JP5171486B2 - Object movement detection method and printer - Google Patents

Description

  The present invention relates to a technique for detecting movement of an object by image processing, and a technical field of a printer that employs the technique.

  When printing while transporting media such as print paper, if there is an error in the transport amount, density unevenness of the halftone image or magnification error will occur, and the quality of the obtained print image will deteriorate. . For this reason, high-precision parts are used and a precise transport mechanism is installed, but the demand for print quality is severe and further accuracy is desired. On the other hand, the demand for cost is strict, and both high accuracy and low cost are required.

  To cope with this, the movement of the media is detected with high accuracy, and in order to realize stable conveyance by feedback control, a part of the media is imaged and the movement of the media conveyed by image processing is detected. Attempts have been made.

Patent Document 1 and Patent Document 2 disclose a method for detecting the movement of the media. The apparatus disclosed herein continuously captures an image of a fine surface pattern of a medium on a platen during transportation using an image sensor. Then, the obtained images are compared by a pattern matching method to detect movement information (movement amount, movement speed, movement direction, etc.) of the media. (See Patent Document 1 and Patent Document 2)
JP 2001-138491 A JP 2007-217176 A

  However, depending on the media to be used, there is a possibility that the transported medium slightly floats (hereinafter referred to as “medium lift”) with respect to the original transport path defined by the platen. When this phenomenon occurs, in the apparatus that detects the movement of the media by the above-described imaging, the interval between the imaging sensor and the media changes. Then, the imaging magnification changes, and the image size (magnification) of the surface pattern of the media obtained by imaging changes. If the image is larger or smaller than the original, the position of the surface pattern before and after the conveyance becomes different from the original, which causes a deterioration in the accuracy of the conveyance amount detection. This is a problem to be solved in pursuing improvement in detection accuracy.

  In Patent Document 1 and Patent Document 2, there is no disclosure or suggestion regarding the recognition and solution of such problems. The present invention has been made in view of recognition of the above-described problems.

  An object of the present invention is to provide a method capable of detecting the movement of an object with higher accuracy than before. Another object of the present invention is to provide a printer capable of detecting the movement of a medium with higher accuracy than before and forming an image with good quality.

  The object movement detection method of the present invention that solves the above problems includes a first step of acquiring an image by imaging the same area of the object from a plurality of different directions using an image sensor, and the image corresponding to the same area. A second step of obtaining a phase difference between a plurality of separated regions on the sensor and correcting a magnification of the acquired image based on the phase difference, and obtaining a corrected image according to the movement of the object; One step and the second step are repeated, and a third step of detecting the movement of the object by comparing the plurality of obtained corrected images is provided.

  Further, the printer of the present invention includes a printing unit that conveys a medium and prints, an imaging unit that has an image sensor that captures an image of the same area of the medium from a plurality of different directions, and acquires the image in the same area. Image processing means for obtaining a phase difference between a plurality of separated regions on the corresponding image sensor and correcting a magnification of the acquired image based on the phase difference; and accompanying the movement of the medium The detection unit for detecting the conveyance information of the medium by comparing the plurality of the corrected images obtained by repeating the image acquisition by the imaging unit and the processing by the image processing unit, and the detection unit It has a control means which controls conveyance of the media based on detection.

  According to the present invention, it is possible to provide a method capable of detecting the movement of an object with higher accuracy than before. Further, it is possible to provide a printer that can detect the movement of the medium with higher accuracy than before and can form an image with good quality.

<First Embodiment>
Exemplary embodiments of the present invention will be described below with reference to the drawings. However, the components described in this embodiment are merely examples, and are not intended to limit the scope of the present invention only to them.

  The object movement detection method of the present invention can be widely applied to a field where it is required to detect the movement of an object with high accuracy, including a printer. For example, the present invention can be applied to devices such as printers and scanners, and devices used in the industrial field, industrial field, physical distribution field, etc., which carry articles for inspection, reading, processing, or various processes. The printer of the present invention can be applied to various types of printers such as an inkjet method, an electrophotographic method, a thermal method, and a dot impact method.

  Hereinafter, an ink jet printer which is an example of an image forming apparatus will be described as an example.

