待驗文件影像檔案中，並取得該待驗文件影像檔案中被各該第一文字框框選的文字；該比對模組比對該原始文件檔案之被各該第一文字框框選的文字與該待驗文件影像檔案之被各該第一文字框的文字資訊，並輸出 一比對結果。 Therefore, the novel document proofreading device is used for proofreading an original document file and an image file of a document to be checked, and includes a storage unit and a processing unit, and the storage unit stores the original document file and the image file of the document to be checked; The processing unit can access the storage unit and includes a text capture module, a text frame matching module, a conversion matrix generation module, a text frame projection module and a comparison module; wherein, the text capture module Obtain a plurality of first text boxes and the text framed by each of the first text frames from the original document file, and obtain a plurality of second text frames and the text framed by each of the second text frames from the image file of the document to be checked ; The text box matching module determines that at least N(N

2 and N is a positive integer) When the text framed by the first text boxes is unique, the M(2

M

N and M are positive integers) second text boxes, and obtain the coordinates of the four corner points and the midpoints of the paired M first text boxes and the M second text boxes; the conversion matrix generation module is based on the paired coordinates The correspondence between the coordinates of the M first text boxes and the coordinates of the M second text boxes calculates a coordinate transformation matrix; the text box projection module converts the original file into the original file according to the coordinate transformation matrix. The first text frames of the original document are projected into the image file of the document to be checked, and the characters selected by the first text frames in the image file of the document to be checked are obtained; the comparison module compares the original document file by the respective The text selected in the first text box is compared with the text information of the image file of the document to be checked by each of the first text boxes, and a comparison result is output.

In some embodiments of the present invention, when the text box matching module determines that the text framed by less than N first text boxes in the original document file is unique, it records the framed text in the original document file The same first text boxes and their number, and record the second text boxes and their number with the same framed text in the image file of the document to be checked, and the same framed text in the original document file. Match the first text box with the second text boxes with the same framed text and the same number as the first text boxes in the image file of the document to be checked, and match the framed text in the original document file with the same framed text. Wait for the first text frame to be framed by a first rectangular frame, and obtain at least two coordinates of the four corners of the first rectangular frame and the position of the midpoint, and the framed text in the image file of the document to be checked is the same The second text boxes are framed by a second rectangular frame, and the coordinates of the four corners and the midpoints of at least two of the second rectangular frames paired with at least two of the first rectangular frames are obtained; the conversion The matrix generation module calculates the coordinate transformation matrix according to the corresponding relationship between the coordinates of the paired at least two first rectangular frames and the coordinates of at least two second rectangular frames.

In some implementation aspects of the present invention, the text capture module further obtains the number of pages of the original document file and the image file of the document to be checked, and the text capture module determines the number of pages of the original document file and the After the number of pages in the image file of the document to be checked is the same, the first text boxes and the characters selected by each of the first text frames are obtained from the original document file, and the second characters are obtained from the image file of the document to be checked framed with and the text framed by each of the second text boxes.

In some embodiments of the present invention, the coordinate transformation matrix is a homography matrix.

The effect of the present invention is: by projecting the first text frames in the original document file into the image file of the document to be checked, and then texting the content framed by the first text frames in the image file of the document to be checked Identification can solve the problem that some characters in the image file of the document to be checked cannot be detected due to displacement, rotation, noise, or handwritten characters/signatures/alterations.

1, 4: Original file archives

11, 11', 41, 41': the first text box

42: The first rectangular frame

2, 5: Image files of documents to be inspected

21, 21', 51, 51': the second text box

52: Second rectangular frame

3: Document proofreading device

31: Storage unit

32: Processing unit

321: Text Capture Module

322: Text box matching module

323: Conversion matrix generation module

324: Text box projection module

325: Comparison Module

S1~S6, S3', S4': Steps

Other features and effects of the present invention will be clearly shown in the embodiments with reference to the drawings, wherein: FIG. 1 is a schematic block diagram of hardware components and modules included in an embodiment of the document proofreading device of the present invention; FIG. 2 is a A schematic diagram of an original document file; FIG. 3 is a schematic diagram of an image file of a document to be checked; FIG. 4 is the main flow of document proofreading in the present embodiment; A schematic diagram of the text; FIG. 6 is a schematic diagram of the detected text in the image file of the document to be checked with a second text frame; 7 is a schematic diagram illustrating the pairing of part of the first text frame and part of the second text frame shown in FIG. 6; Schematic diagram; FIG. 9 is a schematic diagram illustrating the projection of the first text frames in the original document file into the image file of the document to be checked; FIG. 10 illustrates that the coordinates of the first text frame are converted into projections to the to-be-checked document by a coordinate transformation matrix. Coordinates in the document image file; FIG. 11 is a schematic diagram illustrating the pairing relationship between the first text frame in the original document file and the second text frame in the document image file to be checked; FIG. 12 illustrates the frame selection in the original document file The first text boxes of the same text are framed by a first rectangular frame, and the second text boxes of the same text in the image file of the document to be checked are framed by a second rectangular frame; and FIG. 13 is a description of obtaining the A schematic diagram of the coordinates of the positions of the four corners and the midpoints of the first text frame and the second text frame; and FIG. 14 is a diagram illustrating the projection of the first text frames in the original document file to the image file of the document to be checked Schematic in .

