JP2004220371A - Image processing method, image processor, image processing program, and recording medium recorded with image processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To detect a strain in an image, and to precisely correct the strain. <P>SOLUTION: A calibration plate is imaged up in step A1, and the picked-up image is divided into a plurality of individual areas to calculate a transformation conversion coefficient of projection transformation in every of the individual areas. A moving amount of coordinates by the projection transformation is calculated the individual area by the individual area in step A2, based on the transformation coefficient found in step A1. The presence of the strain in the image is diagnosedly determined when the moving amount is larger than a prescribed value. Calibration for calculating a distance per one picture element is carried out based on the distance between two points of a picture element unit in the image and the distance between two points measured actually, in step A3. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an image processing method and an image processing apparatus for performing graphic processing on an image input from an imaging device or the like, and in particular, an image processing method and an image processing apparatus for diagnosing and correcting image distortion, an image processing program, and an image The present invention relates to a recording medium on which a processing program is recorded.
[0002]
[Prior art]
2. Description of the Related Art With the spread of imaging devices such as CCD cameras, image processing for performing various processes based on images obtained by capturing a subject has been performed in factories and homes. For example, in factories, image processing is performed by a digital camera or a scanner, and image processing is performed by a security system using a security camera or the like.
[0003]
Among these, especially in the alignment of parts in the assembly of products and the detection of defective parts in products, it is necessary to realize the position detection and dimensional measurement of objects with high accuracy, and high-precision processing for image processing Is required.
[0004]
The main causes of the deterioration in the accuracy of the image processing include the tilt and lens distortion of the imaging device and the bending of the subject.
[0005]
Such image distortion can be corrected using projective transformation. There are the following three types of typical techniques for correcting image distortion by projective transformation.
[0006]
(1) Various sensors other than the imaging device are used (see Patent Document 1).
Using various sensors, the positional relationship (distance or angle) between the imaging device and the subject is actually measured, and a conversion function is calculated from the measured values. The sensors used include an acceleration sensor, an angular velocity sensor, a magnetic sensor, an ultrasonic sensor, and an infrared sensor.
[0007]
(2) Set manually (see Patent Documents 2 and 3).
Manually set the information required to calculate the conversion function. The necessary information is the coordinates of the four points. The user selects appropriate four points from the image before the conversion, and specifies the coordinates to be moved by these points after the conversion. A conversion function is calculated from the coordinates before and after the conversion.
[0008]
(3) The feature points of the subject are used (see Patent Document 4).
A feature point is searched for from a subject, and a conversion function is calculated so that the feature point moves to a predetermined position. Since the conversion function can be calculated for each subject, it is effective when the positional relationship changes.
[0009]
[Patent Document 1]
Japanese Patent Application Laid-Open No. 2000-307947 [Patent Document 2]
JP-A-7-44698 [Patent Document 3]
JP-A-11-238123 [Patent Document 4]
JP 2001-177716 A
[Problems to be solved by the invention]
In the case of using various sensors other than the imaging device as in (1), it is necessary to separately provide sensors, so that the scale of the device is increased and the cost is increased. In order to perform the correction with high accuracy, the position adjustment between the imaging device and the sensor is complicated.
[0011]
In the case of manual setting as in (2), there is a problem that the burden on the operator is large. In addition, objectivity is poor, for example, the setting changes depending on the operator, and the accuracy is not always high.
[0012]
When using the feature points of the subject as in (3), the subject needs to have appropriate feature points.
[0013]
In any of the above cases, since the entire area is converted by a single conversion function, it is possible to correct the distortion only if the imaging apparatus is tilted, but lens distortion and object deflection may occur. If it is the cause, it cannot be corrected.
[0014]
An object of the present invention is to provide an image processing apparatus and an image processing method that detect distortion of an image and perform distortion correction with high accuracy, an image processing program, and a recording medium that stores the image processing program.
