JP5972715B2 - Image clipping method and image acquisition apparatus - Google Patents

Description

  The present invention relates to an image cutout method and an image acquisition device for cutting out image data for each depression from imaged image data of a sample container having a plurality of depressions.

  In the process of researching and developing pharmaceuticals, screening is performed to narrow down compounds that are drug candidates. In screening, for example, a plurality of culture solutions containing cells are prepared, and the cells are cultured by adding compounds to each culture solution while changing various conditions. Then, based on the culture state of the cells, the compounds that are drug candidates are narrowed down.

  In such screening, conventionally, the culture state of cells has been evaluated by optical measurement. Specifically, a method of measuring the intensity of fluorescence by fluorescing a specific molecule (DNA, protein, chemical substance, etc.) in the cell, or a method of measuring the absorption of a specific molecule in the cell is used. It was. However, these optical measurements require expensive reagents.

  In recent years, three-dimensional culture in which cells are three-dimensionally cultured has been performed in order to examine the effects of pharmaceuticals in an environment closer to the living body. In such a three-dimensional culture, the state of cell clusters (spheroids) grown three-dimensionally is an important observation target. However, the above optical measurement is not suitable for observing cell clumps.

  For this reason, in recent years, an attempt has been made to develop an apparatus for observing the culture state of a cell by photographing the cell with high resolution without using a reagent. In this apparatus, a well plate having a plurality of depressions for culture is photographed every predetermined time, and an image for each depression is cut out from the obtained photographed image. Then, the cultured state of the cells in each depression is evaluated by comparing and analyzing the cut out images over time.

  Here, when the well plate is set in the apparatus, a slight shift may occur in the horizontal position or the rotational position of the well plate. Further, the well plate itself may be slightly expanded and contracted. For this reason, in order to accurately cut out the image of each depression from the photographed image of the well plate, it is necessary to set the cutout position in consideration of these deviations. For example, Patent Document 1 and Patent Document 2 describe conventional techniques in other fields for correcting a deviation of a captured image.

JP 2004-192506 A JP 2012-47463 A

  As a method for setting the cutout position of the image, for example, it is conceivable to specify the position of each depression by pattern matching. However, the well plate contains an empty recess, a recess that holds only the culture solution, and a recess that holds the culture solution and cells. In addition, the culture state of the cells and the scattering state of the illumination light by the culture solution are also different for each depression. For this reason, in order to perform pattern matching with high accuracy, it is necessary to prepare a large number of reference images.

  In addition, a plurality of recesses are arranged close to the well plate. For this reason, even if a large number of reference images are prepared, each reference image does not always match the intended depression. For example, a reference image for a certain depression may be matched with another depression with a higher matching score than the depression.

  Thus, it is difficult to specify the positions of all the depressions only by pattern matching. In particular, as the number of recesses in the well plate increases and each recess becomes smaller, pattern matching becomes more difficult and image processing becomes more complicated.

  The present invention has been made in view of such circumstances, and provides an image cutout method and an image acquisition apparatus that can obtain a cutout image with high accuracy by correcting rotational deviation and positional deviation of a captured image. For the purpose.

  In order to solve the above-mentioned problem, the first invention of the present application is an image cutout method for cutting out image data for each depression from photographed image data of a sample container having a plurality of depressions, and a) an ideal of the sample container In the data, a step of designating a reference point and at least two points of interest corresponding to at least two depressions; b) determining relative coordinates of the reference point with each of the points of interest as an origin; c) in the captured image data of the sample container, a step of identifying an actual target point corresponding to each of the target points; d) a plurality of the actual target points and the relative coordinates corresponding to the actual target points. A step of obtaining a reference point candidate position corresponding to the reference point, and obtaining a rotational deviation amount and a positional deviation amount of the photographed image data from the reference point candidate position; and e) a preset clipping. And a step of cutting out image data for each depression from the photographed image data using the cutout position after correction based on the rotational deviation amount and the positional deviation amount. is there.