Before the present invention is described in detail, it should be noted that in the following description, similar elements are designated by the same reference numerals.

Referring to FIG. 1, it is a block diagram of the main components of an embodiment of the novel document proofreading device, which is used for proofreading an original document file 1 shown in FIG. 2 and an image file of a document to be checked as shown in FIG. 3 2, wherein, the image file 2 of the document to be verified is generated by scanning or photographing the paper copy of the original document file 1; and the document proofreading device 3 of this embodiment realizes the document by executing the process shown in FIG. 4 For proofreading, the document proofreading device 3 is a computer device, and as shown in FIG. 1 , it mainly includes a storage unit 31 (such as a built-in, installed or external memory module in the computer), a storage unit capable of accessing the The processing unit 32 (eg, central processing unit) of 31 and other related components and the like not shown in FIG. 1 . The storage unit 31 stores or temporarily stores the original document file 1 and the image file 2 of the document to be checked; the processing unit 32 is preloaded with a computer-readable recording medium (for example, the storage unit 31) A program read and executable by the processing unit 32, the program includes a text capture module 321, a text frame matching module 322, a conversion matrix generation module 323, a text frame projection module 324 and a ratio to module 325.

Thus, as shown in step S1 in FIG. 4 , when the processing unit 32 executes the program, the text capture module 321 reads the original document file 1 from the storage unit 31 and detects whether the content of the original document file 1 has text, and the detected text is framed by the first text box 11 to obtain a plurality of first text boxes 11 and the content 12 framed by each of the first text boxes 11 from the original document file 1, as shown in FIG. 5 Specifically, the text capture module 321 determines that the original document file 1 is a PDF When formatting, text detection software, such as but not limited to pdfminer and other tools, will be used to select the detected text box with the text box and read the content selected by the text box, that is, the text information; and if the text is extracted When the module 321 determines that the original document file 1 is in a document format such as DOC/ODF, it converts or converts the original document file 1 into a PDF format file, and then uses the above text detection software to detect the original document file 1. and select the text in the text box and read the content selected by the text box, that is, the text information.

And if the text capture module 321 determines that the original document file 1 is an image file, it uses a text detection model trained in advance through deep learning and used to detect text, such as but not limited to RCNN (Region-based Convolutional Neural Networks, region-based convolutional neural network) or YOLO (You Only Look Once) and other deep learning models, perform text detection on the original file file 1 and select the detected text in a text box. Then, the text capture module 321 uses a text recognition model trained in advance through deep learning and used to recognize text to perform text recognition on the content framed by the first text box 11 in the original document file 1 to obtain The content framed by the first text boxes 11, that is, text information, such as Cat, Dog in the first row, Fish, King in the second row, Dog, Egg in the third row, Dog, Egg...etc.

At the same time, as shown in FIG. 6 , the text capture module 321 uses the above-mentioned text detection model to detect the text in the image file 2 of the document to be checked to generate plural numbers A second text box 21 is formed, and then the text recognition model is used to perform text recognition on the content framed by the second text frame 21 in the image file 2 of the document to be checked, so as to obtain the content framed by the second text frame 21, That is, text information, such as Cat and Dog in the first column of FIG. 6 , King in the second column, Dog and Egg in the third column, Egg in the fourth column, etc.; The original document file 1 is generated by scanning or photographing after the handwritten text/signature/alteration, etc., so the image content may appear displacement, rotation, noise, and handwritten text/signature/alteration, etc., which will affect the text detection. The model detects the text and finds the text box. For example, when a cross slash (or noise) appears on the "Dog" in the fourth column of the image file 2 of the document to be checked, the text detection model will misjudge that there is no text there The image "Dog" is not generated in a text box, and because only the content selected by the second text box 21 is input into the text recognition model for text recognition, the word "Dog" is not recognized.

