JP5706556B2 - Overhead scanner device, image acquisition method, and program - Google Patents

Description

  The present invention relates to an overhead scanner device, an image acquisition method, and a program.

  Conventionally, when a scanner scans a plurality of documents such as a book having a large number of pages, the user opens a target page of the document, places the document on a reading table, and presses a scanning start button. Repeatedly.

  Here, Patent Document 1 discloses that an apparatus having both an area sensor and a line sensor recognizes a reading range and a document size from an image pre-scanned by the area sensor and reads a document by a linear sensor. Yes.

  Non-Patent Documents 1 and 2 disclose that a scan image is formed by correcting a frame of a moving image captured by a high-speed area sensor.

JP 2005-167934 A

Ishikawa Komuro Laboratory, “Book Flipping Scanning”, [online], publication date unknown, Ishikawa Komuro Laboratory, [May 25, 2010 search], Internet <URL: http: // www. k2. t. u-tokyo. ac. jp / vision / BookFlipScan / index-j. html> Erico Guizzo, “Superfast Scanner Let's You Digitize a Book By Rapidly Flip Pages”, [online], March 17, 2010, IEEE Spec. spectrum. iee. org / automaton / robotics / robotics-software / book-flipping-scanning>

  However, the conventional scanner has problems such as a complicated operation for reading an image of a document and a low definition of the image.

  For example, although the scanner described in Patent Document 1 can recognize the document size, offset, and the like by pre-scanning with an area sensor, since it is a flatbed scanner, in order to read a plurality of documents, a document pressure plate It is necessary to repeatedly perform operations such as placing a document with pages open and turning the document downward and closing the document pressing plate, which has a problem that the operation is complicated.

  Further, in the scanners described in Non-Patent Documents 1 and 2, a scan image can be acquired from a moving image by an overhead type high-speed area sensor, but expensive equipment must be used to read a document with high definition. Had a point.

  The present invention has been made in view of the above problems, and provides an overhead scanner device, an image acquisition method, and a program that are excellent in document readability and can acquire a high-definition image. Objective.

  In order to achieve such an object, the overhead scanner device of the present invention includes an area sensor, a linear sensor, and a control unit, and the control unit controls the area sensor to continuously acquire a plurality of images. Area image acquisition means, feature point extraction means for extracting feature points from the plurality of images acquired by the area image acquisition means, and the feature points extracted by the feature point extraction means Speed vector calculation means for calculating the speed vector of the feature point by comparing between them, and reading start determination for determining the reading start by the linear sensor based on the speed vector calculated by the speed vector calculation means Means.

  In the overhead scanner device according to the present invention, in the overhead scanner device described above, the reading start determination unit may determine the reading start by the linear sensor when the movement based on the velocity vector stops. Features.

  The overhead scanner device according to the present invention may further include a storage unit in the overhead scanner device described above, and the control unit may be configured to generate a document page based on the speed vector calculated by the speed vector calculating unit. A page direction determining unit that determines a turning direction; a linear image acquiring unit that controls the linear sensor to acquire an image of the document when the reading start is determined by the reading start determining unit; and the linear image Image division means for dividing the image acquired by the acquisition means, and the image divided by the image division means are rearranged according to the page turning direction determined by the page direction determination means and stored in the storage unit And a page rearranging means.

  The overhead scanner device according to the present invention is the overhead scanner device described above, wherein the image dividing unit calculates a boundary line between the document and the background from a gradation difference of the image acquired by the linear image acquiring unit. An inflection point is obtained on each of the upper and lower sides of the boundary line, and the image is divided into right and left by a line connecting the inflection points.

  Further, the overhead scanner device according to the present invention is the overhead scanner device described above, wherein the control unit further includes a curvature correcting unit that corrects a curvature distortion of the image divided by the image dividing unit, The page rearrangement unit rearranges the images corrected by the curvature correction unit according to the page turning direction determined by the page direction determination unit, and stores them in the storage unit.

  Moreover, the overhead scanner device according to the present invention is the overhead scanner device described above, wherein the control unit further includes a skin color detection unit that detects a skin color partial region from the image divided by the image division unit. The page rearrangement unit deletes the image in which the partial area of the skin color is detected by the skin color detection unit without storing the image in the storage unit.

  Further, the overhead scanner device according to the present invention provides a circuit for the linear sensor in the overhead scanner device described above, in the case where there is a movement based on the velocity vector when the reading start by the linear sensor is not determined. The linear sensor circuit is stopped when it is activated and there is no movement based on the velocity vector.

  Moreover, the overhead scanner device according to the present invention further includes a light source in the overhead scanner device described above, and the control unit controls the light source to indicate a controllable state or a reading range of the linear sensor. The light source control means for emitting light is further provided.

  The overhead scanner device according to the present invention is the overhead scanner device described above, wherein the control unit is surrounded by the movement of the feature points based on the history of the velocity vectors calculated by the velocity vector calculating means. An image cutout means for cutting out the image read by the linear sensor within the cutout range is further provided.

  The present invention also relates to an image acquisition method. The image acquisition method according to the present invention includes an area sensor, a linear sensor, and a control unit, and the area sensor is executed by the control unit of an overhead scanner device. An area image acquisition step for acquiring a plurality of images in a controlled manner, a feature point extraction step for extracting feature points from the plurality of images acquired in the area image acquisition step, and the feature point extraction By comparing the feature points extracted in the step between the plurality of images, the speed vector calculation step for calculating the speed vector of the feature points, and the speed vector calculated in the speed vector calculation step And a reading start determination step for determining the reading start by the linear sensor.

  The present invention also relates to a program. The program according to the present invention includes an area sensor, a linear sensor, and a control unit, and controls the area sensor to be executed by the control unit of an overhead scanner device. In the area image acquisition step of continuously acquiring a plurality of images, the feature point extraction step of extracting feature points from the plurality of images acquired in the area image acquisition step, and the feature point extraction step Based on the speed vector calculation step for calculating the speed vector of the feature point by comparing the extracted feature points between the plurality of images, and the speed vector calculated in the speed vector calculation step, A reading start determination step of determining reading start by the linear sensor.

  The present invention also relates to a recording medium, and the computer-readable recording medium according to the present invention records the above-described program according to the present invention.