  The second invention of the present application is the image clipping method of the first invention, wherein the step d) is based on d-1) a plurality of the actual attention points and the relative coordinates corresponding to the actual attention points. A step of repeating a voting process for obtaining a set of a plurality of reference point candidate positions corresponding to the reference point a plurality of times while rotating the photographed image data by a minute angle, and d-2) of the plurality of voting processes And a step of obtaining a rotation angle of the photographed image data when a set of a plurality of reference point candidate positions is most concentrated as an amount of rotational deviation of the photographed image data.

  A third invention of the present application is the image clipping method according to the second invention, wherein in the step d-1), the voting is performed while changing the rotation angle of the photographed image data and the expansion / contraction ratio of the photographed image data. The process is repeated a plurality of times, and in step d-2), among the plurality of voting processes, the rotation angle and the expansion / contraction rate of the captured image data when the plurality of reference point candidate position sets are most concentrated, Respectively obtained as the rotational deviation amount and the expansion / contraction error of the photographed image data, and in the step e), an image for correcting the preset clipping position based on the rotational deviation amount, the expansion / contraction error, and the positional deviation amount. This is a cutting method.

  A fourth invention of the present application is the image clipping method of the second invention or the third invention, wherein the step d) includes d-3) a plurality of sets of reference point candidate positions among the plurality of voting processes. An image clipping method further comprising a step of obtaining a difference between an average coordinate of the plurality of reference point candidate positions when most concentrated and a coordinate of the reference point in the ideal data as a positional deviation amount of the captured image data. is there.

  A fifth invention of the present application is any one of the image cutting methods from the first invention to the fourth invention, and in the step b), polar coordinates of the reference point with the point of interest as the origin are used as the relative coordinates. This is a desired image clipping method.

  A sixth invention of the present application is any one of the image cutting methods from the first invention to the fifth invention, and in step a), a point corresponding to at least two empty depressions is designated as the point of interest. The step c) is an image cutout method for specifying the actual point of interest by pattern matching using reference image data representing an empty depression.

  A seventh invention of the present application is the image clipping method of the sixth invention, wherein in the step a), points corresponding to all depressions in the ideal data are designated as the points of interest, and the step c) Then, it is an image cut-out method for specifying an actual point of interest corresponding to an empty depression in the captured image data by pattern matching using reference image data representing an empty depression.

  An eighth invention of the present application is an image acquisition device that takes an image of a sample container having a plurality of depressions and cuts out image data for each depression from the obtained captured image data, and holds the sample container horizontally. Unit, a light projecting unit that irradiates light to the sample container, an imaging unit that captures the sample container and obtains captured image data of the sample container, and an image of each depression from the captured image data An image processing unit that cuts out data, and the image processing unit has a relative relative of a predetermined reference point with the origin of each of at least two points of interest corresponding to at least two depressions in the ideal data of the sample container. Corresponding to a relative coordinate calculation unit for obtaining coordinates, an actual focused point identifying unit that identifies an actual focused point corresponding to each of the focused points, a plurality of actual focused points, and the actual focused points in the captured image data A deviation amount calculation unit that obtains a reference point candidate position corresponding to the reference point based on the relative coordinates and obtains a rotational deviation amount and a positional deviation amount of the captured image data from the reference point candidate position; An image cutout unit that corrects the set cutout position based on the rotational deviation amount and the positional deviation amount, and uses the cutout position after correction to cut out image data for each depression from the captured image data; Is an image acquisition device.

  According to the first invention of the present application, it is possible to obtain a cutout image with high accuracy in consideration of rotational deviation and positional deviation of captured image data.

  In particular, according to the second invention of the present application, it is possible to appropriately determine the rotational deviation amount of the captured image data based on the variation of the plurality of reference point candidate positions.

  In particular, according to the third invention of the present application, it is possible to appropriately obtain the rotational deviation amount and the expansion / contraction error of the photographed image data based on the variation of the plurality of reference point candidate positions.

