JP2014067426A - Tablet packaging inspection apparatus, and tablet packaging inspection method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a tablet packaging inspection apparatus improving inspection precision of presence of foreign matter sticking to a tablet surface and chipping of the tablet.SOLUTION: A tablet packaging inspection apparatus comprises: a tablet cut unit 2 for cutting an area where one tablet is captured respectively out of a captured image; a coordinate converter 4A for preparing an image that the image capturing cut one tablet is converted into a polar coordinate making the center of the tablet a pole and making the distance from the center and the declination axes respectively; a projection data calculation unit 5 for totalizing gray level every declination from the made image; a mask processing unit 6 for masking a summary value which is less than a value determined based on the gray level of a secant on the tablet in the summary values totalized every declination; and a determination unit 7A for determining whether a value less than the threshold exists among the summary values totalized every declination except the masked summary values.

Description

  The present invention relates to a technique for inspecting whether a foreign substance is attached to a tablet surface or whether there is a defective portion caused by a part of a tablet collapsing.

  A pharmaceutical company that manufactures pharmaceuticals inspects the quality of tablets (uncoated tablets) before shipping. In this inspection process, an inspection is performed to determine whether foreign matter is adhered to the tablet surface and whether there is a defective portion (hereinafter referred to as chipping) caused by a part of the tablet collapsing.

  In addition, a part of the tablet has a secant for easy splitting in half.

  The following technologies are disclosed as related technologies.

JP 2002-99898 A JP 2005-164272 A Japanese Patent Laid-Open No. 11-73508 Japanese Patent Laid-Open No. 10-104076

  In a method for inspecting a tablet using an image, there is a case where foreign matter adhesion and chipping are erroneously detected. Moreover, since a tablet enters the human body, a highly accurate inspection is required.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a technique for inspecting the presence or absence of foreign matter adhesion or chipping of a tablet with higher accuracy.

  In order to solve the above-described problem, a tablet packaging inspection device according to an aspect of the present invention includes an imaging unit that images a plastic sheet in which one or more tablets are supplied to a recess, and an image captured by the imaging unit. A tablet cutout unit that cuts out an area where one tablet is imaged, and an image obtained by imaging one tablet cut out by the tablet cutout unit into polar coordinates centered on the tablet. A coordinate conversion unit that creates an image centered on the distance from the center and the declination, respectively, and a totaling unit that totalizes density values for each declination from the image created by the coordinate conversion unit; A mask that masks aggregate values that are less than a value determined based on the concentration value of the secant in the tablet among the aggregate values that are aggregated for each declination by the aggregation unit. Among the aggregate values aggregated for each declination by the aggregation processor and the aggregate values other than the aggregate values masked by the mask processor, determine whether there is a value below the threshold value A determination unit; and an execution control unit that controls the coordinate conversion unit, the aggregation unit, the mask processing unit, and the determination unit so that each of the images cut out by the tablet cutout unit is processed.

  In order to solve the above-described problem, a tablet packaging inspection device according to one aspect of the present invention is imaged by an imaging unit that images a plastic sheet in which one or a plurality of tablets are supplied to a recess, and the imaging unit. A tablet cutout unit that cuts out an area where one tablet is imaged from the image, a binarization unit that binarizes an image of one tablet cut out by the tablet cutout unit, and the binarization A coordinate conversion unit that creates an image converted into polar coordinates with the center of the tablet as a pole with respect to the image binarized by the unit, the distance from the center, and an image with the declination as an axis, respectively The totaling unit that totalizes the density values for each declination from the image created by the coordinate conversion unit, and the total value that is totaled for each declination by the totaling unit is below the threshold value An execution control for controlling the binarization unit, the coordinate conversion unit, the aggregation unit, and the determination unit so that each of the image cut out by the determination unit and the tablet cutout unit is processed. Part.

  In order to solve the above-described problem, a tablet packaging inspection device according to an aspect of the present invention includes an imaging unit that images a plastic sheet in which one or more tablets are supplied to a recess, and an image captured by the imaging unit. A tablet cutout unit that cuts out an area where one tablet is imaged, and an image obtained by imaging one tablet cut out by the tablet cutout unit into polar coordinates centered on the tablet. The first coordinate conversion unit that creates an image with the distance from the center and the declination as axes, and the density value for each declination from the image created by the first coordinate conversion unit The total which falls below the value determined based on the concentration value of the secant contained in the tablet among the total values calculated for each declination by the first totaling unit and the first totaling unit Among the aggregated values other than the aggregated values masked by the mask processing unit among the aggregated values aggregated for each declination by the mask processing unit and the first aggregation unit, A first determination unit that determines whether there is a binarization unit that binarizes an image obtained by imaging one tablet cut out by the tablet cutout unit, and an image binarized by the binarization unit On the other hand, a second coordinate conversion unit for generating an image converted into polar coordinates with the center of the tablet as a pole, the image having a distance from the center and a declination as axes, and the second coordinate conversion Whether there is a value less than the threshold value among the second totaling unit that totalizes the density values for each declination and the total value totaled for each declination by the second totaling unit from the image created by the unit A second determination unit for determining The first and second coordinate conversion units, the first and second tabulation units, the mask processing unit, the first and second determination units, so that each of the images cut out by the tablet cutting unit is processed. And an execution control unit that controls the binarization unit.