  According to the present invention, the control unit controls the area sensor to continuously acquire a plurality of images, extracts feature points from the acquired plurality of images, and extracts the extracted feature points to the plurality of images. The speed vector of the feature point is calculated by comparing them, and the start of reading by the linear sensor is determined based on the calculated speed vector. As a result, the operability of reading a document is excellent, and a high-definition image can be acquired. For example, a conventional scanner has to be controlled from a button incorporated in the apparatus or a personal computer connected to the scanner, and this has led to a decrease in productivity especially in the case of processing a plurality of documents. According to the present invention, it is possible to acquire a high-definition document image without requiring a special button or tool for starting document reading.

  According to the present invention, the control unit determines the start of reading by the linear sensor when the motion based on the velocity vector stops. As a result, reading is automatically started at the time when the setting of the original such as turning the page or pressing the original with a hand is completed, so that an excellent operability can be exhibited.

  Further, according to the present invention, the control unit determines the page turning direction of the document based on the calculated velocity vector, and controls the linear sensor to acquire the image of the document when the reading start is determined. Then, the acquired image is divided, and the divided images are rearranged according to the determined page turning direction and stored in the storage unit. As a result, it is possible to determine the page order based on the movement caused by the page turning and rearrange the images divided into page units in order of page, so that editing operations such as dividing and rearranging the read images on the computer are unnecessary. This has the effect of significantly increasing productivity. For example, when scanning a document such as a book having a plurality of pages, conventionally, the user 1) turns the page, 2) presses the scan start button, 3) waits for completion of the scan, and 4) loads the page. It was necessary to perform operations such as dividing into left and right, 5) correcting the curvature of the page, and 6) rearranging and saving in the order of the page. According to the present invention, the user operates the scanner. Instead, it is possible to digitize originals such as books by simply flipping through the books one after another. As a result, there is no need to bother the user with scanner control, so that labor can be saved and productivity can be improved. There is an effect.

  Further, according to the present invention, the boundary line between the document and the background is obtained from the gradation difference of the acquired images, the inflection points are obtained on the upper and lower sides of the boundary line, and the image is moved to the left and right by the line connecting the inflection points. Divide into Thereby, when it is composed of left and right spread pages such as a book, there is an effect that an image can be accurately divided in page units.

  Further, according to the present invention, the control unit corrects the curvature distortion of the divided image, rearranges the corrected image according to the determined page turning direction, and stores the rearranged image in the storage unit. Accordingly, it is possible to correct a bending distortion that occurs when a plurality of sheets of paper are bound on one side, such as a book, and to obtain an image in units of pages without distortion.

  Further, according to the present invention, the control unit detects a skin color partial area from the divided image, and deletes the image in which the skin color partial area is detected without storing it in the storage unit. As a result, when the user turns down an unnecessary page with his / her hand, the area of the hand on the image is detected based on the skin color, and an original image that does not include the unnecessary page can be acquired. Play.

  Further, according to the present invention, when the reading start by the linear sensor is not determined, the linear sensor circuit is activated when there is movement based on the velocity vector, and when there is no movement based on the velocity vector, Stop the circuit. As a result, when reading by the linear sensor is not required, such as when no document is placed, the linear sensor can be put in a standby state to save power.

  According to the invention, the control unit controls the light source to emit light indicating the controllable state or reading range of the linear sensor. Thereby, it is possible to notify the user that the linear sensor is ready to be read and the reading range of the linear sensor before reading.

  Further, according to the present invention, the control unit cuts out the image read by the linear sensor in the cutout range surrounded by the movement of the feature points based on the calculated velocity vector history. Thereby, when a user traces the outer periphery of the range which he wants to cut out with a fingertip, there exists an effect that an image can be cut out correctly according to a user's intention.

FIG. 1 is a block diagram illustrating an example of the configuration of the overhead scanner device 100. FIG. 2 is a diagram illustrating an example of the appearance of an apparatus in which the linear imaging unit 110 and the area imaging unit 120 are integrated, and is a diagram illustrating a relationship among the main scanning direction, the sub-scanning direction, and the rotation direction by the motor 12. FIG. 3 illustrates a state in which the linear imaging unit 110 is viewed from the side, and illustrates an example of light irradiation from the line light source 140 under the control of the light source control unit 102k. FIG. 4 is a flowchart showing an example of basic processing in the overhead scanner device 100 of the present embodiment. FIG. 5 is a diagram illustrating an example of a velocity vector (optical flow) calculated between two area images (area image at time t−1 and area image at time t) captured when turning a page. . FIG. 6 is a flowchart illustrating an example of a materialization process in the overhead scanner device 100 of the present embodiment. FIG. 7 is a flowchart showing a subroutine of document monitoring processing. FIG. 8 is a flowchart showing a subroutine of moving object monitoring processing. FIG. 9 is a flowchart showing a subroutine of image correction processing. FIG. 10 is a flowchart showing details of image division processing by the image dividing unit 102g. FIG. 11 is a diagram illustrating a relationship between a boundary line (solid line portion in the figure), an inflection point (center point of the circle in the figure), and a dividing line (broken line part in the figure). FIG. 12 is a flowchart showing details of the curvature correction processing by the curvature correction unit 102h. FIG. 13 is a diagram schematically illustrating an example of a curvature correction process performed by the curvature correction unit 102h. FIG. 14 is a flowchart showing details of the page rearrangement process. FIG. 15 is a diagram schematically illustrating a state in which a page that does not need to be read is turned down by a user. FIG. 16 is a flowchart illustrating an example of the power saving process and the light source control process in the overhead scanner device 100 of the present embodiment.

  Hereinafter, embodiments of an overhead scanner device, an image acquisition method, and a program according to the present invention will be described in detail with reference to the drawings. In addition, this invention is not limited by this embodiment. In particular, in the present embodiment, the reading target may be described as a document such as a book. However, the reading target is not limited to this, and a medium bound with staples, a stack of single sheets, or the like may be read.

[1. Configuration of this embodiment]
The configuration of the overhead scanner device 100 according to the present embodiment will be described with reference to FIG. FIG. 1 is a block diagram illustrating an example of the configuration of the overhead scanner device 100.

  As shown in FIG. 1, the overhead scanner device 100 includes at least a linear imaging unit 110, an area imaging unit 120, and a control unit 102, and in the present embodiment, a storage unit 106, an input / output interface unit 108, and an indicator light source. 130 and a line light source 140 are further provided. Further, these units are communicably connected via an arbitrary communication path.

  The storage unit 106 stores various databases, tables, files, and the like. The storage unit 106 is a storage unit, and for example, a memory device such as a RAM / ROM, a fixed disk device such as a hard disk, a flexible disk, an optical disk, or the like can be used. The storage unit 106 stores computer programs for giving various instructions to a CPU (Central Processing Unit). The storage unit 106 includes a temporary image data file 106a, a processed image data file 106b, and an index object file 106c as shown in the figure.