  In particular, according to the fourth invention of the present application, it is possible to appropriately determine the amount of positional deviation of the captured image data.

  In particular, according to the fifth invention of the present application, the calculation amount of step d) can be reduced by employing polar coordinates.

  In particular, according to the sixth invention of the present application, it is possible to perform highly accurate pattern matching using an empty recess. Therefore, the actual attention point can be specified with high accuracy.

  In particular, according to the seventh invention of the present application, it is possible to extract an empty depression from all the depressions using pattern matching and specify an actual point of interest corresponding to the empty depression.

  Further, according to the eighth invention of the present application, it is possible to obtain a cutout image with high accuracy in consideration of rotational deviation and positional deviation of captured image data.

It is a perspective view which shows an example of a well plate. It is the figure which showed notionally the structure of the image acquisition apparatus. It is a flowchart which shows the flow of an image process when cutting out several hollow part image data from picked-up image data. It is the figure which showed the example of ideal data. It is the figure which showed the example of table data. It is the figure which showed the example of picked-up image data. It is the figure which showed the example of picked-up image data.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings.

<1. Configuration of Image Acquisition Device>
FIG. 1 is a perspective view showing an example of a well plate 9 set in the image acquisition apparatus 1. The well plate 9 is a substantially plate-like sample container having a plurality of recesses 91. As the material of the well plate 9, for example, a transparent resin that transmits light is used. As shown in FIG. 1, the plurality of recesses 91 are regularly arranged on the upper surface of the well plate 9. The depth of each recess 91 is longer than the diameter of the opening located at the top of each recess 91. In the example of FIG. 1, the cells are held together with the culture solution 92 in each of the recesses 91 arranged in the inner region (the region surrounded by the broken line in FIG. 1) 90 excluding the outermost periphery among the plurality of recesses 91. Has been. In addition, compounds having different concentrations and compositions are added to the culture solution 92 in each recess 91.

  In the well plate 9 having a plurality of recesses 91, it is generally recommended to use only the recesses 91 arranged in the inner region 90 excluding the outermost periphery. This is because if all the recesses 91 of the well plate 9 are used, a thermal gradient is generated between the outermost recess 91 and the recess 91 in the inner region 90, and the cell culture conditions tend to be uneven. is there. Also in this embodiment, the recessed part 91 arrange | positioned at the outermost periphery is prohibited. The use-prohibited depression 91 remains in an empty state that does not hold the cells and the culture solution 92 even when the image acquisition device 1 captures an image.

  FIG. 2 is a diagram conceptually showing the configuration of the image acquisition apparatus 1 according to an embodiment of the present invention. The image acquisition device 1 is a device that takes an image of the well plate 9 and cuts out image data (hereinafter referred to as “recessed portion image data”) D2 for each recessed portion 91 from the obtained captured image data D1. The image acquisition device 1 is used, for example, in a screening process for narrowing down compounds that are drug candidates in the field of drug research and development. The person in charge of the screening process acquires a plurality of depression image data D2 of each well plate 9 using the image acquisition device 1. Then, the utility of the compound added to the culture solution 92 is verified by comparing and analyzing the culture state of the cells in each recess 91 based on the acquired plurality of recess image data D2.

  As shown in FIG. 2, the image acquisition apparatus 1 of the present embodiment includes a plate holding unit 10, a light projecting unit 20, an imaging unit 30, a display unit 40, an input unit 50, and a computer 60. The plate holding unit 10, the light projecting unit 20, the imaging unit 30, the display unit 40, and the input unit 50 are each electrically connected to the computer 60.

  The plate holding unit 10 is a mounting table that holds the well plate 9. The well plate 9 is set on the plate holding portion 10 in a horizontal posture with the bottom portion on the lower side. The plate holder 10 has a drive mechanism (not shown) that moves the well plate 9 in the lateral direction. The drive mechanism includes, for example, a motor and a power transmission mechanism such as a ball screw that transmits the driving force of the motor. When the drive mechanism is operated, the well plate 9 moves relative to the imaging unit 30, and the recess 91 that enters the field of view of the imaging unit 30 is replaced.