  In order to solve the above-described problem, in the tablet packaging inspection method according to one embodiment of the present invention, an imaging device images a plastic sheet in which one or a plurality of tablets are supplied to a recess, and a computer images using the imaging device. An image obtained by cutting out a region where one tablet is imaged from the image, and converting the image obtained by capturing one extracted tablet into polar coordinates with the center of the tablet as a pole. Then, an image is created with the distance from the center and the declination as axes, respectively, and from the created image, the density value is totaled for each declination, and the aggregated value aggregated for each declination Of these, the total value below the value determined based on the concentration value of the secant in the tablet is masked, and the total value other than the masked total value among the total values calculated for each declination In out, the process of determining whether there is a value below the threshold value, executes the each of the cut-out image.

  Inspection accuracy for foreign matter adhering to the tablet surface and the presence or absence of chipping of the tablet can be improved.

It is a figure which shows an example of a structure of the tablet packaging apparatus which concerns on Embodiment 1,2. 3 is a block diagram illustrating an example of functions of an image processing unit according to Embodiment 1. FIG. 6 is a flowchart illustrating an example of an operation of the image processing unit according to the first embodiment. 6 is a schematic diagram illustrating an example of a region to be processed (ROI1) according to Embodiments 1 and 2. FIG. It is a schematic diagram which shows an example of the one area | region (ROI) of the tablet which concerns on Embodiment 1,2. It is a figure explaining the polar coordinate conversion and projection data which concern on Embodiment 1,2. It is a figure explaining the smoothing process which concerns on Embodiment 1,2. It is a figure which shows an example of arrangement | positioning of the lighting fixture which concerns on Embodiment 1,2. 6 is a block diagram illustrating an example of functions of an image processing unit according to Embodiment 2. FIG. 10 is a flowchart illustrating an example of an operation of the image processing unit according to the second embodiment. It is a figure explaining an example of the mask process which concerns on Embodiment 2. FIG.

  Embodiments of the present invention will be described below with reference to the drawings. An apparatus and method effective for the inspection for the presence / absence of chipping will be described in the first embodiment, and an apparatus and method effective for the inspection for the presence / absence of foreign matter will be described in the second embodiment.

(Embodiment 1)
In this embodiment, a tablet packaging inspection apparatus effective for inspecting the presence or absence of chipping of a tablet (plain tablet) will be described. Moreover, in this Embodiment, it shall test | inspect using the tablet image imaged using transmitted light (NIR or VIS) illumination.

  The configuration of the tablet packaging device according to the present embodiment will be described with reference to the schematic diagram of FIG. The tablet packaging apparatus 100 includes a plastic sheet conveying unit 10, a recess creating unit 20, a tablet supplying unit 30, a PTP (Press Through Package) sheet conveying unit 40, a sheet combining unit 50, a packaging sheet cutting unit 60, a taking out unit 70, and a CCD camera. 80 (imaging unit) and an image processing unit 90. The CCD camera 80 and the image processing unit 90 constitute a tablet packaging inspection device.

  The plastic sheet transport unit 10 supplies the plastic sheet 200 and transports the plastic sheet 200 to the recess creating unit 20 which is the next step. The plastic sheet transport unit 10 is composed of a plurality of transport rollers. The recess creating unit 20 creates a recess for putting a tablet in the plastic sheet 200 transported from the plastic sheet transport unit 10 using a mold. The plastic sheet 200 with the recesses is conveyed to the board 301 via a plurality of conveying rollers.

  The tablet supply unit 30 supplies the tablets 201 one by one to the recess on the plastic sheet 200 conveyed to the board 301. The PTP sheet conveying unit 40 supplies the PTP sheet 202 and conveys it to the sheet combining unit 50. The PTP sheet conveyance unit 40 is configured by a plurality of conveyance rollers.

  The sheet coupling unit 50 couples the PTP sheet 202 and the plastic sheet 200 in which the tablets 201 are supplied to the recesses.