  FIG. 1 is a schematic view of the main part of an ink jet printer. The medium 1 (a roll sheet is used here, but is not limited thereto) is fed onto the platen 5 by a conveyance mechanism including a conveyance roller 2 and a pinch roller 4. The carriage 7 reciprocates in the main scanning direction (perpendicular to the paper surface) by a motor (not shown). The carriage 7 holds an inkjet print head 8 and performs printing by discharging ink from the print head 8 to the print surface (upper surface) of the medium 1 while moving in the main scanning direction. As the inkjet method, various methods such as a method using a heating element, a method using a piezo element, a method using an electrostatic element, and a method using a MEMS element can be used.

  The conveyance roller 2 is intermittently rotated by the stepping motor 3 in synchronization with the movement of the carriage 7 (main scanning). While printing is performed while the carriage 7 moves in the main scanning direction, the stepping motor 3 stops and the medium 1 does not move. At the timing between printing, the stepping motor 3 moves the medium 1 by one line step by step.

  The conveyance control circuit 11 is a controller that controls these movements. A driver for driving the motors of the stepping motor 3 and the carriage 7 is provided, and the medium conveyance in the sub-scanning direction by the conveyance roller 2 and the movement of the print head 8 in the main scanning direction by the carriage are controlled. The controller 10 controls the entire apparatus and includes a CPU 10a and a memory 10b.

  The platen 5 is provided with a reader 6 that optically reads the medium 1 moving on the platen from the lower surface and detects the amount of movement. The reader 6 is controlled by a reading control circuit 9. The reading control circuit 9 includes a memory 9a and an A / D converter 9b, and also includes an image processing unit that performs various image processing described later.

  FIG. 2 is a diagram illustrating a physical configuration of the reader 6. In general, the integrated optical unit includes an illumination system 1001 that illuminates the medium 1 and an imaging system 1002 that captures an illuminated area of the medium 1. The illumination system 1001 includes a light source 21 including a light source 21 such as an LED or an OLED, and a lens 22a and a reflection portion 22b. The imaging system 1002 includes a lens 23a (23b), a diaphragm 24a (24b), and an image sensor 25. The light source 21 and the image sensor 25 are arranged on a common substrate 26, and these controls are performed by the reading control circuit 9.

  FIG. 3 is a view of the imaging system 1002 of the reader as viewed from the side. Two lenses 23 a and 23 b having the same optical characteristics and diaphragms 24 a and 24 b paired with the lenses 23 a and 23 b are arranged along the conveyance direction 31 (sub-scanning direction) of the medium 1.

  Next, the operation of the above apparatus will be described.

  First, a method for detecting the movement amount of the medium 1 will be described. In FIG. 2, the light emitted from the light source 21 is guided by the light guide 22 to illuminate the medium 1 from an oblique direction. In general, the surface of the media (particularly the back surface of the print surface) has an infinite number of uneven patterns microscopically. Since this is a random pattern without regularity, the uneven pattern at a certain place is unique to that place. When the back surface of the medium 1 is illuminated from an oblique direction, a shadow of a convex portion is generated, and a clear light / dark pattern corresponding to the unevenness is observed. The light / dark pattern image is formed on the image sensor 25 by the lenses 23a and 23b. Here, the light beams are stopped using the stops 24a and 24b to increase the depth of focus of the optical system. For this reason, even if the distance between the medium 1 and the reader 6 varies from the original distance, the blur of the image formed on the image sensor 25 is negligibly small and an image with good contrast can be obtained.

  First, a light / dark pattern is imaged by the image sensor 25 at a timing when the conveying roller 2 is stationary between prints in the main scanning direction. The imaging signal is sent to the reading control circuit 9, and converted into a digital signal by the A / D converter 9b. The image data obtained in this way is stored in the memory 9a as an image before conveyance. Next, one line is printed while the print head 8 is main-scanned by the carriage 7.

  Then, the conveyance roller 2 is rotated by the stepping motor 3, and the medium 1 is moved by a specified amount for one line in the sub-scanning direction and stopped. This specified amount is obtained corresponding to the print width and the number of passes of the print head 8 in the sub-scanning direction, and is stored in advance in the memory as the reference carry amount mo.