  The light projecting unit 20 and the imaging unit 30 are respectively disposed above and below the well plate 9 held by the plate holding unit 10. The imaging unit 30 is realized by, for example, a line sensor having an optical system such as a lens and an imaging element such as a CCD or a CMOS. When photographing the well plate 9, the imaging unit 30 photographs the part of the well plate 9 while irradiating white light from the light projecting unit 20 toward a part of the well plate 9. Then, the well plate 9 is moved in the horizontal direction and the same photographing is repeated. As a result, captured image data D1 of the entire well plate 9 is acquired.

  The light projecting unit 20 may be anything that irradiates light to the well plate 9. Therefore, the light source itself of the light projecting unit 20 may be arranged at a position off the upper side of the well plate 9 and the well plate 9 may be irradiated with light via an optical system such as a mirror. Further, the light projecting unit 20 may be disposed below the well plate 9, and the imaging unit 30 may be disposed above the well plate 9. Further, the light reflected by the well plate 9 may be configured to enter the imaging unit 30.

  The display unit 40 is a part that displays various types of information related to the processing of the image acquisition device 1 (photographed image data D1, hollow image data D2, ideal data D3, GUI, and the like). The display unit 40 is realized by a display device such as a liquid crystal display, for example. The input unit 50 is a part for inputting various commands to the computer 60. The input unit 50 is realized by a keyboard or a mouse, for example. The user of the image acquisition apparatus 1 can transmit various commands to the computer 60 by operating the input unit 50 while checking the display unit 40.

  Note that both the function of the display unit 40 and the function of the input unit 50 may be realized by a single device such as a touch panel display.

  The computer 60 has both a function as a control unit that controls the operation of each unit of the image acquisition device 1 and a function as an image processing unit that cuts out a plurality of depression image data D2 from the acquired captured image data D1. Have. The computer 60 includes a CPU and a memory, and the operation of the drive mechanism, the light projecting unit 20, and the imaging unit 30 described above is performed by the CPU operating in accordance with preset computer programs, input signals, and various data. To control. Thereby, photographing of the well plate 9 in the image acquisition device 1 is executed.

  As conceptually shown in FIG. 2, the computer 60 of the present embodiment includes a storage unit 61, a relative coordinate calculation unit 62, a matching processing unit 63, a deviation amount calculation unit 64, and an image cutout unit 65. Yes. The functions of the relative coordinate calculation unit 62, the matching processing unit 63, the deviation amount calculation unit 64, and the image cutout unit 65 are calculated by the CPU of the computer 60 while referring to the computer program 70 stored in the storage unit 61. It is realized by doing.

  The storage unit 61 is a part that stores various data handled in the image acquisition apparatus 1. The storage unit 61 is realized by a storage device such as a hard disk drive or a RAM, for example. As shown in FIG. 2, the storage unit 61 may be a part of hardware constituting the computer 60, or may be an external storage device connected to the computer 60.

  As shown in FIG. 2, in the storage unit 61 of the present embodiment, a computer program 70, ideal data D3 of the well plate 9, and reference image data D4 of the recess 91 used for matching processing described later are stored in advance. Has been. The storage unit 61 also stores captured image data D1 acquired by the imaging unit 30, table data D5 in which relative coordinates C described later are recorded, and a plurality of depression image data D2 cut out from the captured image data D1. Is done.

<2. Image data cutout process>
FIG. 3 is a flowchart showing a flow of image processing when a plurality of depression image data D2 are cut out from the photographed image data D1 of the well plate 9. Hereinafter, the flow of the image processing will be described with reference to FIGS. 2 and 3.