  The packaging sheet cutting unit 60 cuts out the sheet in which the tablets are packaged by joining the PTP sheet 202 and the plastic sheet 200 so as to have a predetermined size, that is, a predetermined number of tablets. This cut sheet is referred to as a packaging sheet. The take-out unit 70 takes out the cut packaging sheet. Here, the take-out unit 70 removes the packaging sheet determined to be defective by the tablet packaging inspection apparatus configured by the CCD camera 80 and the image processing unit 90.

  The CCD camera 80 images the tablets supplied to the recesses of the plastic sheet 200 for each predetermined area. This predetermined area is determined by the size of each sheet cut out by the packaging sheet cutout section 60.

  The image processing unit 90 is a computer having at least a CPU, a memory, and a non-volatile storage device. The image processing unit 90 detects a foreign substance adhering to the tablet or chipping by processing the image captured by the CCD camera 80.

  Next, the function of the image processing unit 90 will be described with reference to the functional block of FIG. The image processing unit 90 includes a center-of-gravity position calculation unit 1, a tablet cutout unit 2, a binarization unit 3, a coordinate conversion unit 4, a projection data calculation unit 5 (aggregation unit), and a determination unit 7. Each of these functional blocks is realized by a program stored in advance in the above-described nonvolatile storage device being loaded into the above-described memory, and the above-described CPU calculating and executing the loaded program.

  The center-of-gravity position calculation unit 1 calculates the center-of-gravity position of each tablet in the image captured by the CCD camera 80 (the center position of each tablet). The tablet cutout unit 2 creates a cutout image of each tablet based on the barycentric position calculated by the barycentric position calculating unit 1. The tablet cutout unit 2 creates cutout images of all the tablets that have been imaged.

  The binarization unit 3 binarizes the image cut out by the tablet cutout unit 2, that is, an image in which one tablet is captured. The binarization unit 3 selects the maximum density value in the image cut out by the tablet cutout unit 2, calculates a threshold value based on the selected maximum density value, and binarizes using the threshold value.

  The coordinate conversion unit 4 is an image obtained by converting the image binarized by the binarization unit 3 into polar coordinates with the tablet centroid position (tablet center) as a pole, and a distance from the centroid position, and Images are created with the declination being the vertical axis and the horizontal axis, respectively.

  The projection data calculation unit 5 calculates data (projection data) obtained by collecting the density values for each declination of the image subjected to polar coordinate conversion by the coordinate conversion unit 4.

  The determination unit 7 detects the presence / absence of chipping by determining whether there is a value that is less than the threshold value among the density values collected for each declination by the projection data calculation unit 5.

Next, the operation of the image processing unit 90 will be described with reference to the flowchart of FIG. Hereinafter, the input image is F _IN (x, y)
In addition, the entire area of the image obtained from the CCD camera 80 is ROI1.

First, the center-of-gravity position calculation unit 1 binarizes the image obtained from the CCD camera 80 with a predetermined threshold value (T _h ), and creates an image F _{BH in} which the tablet area and other areas are separated. Thus, the tablet is extracted (step S1). In the present embodiment, the obtained image F _BH is an image that is binarized and inverted with respect to the input image.
F _BH =! (Bin (T _h ) · F _IN (x, y))
Note that the notation “! (F)” means a logical negation of F. Further, the notation “A · B” means that the process of A is executed on the target B.

  Next, the barycentric position calculation unit 1 calculates the barycentric position of the tablet (S2).

Here, calculation of the center of gravity position of the tablet will be described. Centroid position calculation unit 1, by labeling processing on the image _{F BH} a (LABEL · _{F BH),} imparts a label for identifying each tablet region in an image _{F BH.} The center-of-gravity position calculation unit 1 excludes tablet regions having a predetermined area (T _SL ) or less. Thereafter, the center-of-gravity position calculation unit 1 calculates the center-of-gravity position (feature amount) of each tablet region to which a label is attached. The center-of-gravity position obtained here is defined as (x _c (i), y _c (i)) (i = 0 to (number of tablet regions−1)). The centroid position calculation unit 1 calculates the centroid position with subpixel accuracy. Therefore, the values of x _c (i) and y _c (i) are both decimal numbers.