[0015]
[Means for Solving the Problems]
An image capturing step of capturing an image of a calibration plate on which a plurality of marks are arranged such that a distance between at least two adjacent marks is within a predetermined error range,
An extraction step of extracting a parameter related to the arrangement of marks in a region having the least distortion in the captured image,
A dividing step of dividing the captured image into a plurality of individual regions;
Based on the extracted parameters, for each of the individual regions, a detection step of detecting the distortion of the arrangement of the included mark,
A diagnosis step of diagnosing whether or not the entire captured image is distorted based on the detected distortion of the arrangement of the marks.
[0016]
According to the invention, in the imaging step, an image of a calibration plate on which a plurality of marks are arranged such that a distance between at least two adjacent marks is within a predetermined error range is taken.
[0017]
In the extraction step, parameters relating to the arrangement of marks in an area having the least distortion in the captured image are extracted. When an image of the calibration plate is captured using a CCD camera or the like, a parameter relating to a mark captured in the central portion of the lens, that is, a mark in the central portion of the image may be extracted.
[0018]
In the dividing step, the captured image is divided into a plurality of individual regions. By dividing, distortion detection can be performed for each individual area.
[0019]
In the detection step, the distortion of the arrangement of the included marks is detected for each individual area based on the extracted parameters.
[0020]
In the diagnosis step, it is diagnosed based on the detected distortion of the arrangement of the marks whether or not the entire captured image is distorted.
[0021]
As described above, since the distortion is detected for each individual region, it is possible to diagnose not only the tilt of the imaging apparatus but also the distortion of the lens and the distortion of the image due to the bending of the subject.
[0022]
Also, in order to diagnose image distortion, it is only necessary to use a calibration plate, and various sensors are not required.Therefore, even if a distortion diagnosis function is added, the scale of the apparatus can be expanded, and adjustment of a complicated apparatus can be performed. Without doing so, costs can be reduced.
[0023]
Also, in the present invention, the extracting step detects a plurality of distances between adjacent marks in an area having the least distortion, calculates an average of the detected distances as a reference distance, and performs imaging using the reference distance. It is characterized in that the coordinates of all the marks obtained are extracted as ideal coordinates.
[0024]
According to the present invention, in the extraction step, first, a plurality of distances between adjacent marks are detected in an area having the least distortion. The average of the detected distances is calculated as a reference distance, and the coordinates of all the captured marks are extracted as ideal coordinates using the reference distance.
[0025]
A comparison target for diagnosing distortion can be easily acquired without using a sensor or the like separately.
[0026]
Further, in the present invention, the detecting step calculates, for each of the individual regions, a projective transformation coefficient from the ideal coordinates and the coordinates of the imaged mark, and, based on the calculated projective transformation coefficient, It is characterized in that a movement amount is detected.
[0027]
According to the invention, in the detection step, for each individual area, a projective transformation coefficient is calculated from the ideal coordinates and the coordinates of the imaged mark, and the amount of movement of the coordinates by the projective transformation is calculated based on the calculated projective transformation coefficient. Is detected.
[0028]
In this way, since the projective transformation is used, distortion can be detected with high accuracy and easily.
[0029]
Also, in the present invention, in the diagnosis step, the movement amount detected for each individual region is compared with a predetermined threshold for the coordinates of all the regions, and diagnosis is performed based on the number of individual regions where the movement amount is equal to or larger than the threshold. It is characterized by doing.
[0030]
According to the present invention, in the diagnosis step, the amount of movement detected for each individual area is compared with a predetermined threshold, and diagnosis is performed based on the number of individual areas where the amount of movement is equal to or greater than the threshold.
[0031]
As described above, since the distortion of the entire image is diagnosed based on the comparison with the threshold value and the number of individual regions corresponding to the threshold condition, by changing the threshold value and the number of individual regions, the diagnostic accuracy can be improved according to the operator and the subject. Can be changed.
[0032]
Further, the present invention further includes a correction step of correcting the captured image by performing projective transformation based on the projective transformation coefficient when the image photographed in the diagnosis step is diagnosed as being distorted. It is characterized by.