  In order to cut out the plurality of depression image data D2 from the photographed image data D1, the computer 60 first refers to the ideal data D3 of the well plate 9 stored in the storage unit 61. As the ideal data D3, for example, specification data of the well plate 9 having no rotation deviation, position deviation, and expansion / contraction error is used. However, image data of the well plate 9 photographed in advance may be used as the ideal data D3 with small rotational deviation, positional deviation, and expansion / contraction error. FIG. 4 is a diagram showing an example of the ideal data D3.

  First, the user of the image acquisition device 1 sets a reference point Oi in the ideal data D3 (step S1). Specifically, while referring to the ideal data D3 displayed on the display unit 40, the input unit 50 is operated to set one point on the ideal data D3 as the reference point Oi. For example, as shown in FIG. 4, the center point of the well plate 9 in the ideal data D3 may be designated as the reference point Oi.

  However, the position of the reference point Oi is not necessarily the center of the well plate 9. In this step S1, an arbitrary point on the ideal data D3 can be designated as the reference point Oi. In addition, the timing for designating the reference point Oi does not necessarily have to be immediately before step S2. For example, information on the reference point Oi designated in advance may be held in the storage unit 61.

  Next, the user of the image acquisition device 1 selects two or more depressions 91 from the plurality of depressions 91 included in the ideal data D3. Specifically, while referring to the ideal data D3 displayed on the display unit 40, the input unit 50 is operated to arbitrarily select two or more depressions 91 on the ideal data D3. Then, the computer 60 designates a point corresponding to each selected depression 91 as the point of interest Wi (step S2).

  In the example of FIG. 4, the recesses 91 at the four corners of the well plate 9 in the ideal data D3 are selected. And the center point of each of the four selected hollows 91 is designated as the points of interest Wi (1) to Wi (4). Note that the point of interest Wi may be a point at a certain relative position with respect to each recess 91. Therefore, the point of interest Wi does not necessarily have to be the center point of the recess 91.

  In step S2, at least two points of interest Wi may be designated. That is, the user may select only some of the recesses 91 of the well plate 9 or may select all of the recesses 91 of the well plate 9.

  However, in step S5, which will be described later, in order to increase the accuracy of pattern matching, it is preferable to select a useless depression (empty depression in the actual well plate 9) 91 in step S2. Further, in steps S5 to S7, which will be described later, in order to reduce an error when calculating various deviation amounts, a plurality of points of interest Wi include a point as far as possible from the reference point Oi. ,preferable. From these viewpoints, in the example of FIG. 4, four depressions are selected from the use-prohibited depressions 91 arranged on the outermost periphery.

  When the reference point Oi and the plurality of points of interest Wi are designated, the relative coordinate calculation unit 62 of the computer 60 obtains the relative coordinates C of the reference point Oi with each point of interest Wi as the origin (step S3). Then, the information of the relative coordinate C is stored as table data D5. FIG. 5 is a diagram showing an example of the table data D5 obtained in step S3. In the table data D5 of FIG. 5, four relative points C (1) to Wi (1) to Wi (4) and relative coordinates C (1) to the reference point Oi having the respective target points Wi (1) to Wi (4) as origins. C (4) is associated one-to-one.

  In the example of FIG. 5, the relative coordinates C are polar coordinates (coordinates defined by the distance r and the deviation angle θ) of the reference point Oi with each point of interest Wi as the origin. If polar coordinates are employed, the calculation of the voting process for obtaining the reference point candidate position Op while changing the rotation angle in step S5 described later becomes simple. For this reason, if polar coordinates are adopted, there is a merit that the amount of calculation can be reduced as compared with the case where orthogonal coordinates are adopted. However, the relative coordinates C may be orthogonal coordinates such as xy coordinates.

  Subsequently, the matching processing unit 63 reads the reference image data D4 stored in the storage unit 61. For the reference image data D4, a typical image showing a single depression 91 is used. The matching processing unit 63 detects the positions of the plurality of depressions 91 selected in step S2 in the captured image data D1 by pattern matching using the reference image data D4. Then, the detected center point of each depression 91 is specified as the actual focus point Wr corresponding to the focus point Wi described above (step S4).