  Among these, the image data temporary file 106a temporarily stores the image data read by the linear imaging unit 110 or the area imaging unit 120.

  The processed image data file 106b stores image data processed or edited by an image dividing unit 102g, a curvature correcting unit 102h, a page rearranging unit 102j, or the like, which will be described later, from image data read by the linear imaging unit 110. .

  The index object file 106c is an index object storage unit that stores the color, shape, and the like of an index object such as a hand or a finger presented by the user. Here, the index object file 106c may store the color (skin color) or shape of the user's hand or finger for each user.

  The input / output interface unit 108 connects the linear imaging unit 110, the area imaging unit 120, the indicator light source 130, and the line light source 140 to the control unit 102. Here, an LED light source, a monitor, or the like can be used as the indicator light source 130. As the line light source 140, an LED light source, a laser light source, or the like can be used, and light is emitted from the light source to the reading range.

  The linear imaging unit 110 scans an original placed upward from above using the linear sensor 13 and reads an image of the original. Here, in this embodiment, the linear imaging part 110 is provided with the controller 11, the motor 12, the linear sensor (line sensor) 13, and the A / D converter 14, as shown in FIG. The controller 11 controls the motor 12, the linear sensor 13, and the A / D converter 14 in response to a command from the control unit 102 via the input / output interface unit 108. The linear sensor 13 photoelectrically converts light reaching from the line in the main scanning direction of the document into an analog charge amount for each pixel on the line. Then, the A / D converter 14 converts the analog charge amount output from the linear sensor 13 into a digital signal, and outputs one-dimensional image data. When the motor 12 is driven to rotate, the original line to be read by the linear sensor 13 moves in the sub-scanning direction. Thus, one-dimensional image data is output from the A / D converter 14 for each line, and the control unit 102 generates two-dimensional image data by combining them. Here, FIG. 2 is a diagram illustrating an example of the appearance of an apparatus in which the linear imaging unit 110 and the area imaging unit 120 are integrated, and is a diagram illustrating the relationship among the main scanning direction, the sub-scanning direction, and the rotation direction by the motor 12. .

  As shown in FIG. 2, when the original is placed upward and the original is picked up by the linear imaging unit 110 from above, the linear sensor 13 reads the one-dimensional image data of the line in the main scanning direction shown in the figure. When the linear sensor 13 is rotated in the illustrated rotational direction by driving the motor 12, the reading line of the linear sensor 13 is moved in the illustrated sub-scanning direction. As a result, the image data of the two-dimensional document is read by the linear imaging unit 110.

  Returning to FIG. 1 again, the area imaging unit 120 scans an original placed upward from above using the area sensor 22 and reads an image of the original. Here, in the present embodiment, the area imaging unit 120 includes a controller 21, an area sensor 22, and an A / D converter 23 as shown in FIG. The controller 21 controls the area sensor 22 and the A / D converter 23 in response to a command from the control unit 102 via the input / output interface unit 108. The area sensor 22 photoelectrically converts light reaching from the original plane (two-dimensional plane in the main scanning direction and the sub-scanning direction shown in FIG. 2) into an analog charge amount for each pixel. Then, the A / D converter 23 converts the analog charge amount output from the area sensor 22 into a digital signal, and outputs two-dimensional image data. As a result, two-dimensional image data is output from the A / D converter 23. In the present embodiment, the linear sensor 13 can read more pixels per line than the area sensor 22, so that the image read by the linear imaging unit 110 is better than the image read by the area imaging unit 120. Is high definition (high image quality). Hereinafter, in order to distinguish between them, an image read by the area imaging unit 120 may be referred to as an “area image”, and an image read by the linear imaging unit 110 may be referred to as a “linear image”.

  The control unit 102 includes a CPU that controls the overhead scanner device 100 in an integrated manner. The control unit 102 has an internal memory for storing a control program, a program defining various processing procedures, and necessary data, and performs information processing for executing various processes based on these programs. As shown in the figure, the control unit 102 is roughly divided into an area image acquisition unit 102a, a feature point detection unit 102b, a velocity vector calculation unit 102c, a page direction determination unit 102d, a reading start determination unit 102e, and a linear image acquisition. A unit 102f, an image division unit 102g, a curvature correction unit 102h, a skin color detection unit 102i, a page rearrangement unit 102j, and a light source control unit 102k.

  The area image acquisition unit 102a controls the area imaging unit 120 to continuously acquire a plurality of area images. For example, as described above, the area image acquisition unit 102a controls the controller 21 of the area imaging unit 120 to acquire image data photoelectrically converted by the area sensor 22 and analog-digital converted by the A / D converter 23. The process of storing in the temporary image data file (buffer) 106a is repeated. In other words, the area image acquisition unit 102a has a function of monitoring (monitoring) the page turning of the document, the movement of the user's hand, and the like by performing motion shooting with the area sensor 22 and acquiring continuous images (moving images). Have.

  The feature point detection unit 102b detects feature points from the area image acquired by the area image acquisition unit 102a. Specifically, the feature point detection unit 102b extracts feature points from the area image based on the image data of the area image stored in the image data temporary file 106a by the area image acquisition unit 102a. For example, the feature point detection unit 102b extracts feature points (characteristic pixels) using a known feature point extraction method (for example, a known optical flow estimation method or a known pattern recognition algorithm feature point extraction method). May be.

  The velocity vector calculation unit 102c calculates the velocity vector of the feature point by comparing the feature points extracted by the feature point extraction unit 102b between a plurality of area images. For example, the velocity vector calculation unit 102c acquires the previous (time t-1) area image acquired by the area image acquisition unit 102a and stored in the image data temporary file 106a, and the newly acquired current (time t) area. The speed vector of the feature point is calculated by calculating the distance and direction in which the feature point extracted by the feature point extraction unit 102b has moved from the image. Here, the velocity vector calculation unit 102c tracks feature points (characteristic pixels) by an optical flow estimation method (gradient method), and the velocity vector field, that is, the behavior of an object (such as a manuscript or a hand) by the optical flow. You may grasp.

  The page direction determination unit 102d determines the page turning direction of the document based on the speed vector calculated by the speed vector calculation unit 102c. For example, the page direction determination unit 102d determines that the pages are turned from left to right when there are many right-direction speed vectors among the speed vectors (optical flows) of the plurality of feature points calculated by the speed vector calculation unit 102c. If there are many left-facing velocity vectors, it may be determined that the pages are turned from right to left.