[0033]
According to the present invention, when the image captured in the diagnosis step is diagnosed as being distorted, the captured image is corrected by performing a projection conversion based on the projection conversion coefficient in the correction step.
[0034]
Since the correction is performed based on the projection conversion coefficient calculated for each individual region, it is possible to accurately correct not only the tilt of the imaging apparatus but also the lens distortion and the image distortion due to the bending of the subject. In addition, the amount of calculation required for correction can be reduced by using the projective transformation coefficient calculated in the detection step.
[0035]
The present invention further includes a distance calculating step of calculating a distance per pixel from a distance between marks based on the number of pixels in a captured image and an actually measured distance between marks arranged on the calibration plate. Features.
[0036]
According to the present invention, in the distance calculation step, a distance per pixel is calculated from a distance between marks based on the number of pixels in a captured image and an actually measured distance between marks arranged on the calibration plate. The distance per pixel is called a so-called calibration factor, and by using this calibration factor, the actual size of the subject can be measured from the captured image.
[0037]
Further, the present invention is an image processing program for causing a computer to execute the above image processing method.
[0038]
According to the present invention, it is possible to provide an image processing program for causing a computer to execute the above image processing method.
[0039]
The present invention is also a computer-readable recording medium on which an image processing program for causing a computer to execute the above-described image processing method is recorded.
[0040]
According to the present invention, it is possible to provide a computer-readable recording medium on which an image processing program for causing a computer to execute the above-described image processing method is recorded.
[0041]
Further, the present invention, imaging means for imaging a calibration plate on which a plurality of marks are arranged such that at least the distance between two adjacent marks is within a predetermined error range,
Extraction means for extracting a parameter related to the arrangement of marks in an area having the least distortion in a captured image,
Dividing means for dividing the captured image into a plurality of individual areas,
Based on the extracted parameters, for each of the individual areas, detecting means for detecting the distortion of the arrangement of the included mark,
Diagnostic means for diagnosing whether or not the entire captured image is distorted based on the detected distortion of the arrangement of the marks.
[0042]
According to the present invention, the imaging means images the calibration plate on which a plurality of marks are arranged such that the distance between at least two adjacent marks is within a predetermined error range.
[0043]
The extracting means extracts a parameter relating to the arrangement of marks in an area having the least distortion in the captured image. When an image of the calibration plate is captured using a CCD camera or the like, a parameter relating to a mark captured in the central portion of the lens, that is, a mark in the central portion of the image may be extracted.
[0044]
The dividing unit divides the captured image into a plurality of individual regions. By dividing, distortion detection can be performed for each individual area.
[0045]
The detecting means detects distortion of the arrangement of the included marks for each individual area based on the extracted parameters.
[0046]
The diagnosis unit diagnoses whether or not the entire captured image is distorted based on the detected distortion of the arrangement of the marks.
[0047]
As described above, since the distortion is detected for each individual region, it is possible to diagnose not only the tilt of the imaging apparatus but also the distortion of the lens and the distortion of the image due to the bending of the subject.
[0048]
Also, in order to diagnose image distortion, it is only necessary to use a calibration plate, and various sensors are not required.Therefore, even if a distortion diagnosis function is added, the scale of the apparatus can be expanded, and adjustment of a complicated apparatus can be performed. Without doing so, costs can be reduced.
[0049]
BEST MODE FOR CARRYING OUT THE INVENTION
FIG. 1 is a block diagram showing the configuration of the image processing system 100. The image processing system 100 includes an imaging device 1, an image processing device 2, and a display device 3, and constitutes, for example, a component positioning system.