  FIG. 6 is a diagram illustrating an example of the captured image data D1. In the example of FIG. 6, the positions of the recesses 91 at the four corners of the well plate 9 are detected by pattern matching. And the center point of each of the four detected hollows 91 is specified as actual attention points Wr (1) to Wr (4).

  When the use-prohibited depression 91 is selected in step S2 described above, pattern matching may be performed using the reference image data D4 representing the empty depression 91 in step S4. Since the empty depression 91 is not affected by the meniscus or refractive index of the culture solution, the change in appearance is smaller than that of the depression 91 holding the culture solution 92. For this reason, if the empty hollow part 91 is utilized, a highly accurate pattern matching can be performed. Therefore, the actual attention point Wr can be specified with high accuracy.

  Further, when performing pattern matching, it is preferable to perform collation with reference image data D4 only around the ideal position of the recess 91 selected in step S2. By doing so, the time required for pattern matching can be further shortened.

  In step S4, it is sufficient that at least two recesses 91 can be detected. That is, in the pattern matching, it is not always necessary to detect all the recesses 91 selected in step S2. As a result of pattern matching, a predetermined number of depressions 91 showing the highest matching score may be used as detection results.

  When the plurality of actual attention points Wr are specified, the deviation amount calculation unit 64 reads the table data D5 stored in the storage unit 61. Then, a reference point candidate position Op corresponding to the reference point Oi is obtained based on each actual attention point Wr and the relative coordinates C corresponding to the actual attention point Wr. As a result, a set of a plurality of reference point candidate positions Op is obtained. In the example of FIG. 6, based on the four actual attention points Wr (1) to Wr (4) and the corresponding relative coordinates C (1) to C (4), four reference point candidate positions Op ( A set of 1) to Op (4) is required.

  A process for obtaining a set of a plurality of reference point candidate positions Op based on the plurality of actual target points Wr and the relative coordinates C corresponding to the actual target point Wr is hereinafter referred to as “voting process”.

  If the photographed image data D1 has a rotational shift or an expansion / contraction error, the plurality of reference point candidate positions Op obtained from the actual attention points Wr do not match as shown in FIG. The deviation amount calculation unit 64 performs the above-described voting process while rotating or expanding / contracting the captured image data D1 by a minute angle. That is, the voting process described above is repeated a plurality of times while changing the combination of the rotation angle and the expansion / contraction rate of the photographed image data D1 (step S5). Of the plurality of voting processes, the set of the plurality of reference point candidate positions Op is most concentrated (when the variation in the coordinates of the plurality of reference point candidate positions Op is minimized). The rotation angle and the expansion / contraction rate are obtained as the rotational deviation amount and expansion / contraction error of the captured image data D1 (step S6).

  In the present embodiment, when the voting process is performed a plurality of times, the two parameters of the rotation angle and the expansion / contraction rate are changed. However, when it can be considered that there is almost no expansion / contraction error, only the rotation angle may be changed.

  Subsequently, the deviation amount calculation unit 64 obtains the average coordinates of the plurality of reference point candidate positions Op when the set of the plurality of reference point candidate positions Op is most concentrated, and takes a point defined by the average coordinates as a captured image. The reference point in the data D1. Then, a difference in coordinates between the reference point in the captured image data D1 and the reference point Oi in the ideal data D3 is obtained as a positional deviation amount in the horizontal plane of the captured image data D1 (step S7).

  Thereafter, the image cutout unit 65 corrects the cutout position set in advance in the ideal data D3 based on the rotational deviation amount and the expansion / contraction error obtained in Step S6 and the positional deviation amount obtained in Step S7. . Then, using the cutout position after correction, a plurality of depression image data D2 is cut out from the photographed image data D1 (step S8).