  The reading start determination unit 102e determines the reading start by the linear sensor 13 based on the speed vector calculated by the speed vector calculation unit 102c. For example, the reading start determination unit 102e may determine the reading start by the linear sensor 13 when the motion based on the speed vector calculated by the speed vector calculation unit 102c stops. Further, the reading start determination unit 102e is linear when an index object such as a hand or a finger represents a specific movement (a gesture such as a piece sign) based on the speed vector calculated by the speed vector calculation unit 102c. The start of reading by the sensor 13 may be determined. Here, the reading start determination unit 102e may perform power management of the linear sensor 13 for power saving. For example, when the reading start by the linear sensor 13 is determined, the circuit of the linear sensor 13 is activated to start reading. If it is not determined, the circuit of the linear sensor 13 may be stopped.

  The linear image acquisition unit 102f controls the linear imaging unit 110 to acquire a linear image by the linear sensor 13 when the reading start determination unit 102e determines the reading start. For example, as described above, the linear image acquisition unit 102f controls the controller 11 of the linear imaging unit 110 to rotate and drive the motor 12, and is photoelectrically converted by the linear sensor 13 and analog / digital converted by the A / D converter 14. By synthesizing the one-dimensional image data for each line, two-dimensional image data (linear image image data) is generated and stored in the image data temporary file 106a.

  The image dividing unit 102g divides the linear image acquired by the linear image acquisition unit 102f to generate a divided image. Specifically, the image dividing unit 102g divides a linear image based on the image data acquired by the linear image acquisition unit 102f and stored in the image data temporary file 106a, and generates image data of the divided image. Here, the image dividing unit 102g obtains a boundary line (document edge) between the original and the background from the gradation difference of the linear image, obtains inflection points on the upper and lower sides of the boundary line, and uses lines connecting the inflection points. The area image may be divided into left and right. In this case, the image dividing unit 102g may function as an image cutout unit that cuts out the document area within the obtained boundary line range. Further, the image dividing unit 102g may cut out the linear image in the cutout range surrounded by the movement of the feature points based on the history of the velocity vector calculated by the velocity vector calculating unit 102c. That is, when the user traces the surroundings with a fingertip so as to surround the range that the user wants to cut out, the image dividing unit 102g overlaps the start point and the end point in the velocity vector group of the feature points of the user's fingertip. It may be determined that the cutout range is specified, and the linear image may be cut out in the cutout range.

  The curvature correcting unit 102h corrects the curvature distortion of the image divided by the image dividing unit 102g. For example, the curvature correction unit 102h may correct the distortion of the divided image so that the curvature of the upper side of the document edge becomes a horizontal line in contact with the upper side.

  The skin color detecting unit 102i detects a skin color partial region from the divided image divided by the image dividing unit 102g. For example, the skin color detection unit 102i may detect a skin color partial area from the divided image by a known pattern recognition algorithm or the like based on the color (skin color) of the user's hand stored in the index object file 106c.

  The page rearrangement unit 102j rearranges the image data of the divided images divided by the image division unit 102g according to the page turning direction determined by the page direction determination unit 102d, and stores it in the processed image data file 106b. Note that the page rearrangement unit 102j stores the image data of the corrected divided image when the curved image correction unit 102h corrects the distortion of the divided image. Further, the page rearrangement unit 102j deletes the divided images in which the skin color partial area is detected by the skin color detection unit 102i without storing them in the processed image data file 106b.

  The light source control unit 102k controls the indicator light source 130 to emit light indicating the controllable state of the linear sensor 13, or controls the line light source 140 to emit light indicating the reading range of the linear sensor 13. . Here, FIG. 3 illustrates a state in which the linear imaging unit 110 is viewed from the side, and is a diagram illustrating an example of light irradiation from the line light source 140 under the control of the light source control unit 102k. As shown in FIG. 3, the light source control unit 102k drives the motor 12 to rotate the line light source 140 in the rotation direction, and the line light in the main scanning direction from the line light source 140 is sub-scanned with the reading range as a limit. The reading range may be notified to the user by moving in the scanning direction.

[2. Processing of this embodiment]
An example of processing executed by the overhead scanner device 100 configured as described above will be described with reference to FIGS.

[2-1. Basic processing]
An example of basic processing in the overhead scanner device 100 of the present embodiment will be described with reference to FIGS. FIG. 4 is a flowchart showing an example of basic processing in the overhead scanner device 100 of the present embodiment.

  As shown in FIG. 4, first, the area image acquisition unit 102a controls the area imaging unit 120 to continuously acquire a plurality of area images (step SA1). For example, the area image acquisition unit 102a controls the controller 21 of the area imaging unit 120 to acquire image data of an area image photoelectrically converted by the area sensor 22 and analog-digital converted by the A / D converter 23 to obtain image data. Store in the temporary file 106a.

  The feature point detection unit 102b detects feature points from the area image acquired by the area image acquisition unit 102a (step SA2). Specifically, the feature point detection unit 102b extracts feature points from the area image based on the image data stored in the image data temporary file 106a by the area image acquisition unit 102a. For example, the feature point detection unit 102b extracts feature points (characteristic pixels) using a known feature point extraction method (for example, a known optical flow estimation method or a known pattern recognition algorithm feature point extraction method). May be.

  Then, the velocity vector calculation unit 102c calculates the velocity vector of the feature point by comparing the feature points extracted by the feature point extraction unit 102b between a plurality of images (step SA3). Specifically, the velocity vector calculation unit 102c acquires the previous (time t-1) area image acquired by the area image acquisition unit 102a and stored in the image data temporary file 106a, and the newly acquired current time (time t). ) To calculate the speed vector of the feature point. For example, the velocity vector calculation unit 102c tracks feature points (characteristic pixels) by an optical flow estimation method (gradient method), and grasps the velocity vector field, that is, the behavior of an object (such as a manuscript or a hand) by the optical flow. May be.

  Here, FIG. 5 shows an example of a velocity vector (optical flow) calculated between two area images (area image at time t−1 and area image at time t) captured at the time of page turning. FIG. The black circle point in FIG. 5 represents the start point of the velocity vector (the position of the feature point at time t−1), and one end on the opposite side of the solid or dashed black circle point is the end point of the velocity vector (feature at the time t). Point position). A solid line represents a trajectory from right to left, and a broken line represents a trajectory from left to right. As shown in FIG. 5, a speed vector is calculated for each feature point between two area images acquired by the area image acquisition unit 102a. In particular, the speed from the right to the left according to the movement of the page in the middle of turning. The vector is increasing. Thus, the velocity vector field, that is, the behavior of the object (original or hand) can be grasped by the optical flow.