[0050]
The imaging apparatus 1 is an imaging unit including a CCD (charge coupled device) camera 11, an A / D (analog / digital) converter 12, a camera controller 13, a D / A converter 14, and a frame memory 15. The CCD camera 11 captures an image of a subject and outputs the amount of received light as an analog image signal. The A / D converter 12 converts an analog image signal output from the CCD camera 11 into digital data and outputs it as digital image data. The camera controller 14 stores the digital image data in the frame memory 15 for each frame and outputs the digital image data to the D / A converter 14 for display on the display device 3. The D / A converter converts digital image data output from the camera controller 14 into an analog image signal corresponding to the display device 3 and outputs the analog image signal to the display device 3. The display device 3 is realized by an LCD (Liquid Crystal Display), a CRT (Cathode Ray Tube) display, or the like, and displays an analog image signal output from the imaging device 1.
[0051]
The image processing device 2 includes a CPU (Central Processing Unit) 21 and a RAM (Random).
An access memory (ROM) 22, a read only memory (ROM) 23, and an input / output (I / O) controller 24. The CPU 21 controls the operation of the image processing device 2 based on a control program stored in the ROM 23. Image data being processed, data being calculated, and the like are temporarily stored in the RAM 22. The I / O controller 24 is connected to input devices such as a keyboard and a mouse, and a device for moving components, and controls the input / output data.
[0052]
The CPU 21 and the ROM 23 constitute an extracting unit, a dividing unit, a detecting unit, and a diagnosing unit, acquire image data from the frame memory 15 via the camera controller 14 of the imaging device 2, and execute image processing described later.
[0053]
The present invention diagnoses the distortion of the lens provided in the CCD 11, the optical axis of the imaging device 1 is tilted with respect to the subject, and diagnoses the distortion of the captured image due to the bending of the subject and outputs the corrected image without distortion.
[0054]
Specifically, the calibration plate 4 on which the reference marks are arranged is imaged by the imaging device 1, and the image processing device 2 diagnoses whether or not the captured image is distorted, and corrects the distortion if any.
[0055]
The image processing method executed by the image processing apparatus 2 includes a “distortion diagnosis mode” for diagnosing distortion of a captured image and a “distortion correction mode” for correcting image distortion when it is determined that correction is necessary in the diagnosis mode. ".
[0056]
In the “distortion diagnosis mode”, three processes of teaching, image distortion diagnosis (creating a conversion coefficient conversion table if diagnosis is necessary), correction confirmation, and calibration are performed. In the “distortion correction mode”, the conversion coefficient created at the time of diagnosis is performed. , A conversion table based on the conversion table, distortion correction based on the conversion table, and confirmation of the distortion correction.
[0057]
First, the distortion diagnosis mode will be described. FIG. 2 is a flowchart showing the processing in the distortion diagnosis mode. The distortion diagnosis mode includes three steps. First, teaching using a calibration plate is performed in step A1, diagnosis is made in step A2 as to whether or not an image is distorted, and calibration is performed in step A3.
[0058]
Hereinafter, each step will be described in detail.
FIG. 3 is a flowchart showing the teaching processing in step A1.
[0059]
In step B1, the imaging device 1 images the calibration plate 4 on which the reference marks are arranged. 4 and 5 are diagrams illustrating examples of the calibration plate 4. FIG. FIG. 4 shows the calibration plate 4 using a uniform grating, and FIG. 5 shows the calibration plate 4 using a composite grating. The calibration plate 4 is provided with a circular reference mark, and is arranged with high precision so that the size of the reference mark, the distance between the reference marks (the distance between the mark centers), and the like fall within a predetermined error range. I have.
[0060]
In the case of the uniform grating shown in FIG. 4, the size of the reference marks and the distance between the reference marks are the same for all the arranged reference marks. In the case of the composite grating shown in FIG. 5, there are a plurality of types of reference marks and the distance between the reference marks. For example, as shown in FIG. Is also a small area, and the peripheral area has a large reference mark and a large distance between the reference marks. If a composite grating is used, the grating can be switched according to the imaging device, the subject, and the like, so that it is not necessary to prepare a plurality of types of calibration plates, and the accuracy can be further improved.