  Then, after transporting one row, the medium 1 is illuminated by the illumination system 1001 and the light / dark pattern is imaged by the image sensor 25 in the same manner as described above. The image data obtained by this imaging is stored in the memory 9a as an image after conveyance for one line.

  Examples of these continuously acquired images before and after conveyance are shown in FIGS. 4 and 5. 4 shows an image 41 before conveyance, and FIG. 5 shows an image 51 after conveyance.

  The image processing unit of the reading control circuit 9 compares the two images before and after the conveyance by the pattern matching (image correlation calculation) method to calculate the movement amount of the medium 1. Actually, the image obtained by imaging is not used as it is, but as will be described later, a corrected image before and after conveyance (IM (n−1)) is obtained by performing magnification or synthesis image correction. (IM (n)) is obtained. Then, the corrected images before and after the conveyance are compared by pattern matching.

  The pattern matching process is as follows. First, an image of a region 42 (correlation window) having a predetermined location and size as shown in FIG. 4 is cut out from the image 41 (IM (n−1)) before transport and stored in the memory 9a as a reference image. Remember. Further, an image having the same size as that of the region 42 is cut out from the image 51 (IM (n)) after conveyance in FIG. Then, this is compared with the reference image of the previous region 42 to obtain a correlation value. This clipping and comparison are repeated while shifting one pixel at a time in the entire area in the image 51 after conveyance. Then, a region 52 where the correlation value is an extreme value is obtained. This area 52 is the place where the pattern most closely matches the area 42 and is determined to be an image of the same part.

  For the two images 42 and 52, the difference (pixel number difference) between the coordinate positions (pixel positions) in the images is calculated. This difference means the actual moving amount m ′ of the medium 1 before and after the conveyance. Furthermore, the movement amount may be obtained with an accuracy of 1 pixel or less using interpolation calculation.

  The actual movement amount m ′ of the medium thus calculated is compared with the reference conveyance amount mo that is the target value in the conveyance roller 2 by the controller 10, and these differences (m′−mo) are regarded as the movement amount error. To do. When there is a slip (slip) during conveyance between the conveyance roller 2 and the medium 1, the calculated actual movement amount becomes smaller than the reference conveyance amount mo, and the movement amount error becomes large. If the slip is integrated, it causes image deterioration such as density unevenness and vertical error of the printed image. In order to prevent this, the obtained movement amount error is fed back to the feed amount of the next step movement of the transport roller 2. That is, the next conveyance is m = mo + (m′−mo). Thereby, the movement amount error in the next step movement can be eliminated or reduced.

== Imaging with two lenses ==
As described in FIG. 3, the imaging system 1002 includes two lenses 23a and 23b. Each of the images 41 and 51 before and after the conveyance is an image obtained by joining and synthesizing two images (IM1 and IM2) formed separately on the image sensor 25 by the two lenses 23a and 23b. . A procedure for creating the composite image will be described.

  FIG. 6 is a diagram illustrating an imaging region by the imaging system. The light / dark pattern image of the medium is divided into a first area 61 and a second area 62 on the medium 1 by the lenses 23a and 23b. The image sensor 25 is divided into a first area 63 and a second area 64. The first area 61 forms an image on the first area 63, and at the same time, the second area 62 forms an image on the second area 64. The first region 61 and the second region 62 partially overlap each other, and this overlapping same region is defined as a common region 65. The common area 65 is divided at both ends on the image sensor 25 and forms an image on the two areas 66 and 67. The lenses 23a and 23b are for reducing the image so that the two images are separated on the image sensor 25 without overlapping. In this manner, the same area (common area 65) of the medium 1 is imaged from a plurality of different directions by a plurality of lenses. On the image sensor 25, there are a plurality of separated regions (regions 66 and 67) corresponding to the common region 65.

  FIG. 7 shows an example of a waveform for one line when only one line is extracted from the image captured by the image sensor 25 along the sub-scanning direction. The horizontal axis represents the position on the image sensor 25, and the vertical axis represents the signal intensity (brightness) in each cell. This waveform includes a waveform 71 corresponding to the first region 61, a waveform 72 corresponding to the second region 62, and a waveform 73 and a waveform 74 corresponding to the common region 65.