（ここで、ＳＳＤは、二乗誤差を表し、２つのエリア画像間の類似度を表す指標となる。また、Ｉ（ｘ，ｙ，ｔ）は、座標（ｘ，ｙ，ｔ）における画素値であり、Ｗは、画素の集合（探索範囲）を表す。） Here, the velocity vector calculation unit 102c may search and track feature points using the gradient method (Lucas Kanade Algorithm) among optical flow estimation methods. The gradient method is a method for obtaining a square error on the assumption that “a characteristic pixel brightness is constant even if time changes”. For example, the square error may be obtained by the following equation. That is, the smaller the square error, the higher the similarity between feature points between the two area images. Therefore, the trajectory connecting the feature points with the highest similarity may be calculated as a velocity vector.

(Here, SSD represents a square error and serves as an index representing the similarity between two area images. Further, I (x, y, t) is a pixel value at coordinates (x, y, t). Yes, W represents a set of pixels (search range).)

  Returning to FIG. 4 again, the reading start determination unit 102e determines the reading start by the linear sensor 13 based on the speed vector calculated by the speed vector calculation unit 102c (step SA4). For example, the reading start determination unit 102e may determine the reading start by the linear sensor 13 when the motion based on the speed vector calculated by the speed vector calculation unit 102c stops. The reading start determination unit 102e reads by the linear sensor 13 when an index object such as a hand or a finger represents a specific movement (gesture) based on the speed vector calculated by the speed vector calculation unit 102c. The start may be determined.

  When the reading start determination unit 102e determines that reading is started (step SA4, Yes), the linear image acquisition unit 102f controls the linear imaging unit 110 to acquire a linear image by the linear sensor 13 (step SA5). ). For example, the linear image acquisition unit 102f controls the controller 11 of the linear imaging unit 110 to rotate and drive the motor 12, and performs primary conversion for each line photoelectrically converted by the linear sensor 13 and analog-digital converted by the A / D converter 14. By synthesizing the original image data, two-dimensional image data is generated and stored in the temporary image data file 106a. If the reading start determination unit 102e does not determine the reading start (No at Step SA4), the control unit 102 returns the process to Step SA1 and repeats the above-described processing from Step SA1 to Step SA4, and then the original page. Continue to monitor turning and user hand movements.

  The above is an example of the basic processing in the overhead scanner device 100 of the present embodiment.

[2-2. Materialization process]
Subsequently, in the basic process described above, an example of a specific process further including a page direction determination process, an image division process, a curvature correction process, a skin color detection process, a page rearrangement process, and the like will be described with reference to FIGS. I will explain. FIG. 6 is a flowchart illustrating an example of a materialization process in the overhead scanner device 100 of the present embodiment.

  As shown in FIG. 6, first, the control unit 102 of the overhead scanner device 100 controls the area imaging unit 120 to perform document monitoring processing (step SB1). Here, FIG. 7 is a flowchart showing a subroutine of document monitoring processing.

  As shown in FIG. 7, the area image acquisition unit 102a controls the area imaging unit 120 to acquire image data of the area image (step SB11).

  Then, the area image acquisition unit 102a detects the original by detecting a white rectangle or the like by a known pattern recognition algorithm based on the acquired image data of the area image (step SB12).

  If the document cannot be detected (No in step SB13), the area image acquisition unit 102a stores the image data of the area image acquired this time in the image data temporary file 106a serving as a buffer (step SB14), and the above-described step. The processes of SB11 to step SB13 are repeated.

  On the other hand, when the document can be detected (step SB13, Yes), the area image acquisition unit 102a finishes the document monitoring process.

  Returning to FIG. 6 again, the control unit 102 of the overhead scanner device 100 performs a moving object monitoring process (step SB2). Here, FIG. 8 is a flowchart showing a subroutine of the moving object monitoring process.

  As shown in FIG. 8, the feature point detection unit 102b obtains the previous (time t-1) area image acquired by the area image acquisition unit 102a and stored in the image data temporary file 106a, and the newly acquired current time ( A feature point is detected from the area image at time t) (step SB21).

  The speed vector calculation unit 102c compares the feature points extracted by the feature point extraction unit 102b between the area images (between the area image at the time t-1 and the area image at the time t), thereby obtaining the feature. A point velocity vector is calculated (step SB22). For example, the velocity vector calculation unit 102c searches for feature points having the highest similarity between the area images using the above-described mathematical formula for calculating the square error, and uses a trajectory connecting the searched feature points as a velocity vector (optical flow). May be calculated.

  Then, the reading start determining unit 102e determines whether or not there is a motion based on the speed vector calculated by the speed vector calculating unit 102c (step SB23). For example, the reading start determination unit 102e determines whether or not the speed vector calculated by the speed vector calculation unit 102c is a zero vector having a length of zero.

  When the reading start determining unit 102e determines that there is a motion (step SB23, Yes), the page direction determining unit 102d determines the page direction of the document based on the speed vector calculated by the speed vector calculating unit 102c. (Step SB24). For example, the page direction determination unit 102d determines that the page is turned from left to right when there are many right direction speed vectors among the speed vectors of the plurality of feature points calculated by the speed vector calculation unit 102c. If there are many vectors, it may be determined that pages are turned from right to left.

  Then, the area image acquisition unit 102a stores the image data of the area image acquired at this time t in the temporary image data file 106a that is a buffer (step SB25).

  Then, the area image acquisition unit 102a controls the area imaging unit 120 to newly acquire the image data of the area image at time t + 1 (step SB26). Thereafter, the processing from step SB21 to step SB23 described above is repeated.

  When it is determined in step SB23 that there is no movement, for example, when the velocity vector is a zero vector (step SB23, No), the area image acquisition unit 102a determines whether the page direction has been determined by the page direction determination unit 102d. If it is determined (step SB27) and the page direction is not determined (No in step SB27), the process returns to step SB25 and the above-described process is repeated.

  On the other hand, when the area image acquisition unit 102a determines that the page direction has been determined by the page direction determination unit 102d (Yes in step SB27), the area image acquisition unit 102a issues a trigger to start reading linear images (step SB28). This completes the moving object monitoring process subroutine.

  Returning to FIG. 6 again, the linear image acquisition unit 102f controls the linear imaging unit 110 in accordance with the reading start trigger issued by the reading start determination unit 102e, and acquires the image data of the linear image by the linear sensor 13 ( Step SB3).