  The obtained plurality of depression image data D2 is stored in the storage unit 61 together with identification information indicating which depression 91 of which well plate 9 corresponds. The user of the image acquisition device 1 reads out the depression image data D2 having the same identification information from the plurality of depression image data D2 accumulated in the storage unit 61, displays the depression image data D2 on the display unit 40, and displays them. Can be compared and analyzed.

  As described above, in the image acquisition device 1 of the present embodiment, a plurality of reference point candidate positions Op are obtained in the captured image data D1, and the rotational deviation amount of the captured image data D1 is determined based on variations in the reference point candidate positions Op. Then, the expansion / contraction error and the positional deviation amount are obtained. Then, a plurality of depression image data D2 is cut out from the photographed image data D1 using the cutout position corrected based on the obtained rotation deviation amount, expansion / contraction error, and position deviation amount. Thereby, it is possible to obtain a cutout image with high accuracy in consideration of rotational deviation, expansion / contraction, and positional deviation of the captured image data.

  In particular, in the present embodiment, it is only necessary that the positions of at least two recesses 91 can be specified in the pattern matching in step S4. For this reason, even if each depression 91 in the well plate 9 is small and it is difficult to accurately specify the positions of all depressions 91 by pattern matching, the individual depression image data D2 is appropriately stored. Can be cut out.

  In the present embodiment, the point of interest Wi and the actual point of interest Wr are determined using the recess 91 in the well plate 9. For this reason, it is not necessary to form on the well plate 9 the alignment mark used as the focused point Wi and the actual focused point Wr separately from the recess 91.

<3. Modification>
Although one embodiment of the present invention has been described above, the present invention is not limited to the above embodiment.

  In the above-described embodiment, the pattern matching of step S4 is performed using the empty depression 91. However, when all the depressions 91 of the well plate 9 are used, pattern matching may be performed using the reference image data D4 that represents the depressions 91 holding the culture solution. Further, pattern matching may be performed using a plurality of types of reference image data D4.

  In step S2, all the depressions 91 in the well plate 9 are selected and points corresponding to all the depressions 91 are designated as points of interest. In step S4, the reference image data representing the empty depressions 91 is displayed. Pattern matching may be performed using D4. In this case, only the empty recess 91 in the captured image data D1 is detected with a high matching score. And the actual attention point Wr corresponding to the empty hollow part 91 is specified. In this manner, the empty depression 91 may be extracted using pattern matching. Then, in step S2, it is possible to save the user from searching for and selecting the empty depression 91 from among the plurality of depressions 91. The empty recess 91 is not necessarily arranged on the outermost periphery as shown in FIG.

  In addition, in step S5 described above, when it is unclear which actual target point Wr corresponds to which target point Wi, each actual target point Wr (1) to Wr (4) is shown in FIG. On the other hand, by applying all the relative coordinates C (1) to C (4), all possible reference point candidate positions Op may be obtained once. In this case, the remaining four reference point candidate positions Op may be left and the other reference point candidate positions Op may be deleted.

  Further, the number of the recesses 91 included in the well plate 9 may be different from the example of FIG. Further, the shape of each recess 91 may be circular as viewed from above as shown in FIG. 1, or may be another shape such as a rectangle.

  Moreover, you may combine suitably each element which appeared in said embodiment and modification in the range which does not produce inconsistency.

DESCRIPTION OF SYMBOLS 1 Image acquisition apparatus 9 Well plate 10 Plate holding part 20 Light projection part 30 Imaging part 40 Display part 50 Input part 60 Computer 61 Memory | storage part 62 Relative coordinate calculation part 63 Matching process part 64 Deviation amount calculation part 65 Image cutout part 70 Computer program 91 depression 92 culture medium D1 photographed image data D2 depression image data D3 ideal data D4 reference image data D5 table data Oi reference point Op reference point candidate position Wi attention point Wr actual attention point C relative coordinates

Claims (8)