  As shown in FIG. 8, the reading control circuit 9 finds a position where the phases of the common portions 73 and 74 coincide by horizontally inverting both the waveform 71 and the waveform 72. Then connect the two waveforms together. Actually, since the two-dimensional images (IM1, IM2) are connected to each other, a position where the phases coincide with each other is similarly found, and the two-dimensional images are connected to each other at an average position. In practice, before performing the joining process, the image is enlarged or reduced by a scaling process to correct the size (magnification). Details will be described later.

  FIG. 9 shows a signal waveform of one line in the image (IM) obtained by connecting the corrected images. A waveform 91 in FIG. 9A is obtained by synthesizing images captured before conveyance, and corresponds to the image 41 before conveyance in FIG. A waveform 92 in FIG. 9B is obtained by synthesizing images captured after the conveyance, and corresponds to the image 51 after the conveyance in FIG. The method for calculating the movement error of the media using these images is as described above.

  In order to make the explanation easy to understand, the image is reversed left and right in order to match the direction of the actual conveyance with the direction of the image. Both can be calculated.

== Media lift correction ==
The medium 1 during printing is configured to keep the distance between the print head 8 and the medium 1 constant by being sucked by the platen 5. However, in reality, there is a case where the media is lifted by about ± 0.3 mm.

  When the media is lifted, the imaging magnification on the image sensor 25 is changed when the reader 6 reads the light / dark pattern of the media. This phenomenon will be described with reference to FIG.

  Respective principal rays (indicated by broken lines) passing from the points 101 to 105 on the medium 1 through the center of the stop 24a are imaged on the points 101a to 105a on the image sensor 25. Here, when the distance between the medium 1 and the lens 23a is increased by ΔZ due to the rising of the medium, the principal rays (indicated by solid lines) from the same points 101 to 105 on the medium 1a whose positions are shifted are image sensors 25. The image is formed on the points 101b to 105b above. Each of these reaching positions is closer to the center point 103a corresponding to the optical axis 103 than the reaching position when there is no lifting of the medium. That is, the larger the distance ΔZ, the smaller the image forming magnification. On the contrary, as the distance ΔZ increases in the minus direction (the media 1 and the lens 23a are closer than the original), the imaging magnification is higher than the original. Then, the moving amount of the light / dark pattern on the image sensor 25 is different from the actual moving amount of the medium, which becomes an error factor of the difference in detecting the moving amount. Therefore, this must be corrected by some method.

  Here, the error of the imaging magnification due to the floating of the medium will be described in more detail. FIG. 11 is a diagram for explaining a change in the phase difference due to the floating of the medium. FIG. 11A shows a case where the distance between the image sensor 25 and the medium 1 is more than the predetermined distance L0 when there is no lifting of the medium by d1. FIG. 11B shows a case where the distance between the image sensor 25 and the medium 1 is equal to the predetermined distance L0. FIG. 11C shows a case where the distance between the image sensor 25 and the medium 1 is closer than the predetermined distance L0 by d2.

  As described above, the light and dark pattern 68 of the common area 65 is imaged in the areas 66 and 67 separately on both ends on the image sensor 25. Since these are images of the same region (common region 65), the signal waveforms are substantially equal. However, the interval between these separated identical waveforms varies depending on the amount of media lifting. In other words, this interval is a phase difference (ΔP) between a plurality of separated areas on the media sensor 25 corresponding to the same area on the medium.

  As shown in FIG. 11B, the interval between the waveforms 71 and 72 corresponding to the light and dark pattern 68 in the common area 65 when the medium is not lifted up (the distance between the medium 1 and the image sensor 25 is the predetermined distance L0). Is L2. L2 is a design reference interval and is a fixed value. As shown in FIG. 11A, when the media is lifted by + d1 and the distance between the media 1 and the image sensor 25 is L0 + d1, the interval L1 between the waveforms 69 and 70 is smaller than the reference interval L2. Conversely, as shown in FIG. 11C, when the media is lifted by −d2 and the distance between the media 1 and the image sensor 25 is L0−d2, the interval L1 between the waveforms 73 and 74 is the reference interval L2. Will be bigger. Note that the interval between positions on the image sensor 25 corresponding to the optical axes of the lenses 23a and 23b is L4. L4 is a design value and is a fixed value.