[0061]
In step B2, the captured image is displayed on the display device 3, and an area to be used for teaching is designated as a distortion correction window from the displayed image. As shown in FIG. 6, the user specifies the distortion correction window W using an input device such as a mouse or a tablet while viewing the image displayed on the display device 3, as shown in FIG. When a uniform grating is used as the calibration plate 4, all the imaged regions may be used for teaching. In this case, no operation by the user is required.
[0062]
In step B3, the user sets the reference mark array pattern of the entire distortion correction window W. For setting, the number of fiducial marks arranged vertically and horizontally is input. For example, the pattern shown in FIG. 6 is an arrangement pattern of six rows and six columns.
[0063]
In step B4, it is determined whether the calibration plate is a uniform grid or a composite grid. The user selects the type of the calibration plate to be used in advance, and makes a determination based on the selection result. If it is a uniform lattice, the process proceeds to step B6, and if it is a composite lattice, the process proceeds to step B5.
[0064]
In the case of a composite grating, since a plurality of types of reference marks are arranged, it is necessary to distinguish between a reference mark to be used and a reference mark not to be used in a captured image. In step B5, an area where the unused reference marks are arranged is set as a mask window M. For example, as shown in FIG. 7, when the reference mark arranged in the center of the distortion correction window W is not used, the mask window M is set in the same manner as the distortion correction window W.
[0065]
A reference mark to be used is specified as an effective mark array except in the mask window M. For example, in FIG. 7, three are specified from the top, three from the bottom, three from the right, and three from the left.
[0066]
In step B6, a center mark for calculating the reference distance is determined from the reference marks in the distortion correction window W. In the determination of the center mark, the user designates the reference mark with the input device while viewing the displayed image, similarly to the designation of each window.
[0067]
In step B7, a binarization condition for detecting a reference mark is set. The binarization conditions include a binarization threshold, a target black and white designation, an area filter for noise removal, and the like.
[0068]
In step B8, teaching is performed.
The binarization process is performed under the binarization condition set in step B7, and the reference marks in the captured image are labeled, and coordinates (X ′, Y ′) representing each of the reference marks, for example, in the present embodiment, If the reference mark is circular, the coordinates of the center of the circle are detected.
[0069]
When the representative coordinates (X ′, Y ′) are detected for all the reference marks, the representative coordinates of the center mark determined in step B6 are set as immovable coordinates.
[0070]
The distance between the center mark and four reference marks adjacent to the center mark is calculated as the distance between the fixed coordinates and the representative coordinates of each reference mark, and the average value is used as the reference distance. When there is no distortion in the captured image, the distance between all the reference marks is constant. It is assumed that the distance between all the reference marks is the calculated reference distance, and the representative coordinates (ideal coordinates) (X, Y) is calculated.
[0071]
Next, the image distortion diagnosis processing in step A2 will be described. FIG. 8 is a flowchart showing the image distortion diagnosis processing.
[0072]
In step C1, the representative coordinates (X ′, Y ′) of each reference mark detected from the image actually picked up in step B7 of the teaching process and the ideal coordinates (X ′, Y ′) of each reference mark calculated on the assumption that the image has no distortion. X, Y) to obtain the conversion coefficient of the projective conversion formula.
[0073]
The projective transformation equation is represented by equation (1).
_{X = (A 3 · X '} + A 4 · Y' + A 5) / (A 0 · X '+ A 1 · Y' + A 2)
_{Y = (A 6 · X '} + A 7 · Y' + A 8) / (A 0 · X '+ A 1 · Y' + A 2) ... (1)
[0074]
Here, (X, Y) is the coordinate after conversion, and (X ′, Y ′) is the coordinate before conversion, and the conversion coefficients A _{0 to} A ₈ can be calculated by giving the coordinates of four points. . Therefore, the conversion coefficient is calculated for each of the four reference marks used. For example, as shown in FIG. 9, in the case of using the four reference marks K 1 to K 4, _A 0 to A ₈ are calculated as a conversion coefficient of discrete regions 110 surrounded by the broken line, four reference marks K3~ When K6 is used, A ₀ ′ to A ₈ ′ are calculated as the conversion coefficients in the individual region 111 surrounded by the broken line. When the four reference marks used in this manner are shifted from left to right, the transform coefficients in each individual area are obtained, and the transform coefficients of the rightmost individual area are calculated. Then, the target individual area is shifted downward, and then left to right again. To calculate the conversion coefficients of all the individual areas.