  Then, the control unit 102 of the overhead scanner device 100 performs image correction processing (step SB4). Here, FIG. 9 is a flowchart showing a subroutine of image correction processing.

  As shown in FIG. 9, the image dividing unit 102g divides the linear image acquired by the linear image acquisition unit 102f to generate a divided image (step SB41). Here, FIG. 10 is a flowchart showing details of the image division processing by the image dividing unit 102g.

  As illustrated in FIG. 10, when the image dividing unit 102g acquires the image data of the linear image acquired by the linear image acquiring unit 102f (step SB411), the image dividing unit 102g determines whether the document and the background are based on the gradation difference of the linear image. A boundary line (document edge) is detected (step SB412).

  Then, the image dividing unit 102g detects inflection points on the upper and lower sides of the boundary line from the detected boundary line (step SB413). Here, FIG. 11 is a diagram showing the relationship between the boundary line (solid line portion in the figure), the inflection point (center point of the circle in the figure), and the dividing line (broken line part in the figure). As shown in FIG. 11, when one side of a page is bound like a book or the like, the upper and lower sides of the boundary line are smoothly distorted by bending the paper when the page is opened. The slope of the tangent line on the upper and lower sides changes sharply at the end. Using this property, the image dividing unit 102g obtains each inflection point from the upper and lower sides of the boundary line.

  Then, the image dividing unit 102g sets a line connecting two inflection points as a dividing line, and sets two left and right page areas surrounded by the dividing line and the boundary line as a cutout range (step SB414).

  Then, the image dividing unit 102g extracts the divided image in the cutout range from the image data of the linear image, cuts out the page (step SB415), and stores the divided image in the image data temporary file 106a (step SB416).

  Returning again to FIG. 9, the curvature correcting unit 102h corrects the curvature distortion of the divided image extracted in units of pages by the image dividing unit 102g (step SB42). Here, FIG. 12 is a flowchart showing details of the curvature correction processing by the curvature correction unit 102h.

  As shown in FIG. 12, the curvature correcting unit 102h creates a horizontal line that touches the upper side of the cutout range (that is, the upper side of the boundary line (document edge)) (step SB421).

  Then, the curvature correcting unit 102h aligns the heights of the vertical lines so that the curvature of the upper side of the cutout range matches the horizontal line, and corrects the distortion of the divided image (step SB422). Here, FIG. 13 is a diagram schematically illustrating an example of a curvature correction process performed by the curvature correction unit 102h. In FIG. 13, the thick broken line represents the created horizontal line, the thin broken line represents the original cutout range (border line of the curved page), and the solid line represents the cutout range after correction.

  As shown in FIG. 13, as an example, the curvature correction unit 102 h divides a divided image of the original cutout range into strips each having a width of one pixel in the vertical direction, and the upper end of the strip-shaped vertical line becomes a horizontal line. You may correct | amend distortion of a divided image by moving vertically until it reaches | attains.

  Returning to FIG. 9 again, the control unit 102 of the overhead scanner device 100 uses the page direction determination unit 102d to determine the image data of the divided image after the correction by the curvature correction unit 102h divided by the image division unit 102g. The pages are rearranged according to the page direction (step SB43). Here, FIG. 14 is a flowchart showing details of the page rearrangement process.

  As shown in FIG. 14, the skin color detection unit 102 i detects a skin color partial region based on the hue of the divided image after distortion correction (step SB 431). For example, the skin color detection unit 102i detects a skin color partial region from the divided image using a known pattern recognition algorithm or the like based on the color (skin color) of the user's hand stored in the index object file 106c. FIG. 15 is a diagram schematically illustrating a state in which a page that does not need to be read is turned down by a user. As shown in FIG. 15, when the user does not want to read an arbitrary page, if the page is covered with a hand, a linear image of the document including the hand is acquired. By detecting, the user's hand can be detected.

  When the hand is not detected by the skin color detection unit 102i (step SB432, No), the page rearrangement unit 102j arranges the image data of the divided images after distortion correction according to the page direction determined by the page direction determination unit 102d. Instead, it is stored in the processed image data file 106b (step SB433). For example, when the page direction determined by the page direction determination unit 102d represents page turning from right to left, the spread left page is stored in the processed image data file 106b in page order so that the left page comes before the right page. If the skin color detection unit 102i detects a hand from the divided image (Yes at step SB432), the page rearrangement unit 102j deletes the image data of the divided image without storing it in the processed image data file 106b.

  Then, the page rearrangement unit 102j determines whether or not the above-described processing has been completed for both the divided images of the left and right pages (step SB434). If there is an unprocessed divided image (step SB434, No), the above-described processing is performed. repeat. On the other hand, when there is no unprocessed divided image (step SB434, Yes), the page rearrangement process is finished.

  Returning to FIG. 6 again, the overhead scanner 100 performs processing on the document including a plurality of pages as described above, and the image data after various processing is stored in the processed image data file 106b (step SB5). Finish the materialization process.

  The above is an example of the specific processing in the overhead scanner device 100 of the present embodiment. In the above-described specific processing, the flow of one process when an image of two pages of left and right spread pages is read has been described. However, by repeating the above-described processing of step SB1 to step SB5, another spread is performed. The page can be automatically read continuously.

[2-3. Power saving process and light source control process]
Next, an example of processing that further includes power saving processing, light source control processing, and the like in the above-described specific processing will be described with reference to FIG. FIG. 16 is a flowchart illustrating an example of the power saving process and the light source control process in the overhead scanner device 100 of the present embodiment.

  As described in the above embodiment, the overhead scanner device 100 controls the area imaging unit 120 to perform document monitoring processing and moving object monitoring processing, and based on the area image acquired via the area imaging unit 120. Then, a trigger for starting to read a high-definition linear image by the linear imaging unit 110 is issued. Therefore, since it is not necessary to always activate the linear imaging unit 110, it is possible to save power by setting the standby state before issuing a trigger. For example, when the user turns the page and the linear image is read according to the issued trigger, but the user does not show the next movement, the user enters the standby state and the user moves. Processing for returning from the standby state may be performed.

  That is, as shown in FIG. 16, the overhead scanner device 100 performs the above-described steps SB21 to SB23 when the trigger for starting the reading of the linear image is not issued by the reading start determination unit 102e (No in step SC1). Similar processing is performed (step SC2 to step SC4). The subsequent processing is performed in parallel with the above-described materialization processing.