An image cutout method for cutting out image data for each depression from the image data of a sample container having a plurality of depressions,
a) designating a reference point and at least two points of interest corresponding to at least two depressions in the ideal data of the sample container;
b) obtaining relative coordinates of the reference point with each of the points of interest as the origin;
c) identifying actual points of interest corresponding to each of the points of interest in the captured image data of the sample container;
d) Based on the plurality of actual target points and the relative coordinates corresponding to the actual target points, a reference point candidate position corresponding to the reference point is obtained, and from the reference point candidate position, the captured image data Obtaining a rotational deviation amount and a positional deviation amount;
e) correcting a preset cutout position based on the rotational deviation amount and the positional deviation amount, and using the cutout position after correction, cutting out image data for each depression from the photographed image data; ,
Image clipping method including The image clipping method according to claim 1,
Said step d)
d-1) A voting process for obtaining a set of a plurality of reference point candidate positions corresponding to the reference point based on the plurality of actual attention points and the relative coordinates corresponding to the actual attention points, A process of repeating the data several times while rotating the data by a minute angle;
d-2) obtaining a rotation angle of the photographed image data when a set of a plurality of reference point candidate positions is most concentrated among the plurality of voting processes as an amount of rotational deviation of the photographed image data;
Image clipping method including The image clipping method according to claim 2,
In the step d-1), the voting process is repeated a plurality of times while changing the rotation angle of the captured image data and the expansion / contraction rate of the captured image data,
In the step d-2), the rotation angle and the expansion / contraction rate of the photographed image data when the plurality of reference point candidate position sets are most concentrated among the plurality of voting processes are respectively set to the photographed image data. Calculated as the amount of rotational deviation and expansion / contraction error of
In the step e), an image cutout method for correcting a preset cutout position based on the rotational deviation amount, the expansion / contraction error, and the positional deviation amount. The image clipping method according to claim 2 or claim 3, wherein
Said step d)
d-3) Among the plurality of voting processes, the average coordinates of the plurality of reference point candidate positions when the set of the plurality of reference point candidate positions is most concentrated, and the coordinates of the reference points in the ideal data An image cut-out method further comprising a step of obtaining a difference between the two as a positional deviation amount of the captured image data. An image clipping method according to any one of claims 1 to 4, wherein
In the step b), an image cutout method for obtaining polar coordinates of the reference point with the point of interest as an origin as the relative coordinates. An image cutout method according to any one of claims 1 to 5,
In step a), a point corresponding to at least two empty depressions is designated as the point of interest,
In the step c), an image cut-out method for specifying the actual attention point by pattern matching using reference image data representing an empty depression. The image cutout method according to claim 6,
In step a), the points corresponding to all the depressions in the ideal data are designated as the points of interest,
In the step c), an image cut-out method for specifying an actual point of interest corresponding to an empty depression in the captured image data by pattern matching using reference image data representing an empty depression. An image acquisition device for photographing a sample container having a plurality of depressions and cutting out image data for each depression from the obtained captured image data,
A holding unit for holding the sample container horizontally;
A light projecting unit for irradiating light to the sample container;
An imaging unit that images the sample container and acquires captured image data of the sample container;
An image processing unit that cuts out image data for each depression from the captured image data;
With
The image processing unit
In the ideal data of the sample container, a relative coordinate calculation unit that obtains relative coordinates of a predetermined reference point with the origin of each of at least two points of interest corresponding to at least two depressions;
In the captured image data, an actual point of interest specifying unit for specifying an actual point of interest corresponding to each of the points of interest;
Based on the plurality of actual target points and the relative coordinates corresponding to the actual target points, a reference point candidate position corresponding to the reference point is obtained, and the rotational deviation of the captured image data is determined from the reference point candidate position. A deviation amount calculation unit for obtaining an amount and a positional deviation amount;
An image cutout unit that corrects a preset cutout position based on the rotation shift amount and the position shift amount, and cuts out image data for each depression from the captured image data using the cutout position after correction; ,
An image acquisition apparatus.

Images