[0075]
In step C2, the movement amount is calculated for the coordinates of the entire area. The movement amount is a distance between the coordinates before the conversion and the coordinates after the conversion, and indicates that the larger the movement amount, the greater the distortion of the captured image.
[0076]
Specifically, the coordinates after the conversion are calculated using the conversion coefficients calculated in step C1, and the movement amount is obtained.
[0077]
In step C3, the user inputs and sets an allowable value of the movement amount as the threshold value. Since the greater the amount of movement, the greater the distortion, the amount of distortion up to which no distortion is considered is set by the amount of movement.
[0078]
In Step C4, it is determined whether the distortion needs to be corrected. All the movement amounts calculated in step C2 are compared with the allowable value set in step C3. If only one movement amount exceeds the allowable value, it is determined that correction is necessary, and the process proceeds to step C5. If all the movement amounts are equal to or less than the allowable value, it is determined that no correction is necessary, and the process proceeds to step C6.
[0079]
The result of the determination in step C4 is stored as a correction flag because it is necessary for a correction process described later. In step C5, "correction required" is stored in the correction flag, and in step C6, "correction unnecessary" is stored in the correction flag.
[0080]
FIG. 10 is a flowchart illustrating the calibration process.
In step D1, the distance between two reference marks on the calibration plate is measured by the number of pixels. This may be measured using the image captured in the above-described processing. Alternatively, a separately calibrated sample may be taken, and the distance between two points in the image may be measured.
[0081]
In step D2, the corrected distance is calculated using the conversion coefficient calculated in the image distortion diagnosis process, and the difference between the pre-corrected distance and the corrected distance is calculated. The effectiveness of the correction is confirmed based on this difference.
[0082]
The validity is confirmed, for example, by displaying a difference value, and the user confirms the displayed difference value.
[0083]
In Step D3, it is determined whether or not the confirmation of the correction validity has been completed. If it has been completed, the process proceeds to Step D4, and if not, the process returns to Step D1.
[0084]
When it is determined in step C4 that the correction is unnecessary, the processing of steps D2 and D3 for confirming the validity may not be performed.
[0085]
In step D4, the actual measured value of the distance measured in step D1 is input and set. In the calibration plate, since the distance between the reference marks is fixed, the value may be input.
[0086]
In step D5, a calibration factor is calculated from the measured value and the number of pixels set in step D4. The calibration factor is a length per unit pixel such as Xm per pixel.
[0087]
Hereinafter, the distortion correction mode will be described.
In the distortion correction mode, there are a process of creating a conversion table based on the conversion coefficients created in the distortion diagnosis mode, a process of correcting distortion based on the conversion table, and a process of confirming distortion correction.
[0088]
1) Conversion Table Creation Processing In order to store the conversion tables for all areas, a huge storage capacity is required. For example, when the power is turned on or when the user specifies, conversion coefficients are calculated and converted. Create a table. Note that the conversion table is created only in the image distortion diagnosis processing when “correction required” is stored in the correction flag.
[0089]
Specifically, the conversion table is a table in which coordinates before conversion and coordinates after conversion are described.
[0090]
As described above, by storing the conversion coefficients, the storage capacity can be reduced, and the amount of calculation can be reduced by creating the conversion table from the conversion coefficients.
[0091]
2) Image distortion correction processing Using the conversion table created in the conversion table creation processing, each pixel of the image captured by the imaging device 1 is moved to create a corrected image.
[0092]
3) Correction Check Processing The image before correction and the image after correction are displayed on the display device 3, and the user checks the effect of the correction processing.