  If the reading start determination unit 102e determines that there is movement based on the velocity vector (step SC4, Yes), the circuit around the linear sensor 13 is activated (step SC5). Here, the circuit around the linear sensor 13 may include the circuit of the controller 11, the motor 12, and the A / D converter 14 in addition to the circuit of the linear sensor 13.

  Then, the light source control unit 102k controls the indicator light source 130 to turn on the light indicating the controllable state of the linear sensor 13 (step SC6).

  Then, the light source control unit 102k controls blinking of the line light source 140 to emit light indicating the reading range of the linear sensor 13 (step SC7). For example, the light source control unit 102k drives the motor 12 to rotate the line light source 140 in the rotation direction and blink the line light in the main scanning direction from the line light source 140, while limiting the reading range to the sub scanning direction. The reading range may be notified to the user by moving to.

  On the other hand, in step SC4, when it is determined that there is no movement based on the velocity vector (step SC4, No), the reading start determination unit 102e stops the circuits around the linear sensor 13 (step SC8).

  Then, the light source control unit 102k turns off the indicator light source 130 (step SC9) and turns off the line light source 140 (step SC10).

  The overhead scanner device 100 repeatedly executes the above processing until the reading start determination unit 102e issues a linear image reading start trigger (No in step SC1), and saves power. The controllable state and reading range of the sensor 13 are notified by light. Each time a scan start trigger is issued (step SC1, Yes), the above-described processing may be returned to step SC1, and the above-described processing may be repeated until a new trigger is issued.

  The above is an example of the power saving process and the light source control process.

[3. Summary of this embodiment and other embodiments]
As described above, according to the present embodiment, the overhead scanner device 100 controls the area imaging unit 120 to continuously acquire area images, extracts feature points from the acquired area images, and features between area images. By comparing the points, the velocity vector (optical flow) of the feature point is calculated, and the start of reading by the linear sensor 13 is determined based on the calculated velocity vector. Thereby, it is possible to obtain a high-definition image that is excellent in manuscript reading operability. For example, a conventional scanner has to be controlled from a button incorporated in the apparatus or a personal computer connected to the scanner, and this has led to a decrease in productivity especially in the case of processing a plurality of documents. According to the present invention, it is possible to acquire a high-definition document image without requiring a special button or tool for starting document reading.

  Further, according to the present embodiment, the overhead scanner device 100 determines the reading start by the linear sensor 13 when the movement based on the velocity vector stops. As a result, reading is automatically started upon completion of document placement such as turning pages or pressing a document with a hand, so that excellent operability can be exhibited.

  Further, according to the present embodiment, the overhead scanner device 100 determines the page turning direction of the document based on the calculated speed vector, and controls the linear imaging unit 110 when the reading start is determined to control the document image. A linear image is acquired, the linear image is divided, and the divided divided images are rearranged according to the determined page turning direction and stored in the processed image data file 106b. As a result, it is possible to determine the page order based on the movement caused by the page turning and rearrange the images divided into page units in order of page, so that editing operations such as dividing and rearranging the read images on the computer are unnecessary. Thus, productivity can be significantly increased. That is, according to the present embodiment, the user can digitize a document such as a book by simply turning the book one after another without operating the scanner, and as a result, the user can control the user with scanner control. Since there is no need to bother, it is possible to save time and improve productivity.

  Further, according to the present embodiment, the overhead scanner device 100 obtains a boundary line (document edge) between the original and the background from the gradation difference of the acquired linear image, obtains inflection points on the upper and lower sides of the boundary line, and The linear image is divided into right and left by a dividing line connecting inflection points. As a result, when the page is composed of left and right spread pages such as a book, the image can be accurately divided in units of pages.

  Further, according to the present embodiment, the overhead scanner device 100 corrects the curved distortion of the divided divided images, rearranges the corrected divided images according to the determined page turning direction, and stores them in the processed image data file 106b. As a result, it is possible to correct a curved distortion that occurs when a plurality of sheets are bound on one side, such as a book, and to acquire an image in page units without distortion.

  Further, according to the present embodiment, the overhead scanner device 100 detects a skin color partial area from the divided divided images, and deletes the divided image from which the skin color partial area is detected without storing it in the processed image data file 106b. To do. As a result, when the user turns down an unnecessary page with his / her hand, the area of the hand on the image can be detected based on the skin color, and an original image not including the unnecessary page can be acquired.

  Further, according to the present embodiment, the overhead scanner device 100 activates the circuit around the linear sensor 13 when there is a motion based on the velocity vector when the reading start by the linear sensor 13 is not determined, and the velocity vector When there is no movement based on, the circuit around the linear sensor 13 is stopped. As a result, when reading by the linear sensor 13 is unnecessary, such as when no document is placed, the linear sensor 13 can be put in a standby state to save power.

  In addition, according to the present embodiment, the overhead scanner device 100 controls the indicator light source 130 to emit light indicating the controllable state of the linear sensor 13, and controls the line light source 140 to indicate light to indicate the reading range. Flashes. Thereby, it is possible to notify the user that the linear sensor 13 is ready to be read and the reading range of the linear sensor 13 before reading.

  In addition, according to the present embodiment, the overhead scanner device 100 cuts out a linear image in a cutout range surrounded by movement of feature points based on the calculated history of velocity vectors. Thereby, when a user traces the outer periphery of the range which he wants to cut out with a fingertip, an image can be cut out correctly according to a user's intention.

  Furthermore, the present invention may be implemented in various different embodiments other than the above-described embodiments within the scope of the technical idea described in the claims. For example, the case where the overhead scanner device 100 performs processing in a stand-alone form has been described as an example. However, processing is performed in response to a request from a client terminal in a separate housing from the overhead scanner device 100, and the processing result is You may make it return to a client terminal. In addition, among the processes described in the embodiment, all or a part of the processes described as being automatically performed can be manually performed, or all of the processes described as being manually performed can be performed. Alternatively, a part can be automatically performed by a known method. In addition, the processing procedures, control procedures, specific names, information including registration data for each processing, screen examples, and database configuration shown in the above documents and drawings may be arbitrarily changed unless otherwise specified. Can do.