It is worth mentioning that, before performing the above step S1, the text capture module 321 can also obtain the page numbers of the original document file 1 and the image file 2 of the document to be checked, and the text capture module 321 determines the Step S1 is performed only after the number of pages of the original document file 1 is the same as the number of pages of the image file 2 of the document to be checked. Otherwise, it is determined that the two files may not be related, and step S1 is not performed and a file error message is output.

2且N為正整數)個第一文字框11所框選的文字在該原始文件檔案1中 是唯一的，若是，例如圖7所示，其中有五個第一文字框11’所框選的文字在該原始文件檔案1中是唯一出現的，即Cat、Fish、King、Car和Apple，則如圖4的步驟S3，該文字框配對模組322根據該五個第一文字框11’配對到該待驗文件影像檔案2中框選唯一出現的文字(例如圖7中的Cat、King、Car和Apple)的M(2

M

N)個第二文字框21’，在此M等於4，即該待驗文件影像檔案2中有四個第二文字框21’框選唯一出現的文字且所框選的文字與該五個第一文字框11’其中的四個第一文字框11’框選的文字配對；值得一提的是，該待驗文件影像檔案2中框選唯一出現的文字的第二文字框21’的數量不一定會與第一文字框11’相同(如同上述)，不過只要該待驗文件影像檔案2中框選唯一出現的文字的第二文字框21’的數量至少兩個且所框選的文字和該五個第一文字框11’所框選的文字其中至少兩個配對即可。 Next, the processing unit 32 executes step S2 in FIG. 4 , so that the text box matching module 322 determines whether the original document file 1 has at least N(N

2 and N is a positive integer) the text framed by the first text boxes 11 is unique in the original document file 1. If, for example, as shown in FIG. 7, there are five text frames selected by the first text boxes 11'. The original document file 1 is the only one that appears, namely Cat, Fish, King, Car and Apple, as shown in step S3 in FIG. 4 , the text box matching module 322 is paired with the five first text boxes 11 ′ In the image file 2 of the document to be checked, select the M(2) of the only text (such as Cat, King, Car and Apple in Figure 7)

M

N) second text boxes 21 ′, where M is equal to 4, that is, there are four second text boxes 21 ′ in the image file 2 of the document to be checked to select the only text that appears, and the selected text is the same as the five The four first text boxes 11 ′ in the first text box 11 ′ are frame-selected text pairs; it is worth mentioning that the number of the second text boxes 21 ′ which frame the only text appearing in the image file 2 of the document to be checked is not equal. It must be the same as the first text box 11' (as described above), but as long as the number of the second text boxes 21' in which the only text that appears in the image file 2 of the document to be checked is framed is at least two, and the framed text is the same as the At least two of the texts selected by the five first text boxes 11' may be matched.

Then, as shown in FIG. 8 , the text box pairing module 322 obtains the four corner points and the midpoint positions of the paired four first text boxes 11 ′ and four second text boxes 21 ′ (black in FIG. 8 ) point) coordinates; and the coordinates described here refer to the pixel positions on the original document file 1 and the image file 2 of the document to be checked.

Next, as shown in step S4 in FIG. 4 , the processing unit 32 makes the conversion matrix generation module 323 according to the coordinates of the paired four first text boxes 11 ′ (25 coordinates in total) and the four second text boxes 21 ′ between these coordinates (a total of 25 coordinates) of Corresponding relationship, calculate a coordinate transformation matrix for projecting the first text boxes 11 in the original document file 1 to the image file 2 of the document to be checked, and in this embodiment, the coordinate transformation matrix can be Not limited to the Homography matrix.

Then, as shown in step S5 in FIG. 4 , the processing unit 32 instructs the text frame projection module 324 to project the first text frames 11 in the original document file 1 to the image file 2 of the document to be checked according to the coordinate transformation matrix , as shown in FIG. 9 ; for example, as shown in FIG. 10 , taking the first text box 11 of “Cat” in the first column of the original document file 1 as an example, the four text boxes of the first text box 11 After the coordinates (3,3), (6,3), (3,5), (6,5) of the corner points are multiplied by the coordinate conversion matrix H, the coordinates of the four corner points of the first text box 11 will be converted (2,2), (5,2), (2,4), (5,4), and when projected to the image file 2 of the document to be checked, the first text box 11 will just frame the document to be checked "Cat" in Image File 2.

Therefore, even if the “Dog” with noise (cross-hatched) in the fourth row of the image file 2 of the document to be checked was not detected and not selected by the second text box 21, the above-mentioned In the step of projecting all the first text boxes 11 in the original document file 1 to the image file 2 of the document to be checked, the fourth row of the image file 2 of the document to be checked is not detected but is not detected in the original document file 1 The "Dog" text in the fourth column of the In this case, the defect that the text cannot be accurately detected.