  As described above, the interval (phase difference) between the separated identical waveforms and the imaging magnification change with correlation. In view of this, an enlargement or reduction process of the image is performed on the image obtained by imaging so as to correct the magnification error corresponding to the change in the imaging magnification.

  A method for calculating the conversion magnification for correcting the magnification error will be described below.

  Assuming that ΔLa and ΔLb are the amounts of change at the two positions of the image formation position of the common area 65 corresponding to the media lift amount + d1, the following relationship (Equation 1) is established.

L1 = L2−ΔLa−ΔLb (Expression 1)
When the common area 65 is set at an equal distance from the optical axes of the lenses 23a and 23b, the common area 65 has the same angle of view with respect to the lenses 23a and 23b. Therefore, since ΔLa = ΔLb, when ΔLa = ΔLb = ΔL, the following relationship (Equation 2) is established.

ΔL = (L1−L2) / 2 (Expression 2)
Since the angle of view of the common area 65 is L4 / 2, the imaging magnification Mo when there is no media lift is as shown in the following (Equation 3).

Mo = (L2-L4) / 2 / (L4 / 2) = (L2-L4) / L4 (Formula 3)
The magnification Md1 shown in FIG. 11A in the case where the media lifting amount = + d1 is as shown in the following (Expression 4).

Md1 = ((L2-L4) / 2-ΔL) / (L4 / 2)
= ((L2-L4) / 2 + (L1-L2) / 2) / (L4 / 2)
= (L1-L4) / L4 (Formula 4)
It can be considered that an image that should originally be imaged at the imaging magnification Mo is formed at an imaging magnification of Md1. Therefore, in order to correct the magnification error due to the floating of the media, the image may be converted at the conversion magnification as shown in the following (Formula 5).

Mo / Md1 = (L2-L4) / (L1-L4) (Formula 5)
In the image processing unit of the reading control circuit 9, the pre-combination images (IM1, IM2) described above are enlarged or reduced at the conversion magnification of (Equation 5) by the image scaling process to obtain a corrected image. . Then, a combination of these corrected images as described above is used as the previous images 41 and 51 to detect the amount of media movement.

  Here, since L2 and L4 are design fixed values, these values are stored in the memory in advance. Therefore, the required conversion magnification can be calculated simply by obtaining the common region interval L1 (phase difference ΔP). Note that (Equation 5) is the same in the case of FIG. 11C.

  In order to calculate the interval L1 (phase difference ΔP), the same method as in the correlation calculation described above can be used. Using a part of the image 70 corresponding to the common area 65 as an image of the correlation window, a correlation value with an image of the same size cut out from the image 69 corresponding to the area 66 is obtained. Then, the search is performed by shifting one pixel at a time, and the distance (number of pixels) between the two when the correlation value becomes an extreme value is defined as an interval L1.

  A summary of the operation procedure described so far is shown in the flowcharts of FIGS. FIG. 12 shows an overall procedure for performing printing while correcting the media transport amount. FIG. 13 shows the details of the procedure (S3 and S7 in FIG. 12) for creating a light and dark pattern corrected image.

<Second Embodiment>
In the above embodiment, the image of the common part of the media is picked up by two lenses arranged eccentrically, and the error of the imaging magnification due to the floating of the medium is corrected based on the phase difference of the image of the common part. It is. On the other hand, in this example, the magnification error is corrected more accurately by taking images of two common areas (same areas) using three lenses. In each case, the amount of floating of the medium is grasped from the phase difference, and the inclination of the paper in the transport direction is obtained using the interval between the two common areas.

  A phenomenon in which a measurement error occurs due to media tilt will be described with reference to FIG.

  The medium 1 (broken line) is in a state where there is no inclination in the transport direction. The chief rays (shown by broken lines) from the points 1201 to 1205 on the medium 1 are imaged on the points 1201a to 1205a on the image sensor 25.