  Further, regarding the overhead scanner device 100, the illustrated components are functionally conceptual and do not necessarily need to be physically configured as illustrated. For example, the processing functions provided in each apparatus of the overhead scanner apparatus 100, in particular, the processing functions performed by the control unit 102, all or any part thereof are interpreted and executed by a CPU (Central Processing Unit) and the CPU. It may be realized by a program to be executed, or may be realized as hardware by wired logic. The program is recorded on a recording medium to be described later, and is mechanically read by the overhead scanner device 100 as necessary. That is, a computer program for performing various processes is recorded in the storage unit 106 such as a ROM or an HD. This computer program is executed by being loaded into the RAM, and constitutes a control unit in cooperation with the CPU. The computer program may be stored in an application program server connected to the overhead scanner device 100 via an arbitrary network, and may be downloaded in whole or in part as necessary. is there.

  In addition, the program according to the present invention may be stored in a computer-readable recording medium, and may be configured as a program product. Here, the “recording medium” means a memory card, USB memory, SD card, flexible disk, magneto-optical disk, ROM, EPROM, EEPROM (registered trademark), CD-ROM, MO, DVD, and Blu-ray. (Registered Trademark) Any “portable physical medium” such as Disc is included. The “program” is a data processing method described in an arbitrary language or description method, and may be in any format such as source code or binary code. The “program” is not necessarily limited to a single configuration, but is distributed in the form of a plurality of modules and libraries, or in cooperation with a separate program represented by an OS (Operating System). Including those that achieve the function. In addition, a well-known structure and procedure can be used about the specific structure for reading a recording medium in each apparatus shown in embodiment, a reading procedure, or the installation procedure after reading.

  Various databases and the like (image data temporary file 106a, processed image data file 106b, and index object file 106c) stored in the storage unit 106 are a memory device such as a RAM and a ROM, a fixed disk device such as a hard disk, a flexible disk, The storage means such as an optical disk stores various programs, tables, databases, and the like used for various processes.

  The overhead scanner device 100 may be configured as an information processing device such as a known personal computer or workstation, or may be configured by connecting an arbitrary peripheral device to the information processing device. Further, the overhead scanner device 100 may be realized by installing software (including programs, data, and the like) that causes the information processing device to realize the method of the present invention. Furthermore, the specific form of distribution / integration of the devices is not limited to that shown in the figure, and all or a part of them may be functional or physical in arbitrary units according to various additions or according to functional loads. Can be distributed and integrated. That is, the above-described embodiments may be arbitrarily combined and may be selectively implemented.

  As described above, the overhead scanner device, the image acquisition method, and the program according to the present invention can be implemented in many industrial fields, particularly in the image processing field that handles images read by a scanner, and are extremely useful. is there.

DESCRIPTION OF SYMBOLS 100 Overhead scanner apparatus 102 Control part 102a Area image acquisition part 102b Feature point detection part 102c Speed vector calculation part 102d Page direction determination part 102e Reading start determination part 102f Linear image acquisition part 102g Image division part 102h Curvature correction part 102i Skin color detection part 102j Page rearrangement unit 102k Light source control unit 106 Storage unit 106a Temporary image data file 106b Processed image data file 106c Index object file 108 Input / output interface unit 110 Linear imaging unit 11 Controller 12 Motor 13 Linear sensor 14 A / D converter 120 Area imaging unit 21 Controller 22 Area sensor 23 A / D converter 130 Indicator light source 140 Line light source

Claims (3)

An area sensor, a linear sensor, a control unit, and a storage unit ;
The controller is
Area image acquisition means for controlling the area sensor to acquire a plurality of images continuously;
Feature point extraction means for extracting feature points from the plurality of images acquired by the area image acquisition means;
A speed vector computing means for computing a speed vector of the feature point by comparing the feature points extracted by the feature point extracting means between the plurality of images;
Based on the speed vector calculated by the speed vector calculating means, reading start determining means for determining the reading start by the linear sensor;
Page direction determining means for determining a page turning direction of the document based on the speed vector calculated by the speed vector calculating means;
A linear image acquisition unit that controls the linear sensor to acquire an image of the original when the reading start determination unit determines that reading is started;
Image dividing means for dividing the image acquired by the linear image acquisition means;
Skin color detecting means for detecting a skin color partial area from the image divided by the image dividing means;
Page reordering means for reordering the images divided by the image dividing means according to the page turning direction determined by the page direction determining means and storing them in the storage unit;
Equipped with a,
The page sorting means
An overhead scanner device, wherein the skin color detection unit deletes the image in which the partial area of the skin color is detected without storing the image in the storage unit . An area sensor, a linear sensor, a control unit, and a storage unit, and executed by the control unit of the overhead scanner device;
An area image acquisition step of acquiring a plurality of images continuously by controlling the area sensor;
A feature point extracting step of extracting feature points from the plurality of images acquired in the area image acquiring step;
A speed vector calculation step of calculating a speed vector of the feature point by comparing the feature points extracted in the feature point extraction step between the plurality of images;
A reading start determination step of determining reading start by the linear sensor based on the speed vector calculated in the speed vector calculation step;
A page direction determining step for determining a page turning direction of the document based on the speed vector calculated in the speed vector calculating step;
A linear image acquisition step of controlling the linear sensor to acquire an image of the document when the reading start is determined in the reading start determination step;
An image dividing step of dividing the image acquired in the linear image acquiring step;
A skin color detection step for detecting a skin color partial region from the image divided in the image division step;
A page rearrangement step of rearranging the images divided in the image division step according to the page turning direction determined in the page direction determination step and storing them in the storage unit;
Including
The page rearranging step includes
An image acquisition method, wherein the image in which the partial area of the skin color is detected in the skin color detection step is deleted without being stored in the storage unit. An area sensor, a linear sensor, a control unit, and a storage unit are provided, and are executed by the control unit of the overhead scanner device.
An area image acquisition step of acquiring a plurality of images continuously by controlling the area sensor;
A feature point extracting step of extracting feature points from the plurality of images acquired in the area image acquiring step;
A speed vector calculation step of calculating a speed vector of the feature point by comparing the feature points extracted in the feature point extraction step between the plurality of images;
A reading start determination step of determining reading start by the linear sensor based on the speed vector calculated in the speed vector calculation step;
A page direction determining step for determining a page turning direction of the document based on the speed vector calculated in the speed vector calculating step;
A linear image acquisition step of controlling the linear sensor to acquire an image of the document when the reading start is determined in the reading start determination step;
An image dividing step of dividing the image acquired in the linear image acquiring step;
A skin color detection step for detecting a skin color partial region from the image divided in the image division step;
A page rearrangement step of rearranging the images divided in the image division step according to the page turning direction determined in the page direction determination step and storing them in the storage unit;
Including
The page rearranging step includes
The program in which the image in which the partial area of the skin color is detected in the skin color detection step is deleted without being stored in the storage unit.

Images