Then, the text box projection module 324 uses the above-mentioned text recognition model to perform text recognition on the content framed by the first text boxes 11 in the image file 2 of the document to be checked, so as to obtain the image file 2 of the document to be checked The text information framed by each of the first text boxes 11 in the .

Finally, as shown in step S6 in FIG. 4 , the processing unit 32 enables the comparison module 325 to compare the text framed by the first text boxes 11 of the original document file 1 with the text of the image file 2 of the document to be checked. The first text box 11 frames the selected text, and outputs a comparison result. The comparison result can be outputting or marking the content of the original document file 1 and the document file 2 to be checked for differences, or further judging the two documents. Whether the difference between the two documents is greater than a set threshold value, if so, it is determined that there are too many differences between the two documents, the two may not be related, and an error message is output.

In addition, returning to the above-mentioned step S2, when the text box matching module 322 determines that the text framed by less than N first text boxes 41 in the original document file 4 as shown in FIG. 11 is unique, then Step S3 ′ in FIG. 4 is executed, and the text box matching module 322 records the first text boxes 41 with the same text boxed in the original document file 4 and the number thereof (for example, the first text box 41 of the box-selected "Cat" has Two, there are two first text boxes 41 for box selection "Dog", two first text boxes 41 for box selection "Fish", three first text boxes 41 for box selection "Egg", and three first text boxes 41 for box selection "King" There is one first text box 41, and there are three first text boxes 41 for “Apple”, and “Car” is selected. There are two first text boxes 41 ), and record the second text boxes 51 and their numbers with the same text box-selected in the image file 5 of the document to be checked as shown in FIG. 11 (for example, box-selected “Cat” There is one second text box 51, two second text boxes 51 for selecting “Dog”, two second text boxes 51 for selecting “Fish”, and two second text boxes 51 for selecting “Egg” Two, there is one second text box 51 for box-selecting "King", two second text boxes 51 for box-selecting "Apple", and two second text boxes 51 for box-selecting "Car").

Then, the text box matching module 322 has the same first text boxes 41 with the same text framed in the original document file 4 as the framed text in the image file 5 of the document to be checked, and the number is the same as the first text boxes 41 are paired with the same second text boxes 51. For example, as shown in FIG. 11, the first text boxes 41' of "Dog", "Fish", "King", and "Car" in the original document file 4 are selected. It can be paired with the second text boxes 51 ′ of “Dog”, “Fish”, “King”, and “Car” in the image file 5 of the document to be checked; and as shown in FIG. The group 322 has the same text framed in the original document file 4 and the first text frames 41 ′ that are paired with the second text frames 51 ′ in the image file 5 of the document to be checked are framed by a first rectangular frame 42 . Wherein, that is, the first rectangular frame 42 contains all the same characters therein; similarly, the character frame matching module 322 also frames the selected characters in the image file 5 of the document to be checked and is the same as the original document The second text boxes 51 ′ paired with the first text boxes 41 ′ in the file 4 are framed by a second rectangular frame 52 , that is, a second rectangular frame The box 52 contains all the same text in it.

Then, as shown in FIG. 13 , the text box matching module 322 obtains the positions of the four corners and the midpoints of the first rectangular frames 42 (four in total) and the second rectangular frames 52 (four in total). (the black dots in FIG. 13 ), then, the processing unit 32 executes step S4 ′ in FIG. 4 , so that the conversion matrix generation module 323 is based on the coordinates of the first rectangular frames 42 and the For the correspondence between the coordinates of the second rectangular frame 52, the coordinate transformation matrix is calculated. It is worth mentioning that the text box matching module 322 can calculate the coordinates without taking all the coordinates of the four corners of the first rectangular frame 42 and all of the second rectangular frames 52 and the positions of their midpoints. The transformation matrix, that is, it is also possible to only take at least two of the first rectangular frames 42 and at least two second rectangular frames 52 paired with the at least two first rectangular frames 42 The four corners and their midpoint positions coordinates to calculate the coordinate transformation matrix.

Next, the above-mentioned step S5 is performed, and the text frame projection module 324 projects the first text frames 41 in the original document file 4 to the image file 5 of the document to be checked according to the coordinate transformation matrix calculated in step S4 ′ 14, whereby the text “Egg” in the third column of the image file 5 of the document to be checked that was not originally framed by the second text box 51 can be framed by the first text box 41, Therefore, when the content framed by the first text boxes 41 in the image file 5 of the document to be verified is input into the text recognition model for recognition, the text that was originally missed in the third row of the image file 5 of the document to be verified "Egg" can be recognized by the text recognition model.