  On the other hand, the medium 1a is inclined by an angle Δθ in the transport direction. Principal rays (indicated by solid lines) from the same points 1201 to 1205 on the medium 1a are imaged on the points 1201b to 1205b on the image sensor 25. That is, like the points 1204 and 1205, the imaging magnification of the portion closer to the image sensor 25 than the original due to the inclination is enlarged. On the contrary, the imaging magnification of a portion farther from the image sensor 25 than the original due to the inclination, such as points 1201 and 1202, is reduced. What is different from the above-described case of media floating is that the imaging magnification differs depending on the area of the media. Therefore, in this embodiment, an error in the imaging magnification that continuously changes due to the inclination of the medium is corrected.

  FIG. 15 is a graph showing an error in the imaging position when the tilt angle Δθ = 1 °. The horizontal axis is the position (mm) on the media surface, and the intersection with the optical axis of the lens 23a is zero. The vertical axis represents the error (μm) in the imaging position. In the region on the left side of the optical axis 1203 in FIG. 14, the medium is further away from the image sensor 25 than the predetermined position, so the imaging magnification is small. Therefore, the distance from the optical axis of the imaging position is small. The value of the imaging position deviation in the graph of FIG. 15 is negative. On the other hand, since the portion on the right side of the optical axis 1203 in FIG. 14 is closer to the image sensor 25 than the predetermined position, the imaging magnification is large. Since the horizontal axis of the graph of FIG. 15 has the optical axis position set to 0, when the image is largely formed on the minus side, the value of the imaging position deviation is also negative.

  Thus, if the inclination angle Δθ of the medium is known, the amount of error corresponding to each part can be calculated by calculating the intersection of the straight line from each point and the sensor. The magnification of the image is corrected using this error amount.

  In order to obtain the inclination angle of the medium 1, the distance to the medium may be obtained at two or more points separated on the medium. This finding method will be described below.

== How to find the distance to the media ==
FIG. 16 is a diagram illustrating a configuration in which an imaging region on the medium 1 is divided into three and imaged using three lenses.

  The areas 1401, 1402, and 1403 on the medium 1 are imaged on the areas 1411, 1412, and 1413 on the image sensor 25 by the corresponding lenses 23a, 23b, and 23c through the diaphragms 24a, 24b, and 24C, respectively. The areas 1401 and 1402 have a first common area 1404 (the same area) where they overlap. The areas 1402 and 1403 have a second common area 1405 (the same area) where they overlap. The first common area 1404 is imaged on the two separated areas 1414a and 1414b on the image sensor 25 by the lens 23a and the lens 23b. Similarly, the second common area 1405 is imaged by two areas 1415a and 1415b separated on the image sensor by the lens 23b and the lens 23c. In this way, a plurality of identical areas (common areas 1404 and 1405) on the medium 1 are respectively imaged from a plurality of different directions by a plurality of lenses.

  Using this relationship, an error in the distance from the image sensor 25 is obtained for each of the common areas 1404 and 1405. The common region 1404 is calculated from the distance (phase difference) between the two separated regions 1414a and 1414b, and the common region 1405 is calculated from the distance (phase difference) between the two separated regions 1415a and 1415b.

  The interval between the two common areas 1404 and 1405 is a fixed value in design and is known in advance. The inclination angle can be obtained by using the distance error obtained in each common area and the interval (fixed value) thereof.

  With reference to FIG. 17, a method for obtaining an error in the distance from the image sensor from the phase difference of the images in the common area will be described.

  The lens 23 a forms an image of the light / dark pattern of the medium 1 on the image sensor 25. Reference numeral 1501 denotes the optical axis of the lens 23a. When there is no lift of the medium, a point 1502 on the medium 1 away from the optical axis forms an image at a point 1503 on the image sensor 25. The distance of the point 1503 from the optical axis 1501 is assumed to be hi. When the medium 1 approaches the position of the medium 1a by approaching Δa in the direction of the image sensor 25, a chief ray (shown by a broken line) from the same point 1502 on the medium 1a forms an image at a point 1503 on the sensor. . The distance from this point to the optical axis 1501 is defined as hi ′. Assuming that the distances from the lens 23 to the medium 1 and the image sensor 25 are a and b, respectively, the media lifting amount Δa can be obtained by the equation shown in FIG.