[0093]
FIG. 11 is a display example of an image before and after correction. FIG. 11A shows an image before correction, in which the four corners are particularly distorted. FIG. 11B shows an image after correction, in which all reference marks are arranged at equal intervals.
[0094]
As described above, since the conversion function is calculated for each area, the image distortion can be corrected not only for the tilt of the imaging apparatus 1 but also for the lens distortion and the object bending.
[0095]
Further, since it is not necessary to provide a separate sensor for performing diagnosis and correction of image distortion, even if an image diagnosis and correction function is added, the apparatus scale is not enlarged, and adjustment of a complicated apparatus is not performed. Costs can be reduced.
[0096]
Further, since a calibration plate on which a reference mark is arranged in advance is used, there is no need to manually input a reference point, so that the burden on the operator can be reduced and the correction accuracy can be improved.
[0097]
Further, another embodiment of the present invention is an image processing program for causing a computer to function as the image processing device 2, and a computer-readable recording medium storing the image processing program. Thus, the image processing program and the recording medium storing the image processing program can be provided in a portable manner.
[0098]
The image processing program is executed by reading the recording medium by a program reading device provided in a printer or a computer system.
[0099]
As an input means of the computer system, a flatbed scanner, a film scanner, a digital camera, or the like may be used. The computer system includes these input means, a computer that executes image processing by loading a predetermined program, an image display device such as a CRT display or a liquid crystal display that displays the processing results of the computer, and a computer processing It consists of a printer that outputs the results on paper or the like. Further, a modem or the like is provided as communication means for connecting to a server or the like via a network.
[0100]
The recording medium is not limited to a medium that can be read by a program reading device, but may be a memory of a microcomputer, for example, a ROM. The recorded program may be accessed and executed by a microprocessor, or a program read from a recording medium may be downloaded to a program storage area of a microcomputer and executed. It is assumed that the microcomputer has this download function in advance.
[0101]
Specific examples of the recording medium include a tape system such as a magnetic tape and a cassette tape, a magnetic disk such as a flexible disk and a hard disk, and a CD-ROM (Compact Disc-Read Only Memory) / MO (Magneto Optical) disk / MD (MD). Mini Disc) / DVD (Digital Versatile Disc) and other optical disc disc systems, IC (Integrated Circuit) card (including memory card) / optical card and other card systems, or mask ROM, EPROM (Erasable Programmable Read Only Memory) (Electrically Erasable Programmable Read Only Memory ), It is a medium fixedly carrying the program, including semiconductor memories such as a flash ROM.
[0102]
In the present embodiment, the computer may have a system configuration connectable to a communication network including the Internet, and may download the image processing program via the communication network. When the program is downloaded from the communication network as described above, the download function may be provided in the computer in advance, or may be installed from another recording medium. The download program may be executed through a user interface, or may be a predetermined URL (Uniform Resource).
Locator).
[0103]
【The invention's effect】
As described above, according to the present invention, since distortion is detected for each individual region, it is possible to diagnose lens distortion and image distortion due to bending of a subject in addition to tilt of the imaging device. Also, in order to diagnose image distortion, it is only necessary to use a calibration plate, and various sensors are not required.Therefore, even if a distortion diagnosis function is added, the scale of the apparatus can be expanded, and adjustment of a complicated apparatus can be performed. Without doing so, costs can be reduced.
[0104]
Further, according to the present invention, a comparison target for diagnosing distortion can be easily obtained without separately using a sensor or the like.
[0105]
Further, according to the present invention, since projective transformation is used, distortion can be detected with high accuracy and easily.
[0106]
According to the present invention, since the distortion of the entire image is diagnosed based on the comparison with the threshold value and the number of the individual regions corresponding to the condition of the threshold value, the threshold value and the number of the individual regions are changed to change the threshold value and the number of the individual regions according to the worker and the subject. The diagnostic accuracy can be changed.