Finally, the above-mentioned step S6 is performed, and the processing unit 32 makes the comparison module 325 compare the text information of the original document file 4 framed by each of the first text boxes 41 with the image file 5 of the document to be checked. The text information framed by each of the first text boxes 41 is outputted as a comparison result.

To sum up, since the content in the original document files 1 and 4 is usually clearer and more accurate than the content in the image files 2 and 5 of the document to be inspected generated after the original document files 1 and 4 are scanned or photographed, the corresponding Compared with the image files 2 and 5 of the document to be checked, the text content in the original document files 1 and 4 can be read directly (for example, the original document files 1 and 4 are PDF files) or can be accurately detected by text detection software. Detect (for example, the original document files 1, 4 are image files) and select them with text boxes; therefore, in the above embodiment, the first text boxes 11, 41 in the original document files 1, 4 are projected to the to-be-to-be Check the document image files 2, 5, and then perform text recognition on the contents framed by the first text boxes 11, 41 in the document image files 2, 5 to be checked, and solve the problem in the document image files 2, 5. Some characters cannot be accurately detected due to displacement, rotation, noise, or handwritten characters/signatures/alterations, etc., the functions and purposes of the new model are indeed achieved.

However, the above are only examples of the present invention, which should not limit the scope of the present invention. Any simple equivalent changes and modifications made according to the scope of the patent application for this new model and the contents of the patent specification are still within the scope of the present invention. within the scope of this new patent.

1, 4: Original file archives

2, 5: Image files of documents to be inspected

3: Document proofreading device

31: Storage unit

32: Processing unit

321: Text Capture Module

322: Text box matching module

323: Conversion matrix generation module

324: Text box projection module

325: Comparison Module

Claims (4)

A document proofreading device is used for proofreading an original document file and an image file of a document to be checked, and includes: a storage unit, in which the original document file and the image file of the document to be checked are stored; and a processing unit capable of accessing the storage unit and comprising a text capturing module, a text frame matching module, a conversion matrix generating module, a text frame projecting module and a matching module; wherein The text extraction module obtains a plurality of first text boxes and the text selected by each of the first text boxes from the original document file, and obtains a plurality of second text boxes and each of the first text boxes from the image file of the document to be checked. Two text boxes to select the text; When the text box matching module determines that the text selected by at least N (N≥2 and N is a positive integer) first text boxes in the original document file is unique, it is matched to the pending test according to the N first text boxes Select M (2≤M≤N and M is a positive integer) second text boxes of the only text in the document image file, and obtain the four corners of the matching M first text boxes and M second text boxes the coordinates of the position of the point and its midpoint; The conversion matrix generation module calculates a coordinate conversion matrix according to the correspondence between the coordinates of the paired M first text boxes and the coordinates of the M second text boxes; The text box projection module projects the first text boxes in the original document file into the image file of the document to be checked according to the coordinate transformation matrix, and obtains the image files of the document to be checked that are selected by the first text frames Text; The comparison module compares the text selected by each of the first text boxes of the original document file with the text information of the image file to be checked by each of the first text boxes, and outputs a comparison result. The document proofreading device according to claim 1, wherein when the text box matching module determines that the text framed by less than N first text boxes in the original document file is unique, it records the frame selection in the original document file the first text boxes with the same text and their number, and record the second text boxes and their number with the same text boxed in the image file of the document to be checked, and the same text boxed in the original document file The first text boxes in the document image file to be checked are matched with the second text boxes with the same framed text and the same number as the first text boxes, and the framed text in the original document file is the same The first text boxes are framed by a first rectangular frame, and the coordinates of the four corner points and the midpoint of at least two of the first rectangular frames are obtained, and the text framed in the image file of the document to be checked is framed The same second text frames are framed by a second rectangular frame, and the coordinates of the four corners and the midpoint of at least two of the second rectangular frames paired with at least two of the first rectangular frames are obtained; The transformation matrix generating module calculates the coordinate transformation matrix according to the corresponding relationship between the coordinates of the paired at least two first rectangular frames and the coordinates of at least two second rectangular frames. The document proofreading device according to claim 1 or 2, wherein the text capture module further obtains the page numbers of the original document file and the image file of the document to be checked, and the text capture module determines the Only after the number of pages is the same as the number of pages of the image file of the document to be checked, the first text boxes and the text information selected by each of the first text boxes are obtained from the original document file, and obtained from the image file of the document to be checked the second text boxes and text information framed by each of the second text boxes. The document proofreading apparatus according to claim 1 or 2, wherein the coordinate transformation matrix is a homography matrix.

Images