  The media inclination is obtained by the above procedure. Then, the distribution of the imaging magnification at each position on the image sensor 25 is obtained, and the captured image is inversely converted according to the distribution of the imaging magnification, thereby correcting the imaging magnification error due to the tilt of the media. Other than this correction, the procedure is the same as in the previous embodiment, and thus detailed description thereof is omitted.

  By using the method of each embodiment described above, it is possible to correct an imaging magnification error due to floating of a medium by image processing without using a complicated optical system such as a focusing lens or a variable power lens. This makes it possible to achieve both high accuracy and low cost of the apparatus.

  Although the example in which the moving amount of the medium 1 is detected using the reader 6 has been described above, instead of the medium 1, a moving body (conveying roller or conveying belt) included in the conveying mechanism that conveys the medium 1 in the printer. Similarly, the movement may be detected by imaging and image processing. Further, not only the movement amount but also movement information such as movement speed, movement acceleration, and movement direction may be detected. The moving speed can be obtained from the moving amount per unit time. The moving acceleration can be obtained from the change in speed by differential processing.

  In addition, when the sub-scan is stationary between prints, the medium is not picked up and the amount of movement is detected, but the image is picked up while the medium is moving in the sub-scan direction. Good.

Schematic diagram of the main parts of an inkjet printer Diagram showing the physical configuration of the reader View of the imaging system of the reader from the side Illustration of light and dark pattern acquired before transport Illustration of light and dark pattern acquired before transport Diagram showing the imaging area by the imaging system The figure which shows the example of the waveform for 1 line of the imaged image Diagram explaining the concept of connecting waveforms The figure which shows the example of the waveform for one line of the joined image Diagram explaining magnification error due to media floating Diagram explaining changes in phase difference due to media floating Movement amount detection flowchart Flowchart diagram showing the overall procedure for printing with the carry amount corrected Diagram explaining the occurrence of errors due to media tilt Graph showing the error of the imaging position The figure which shows the imaging area by an imaging system in 2nd Embodiment Diagram explaining how to calculate distance error Figure showing the formula for calculating the amount of media lifting

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Media 2 Conveyance roller 3 Stepping motor 6 Reader 7 Carriage 8 Print head 9 Reading control circuit 10 Controller 11 Conveyance control circuit

Claims (7)

A first step of acquiring an image by imaging the same region of the object from a plurality of different directions using an image sensor;
A second step of obtaining a corrected image by obtaining a phase difference between a plurality of separated regions on the image sensor corresponding to the same region and correcting a magnification of the acquired image based on the phase difference;
In accordance with the movement of the object, the first step and the second step are repeated, and a plurality of correction images obtained are compared to detect the movement information of the object. An object movement detection method. In the first step, a plurality of regions where the same region overlaps as a common region is imaged from different directions, and an image of each region is obtained.
The second step is characterized in that the corrected image is obtained by correcting the magnification of the image in each region based on the obtained phase difference and connecting the corrected images in each region. The object movement detection method according to 1.   The second step is characterized in that an inclination in the moving direction of an object is obtained by obtaining the phase difference for each of the plurality of the same regions, and the magnification is corrected based on the obtained inclination. 3. A method for detecting movement of an object according to 2.   4. The object movement detection method according to claim 1, wherein the third step detects the movement amount of the object by comparing the plurality of corrected images by pattern matching. 5.   5. The object movement detection method according to claim 1, wherein the object is a medium used in a printer or a part of a mechanism for conveying the medium used in the printer. 6. Printing means for transporting media and printing;
An imaging unit having an image sensor that captures an image of the same area of the media from a plurality of different directions;
Image processing means for obtaining a corrected image by obtaining a phase difference between a plurality of separated regions on the image sensor corresponding to the same region and correcting the magnification of the acquired image based on the phase difference;
Detecting means for detecting the conveyance information of the media by repeating the image acquisition by the imaging means and the processing by the image processing means in accordance with the movement of the media, and comparing the plurality of obtained corrected images; ,
A printer comprising control means for controlling conveyance of the media based on detection by the detection means.   The printer according to claim 6, wherein the printing unit performs printing by an inkjet method.

Images