[0107]
Further, according to the present invention, since the correction is performed based on the projection conversion coefficient calculated for each individual area, it is possible to accurately correct lens distortion and image distortion due to bending of the subject in addition to tilt of the imaging device. Can be. In addition, the amount of calculation required for correction can be reduced by using the projective transformation coefficient calculated in the detection step.
[0108]
Further, according to the present invention, by using a so-called calibration factor, the actual dimensions of the subject can be measured from the captured image.
[0109]
Further, according to the present invention, it is possible to provide an image processing program for causing a computer to execute the above-described image processing method, and a computer-readable recording medium storing the image processing program.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of an image processing system 100.
FIG. 2 is a flowchart illustrating processing in a distortion diagnosis mode.
FIG. 3 is a flowchart showing a teaching process in step A1.
FIG. 4 is a view showing an example of a calibration plate 4;
FIG. 5 is a view showing an example of a calibration plate 4;
FIG. 6 is a diagram illustrating a display example when a distortion correction window W is designated.
FIG. 7 is a diagram showing a display example when specifying a mask window M and specifying an effective array.
FIG. 8 is a flowchart illustrating an image distortion diagnosis process.
FIG. 9 is a diagram illustrating an example of an individual area.
FIG. 10 is a flowchart illustrating a calibration process.
FIG. 11 is a display example of an image before and after correction.
[Explanation of symbols]
REFERENCE SIGNS LIST 1 imaging device 2 image processing device 3 display device 11 CCD camera 12 A / D converter 13 camera controller 14 D / A converter 15 frame memory 21 CPU
22 RAM
23 ROM
24 I / O Controller 100 Image Processing System

Claims (9)

An imaging step of imaging a calibration plate on which a plurality of marks are arranged such that a distance between at least two adjacent marks is within a predetermined error range;
An extraction step of extracting a parameter related to the arrangement of marks in a region having the least distortion in the captured image,
A dividing step of dividing the captured image into a plurality of individual regions;
Based on the extracted parameters, for each of the individual regions, a detection step of detecting the distortion of the arrangement of the included mark,
A diagnostic step of diagnosing whether or not the entire captured image is distorted based on the detected distortion of the arrangement of the marks. The extraction step detects a plurality of distances between adjacent marks in an area having the least distortion, calculates an average of the detected distances as a reference distance, and uses the reference distance to calculate all of the captured marks. 2. The image processing method according to claim 1, wherein the coordinates are extracted as ideal coordinates. The detecting step calculates, for each of the individual regions, a projective transformation coefficient from the ideal coordinates and the coordinates of the imaged mark, and detects a moving amount of the coordinate by the projective transformation based on the calculated projective transformation coefficient. 3. The image processing method according to claim 2, wherein: The diagnosing step is characterized in that, for the coordinates of all the regions, the movement amount detected for each individual region is compared with a predetermined threshold, and diagnosis is performed based on the number of individual regions in which the movement amount is equal to or larger than the threshold. The image processing method according to claim 3. When the image captured in the diagnosis step is diagnosed as being distorted, the image processing apparatus further includes a correction step of correcting the captured image by performing projective transformation based on the projective transformation coefficient. Item 5. The image processing method according to item 3 or 4. A distance calculating step for calculating a distance per pixel from a distance between marks based on the number of pixels in a captured image and an actually measured distance between marks arranged on the calibration plate. 6. The image processing method according to any one of 1 to 5. An image processing program for causing a computer to execute the image processing method according to claim 1. A computer-readable recording medium storing an image processing program for causing a computer to execute the image processing method according to claim 1. Imaging means for imaging a calibration plate on which a plurality of marks are arranged such that a distance between at least two adjacent marks is within a predetermined error range;
Extraction means for extracting a parameter related to the arrangement of marks in an area having the least distortion in a captured image,
Dividing means for dividing the captured image into a plurality of individual areas,
Based on the extracted parameters, for each of the individual areas, detecting means for detecting the distortion of the arrangement of the included mark,
A diagnostic means for diagnosing whether or not the entire captured image is distorted based on the detected distortion of the arrangement of the marks.

Images