The center-of-gravity position calculation unit 1 aligns the obtained center-of-gravity positions (x _c (i), y _c (i)) as shown in FIG. That is, the center-of-gravity position calculation unit 1 aligns the center-of-gravity positions (x _c (i), y _c (i)) so that the tablets are arranged in the ROI 1 captured by the CCD camera 80. Here, the final result of the centroid position is (x _c ^T (k, l), y _c ^T (k, l))
And As shown in FIG. 4, when m is the number of tablets in the horizontal direction (x direction) in ROI1, and n is the number of tablets in the vertical direction (y direction) in ROI1, k = 0 to m−. 1, l = 0 to n−1. The center-of-gravity position calculation unit 1 sets (x _c ^T (k, l) = − 1, y _c ^T (k, l) = − 1) for missing tablets (those in which tablets are not stored in the recesses).

  Since the brightness, contrast, and the like of the image obtained for each arrangement in FIG. 4 are different, each parameter described in the present embodiment is prepared for each arrangement unit in FIG.

Returning to the flowchart of FIG. Next, the binarizing unit 3 calculates the maximum density of each tablet in the input image F _IN (x, y) (S3). The maximum tablet concentration is calculated for each tablet.

  A method for calculating the maximum concentration of the tablet in the binarizing unit 3 will be described. Since the maximum concentration of the tablet is in the vicinity of the center of the tablet, the processing in this embodiment takes this point into consideration.

First, the binarization unit 3 performs a process for shrinking each tablet area (a process for shrinking the area of the tablet area to the approximate center position of the tablet) on the binarized image F _BH , and the image F _BH after the shrink process is performed. ^{Get E.} The binarization unit 3 performs the degeneration process a predetermined number of times (Ne).
F _BH ^E = ERODE (Ne) ・ F _BH

Next, the binarizing unit 3 creates an image obtained by _ANDing the input image F _IN (x, y) and F _BH ^E , that is, an image only near the center of the tablet, and uses this image for each tablet. Select and obtain the maximum concentration value. The acquired maximum concentration value for each tablet is aligned with the label (k, l) as in the above-described center of gravity position. Calculate the maximum concentration value as G _max (k, l)
And

Returning to the flowchart of FIG. The binarization unit 3 uses G _max (k, l) to calculate a binarization threshold (T _B (k, l)) for each tablet by the following equation (S4).
T _B (k, l) = G _max (k, l) + T _off
T _off is a preset parameter.

  Returning to the flowchart of FIG. The subsequent processing loops until the number of tablets (for (m × n) in the example of FIG. 4) is executed (S5). That is, the process is repeated until k becomes 0 to m−1 and l becomes 0 to n−1.

The tablet cutout unit 2 cuts out an area where one tablet is imaged (hereinafter referred to as ROI) from the input image F _IN (x, y) (S6).

Here, the process of the tablet cutout unit 2 will be described with reference to FIG. Based on the centroid position (x _c ^T (k, l), y _c ^T (k, l)) calculated by the centroid position calculation unit 1, the tablet cutting unit 2 uses the following formula to calculate the ROI ((x ₁ , Y ₁ ) − (x ₂ , y ₂ )).
x ₁ = [x _c ^T (k, l)] − w _h ^T
y ₁ = [y _c ^T (k, l)] − w _h ^T
x ₂ = x ₁ + 2wh _h ^T
y ₂ = y ₁ + 2wh _h ^T
Here, w _h ^T is a parameter determined in advance so as to obtain a cut-out image in which one tablet can be entirely accommodated, and the barycentric position (x _c ^T (k, l), y _c ^T (k, l)). X-axis and y-axis component distance from Moreover, the notation [F] means an integer value of the floating-point F.

Returning to the flowchart of FIG. The binarization unit 3 binarizes the image cut out in step S6 with the binarization threshold value T _B (k, l) calculated in step S4, so that the binarized image (F _BIN ^{T (k , L)} ) is created (S7).
F _BIN ^{T (k, l)} = Bin (T _B (k, l)) · F _IN (ROI)

Next, the coordinate conversion unit 4 converts the binarized image F _BIN ^{T (k, l)} to the center ((x _c ^T (k, l) −x ₁ ), (y _c ^T (k, l) −y). ₁ )) The image (F _POL ^{T (k, l)} ) that has been _subjected to polar coordinate conversion in step ¹⁾ ) is created (S8). Note that ((x _c ^T (k, l) −x ₁ ), (y _c ^T (k, l) −y ₁ )) is a coordinate system based on the upper left corner of the input image F _IN (x, y). It is a notation. The coordinate conversion unit 4 performs Bilinear interpolation when performing polar coordinate conversion.
F _POL ^{T (k, l)} = POLAR · F _BIN ^{T (k, l)}
Here, the polar coordinate conversion image F _POL ^{T (k, l) has} a width of (2 × π × Re) and a height of (Re−Rs + 1). Rs is the starting point (lower limit) of the vertical axis r of polar coordinates, Re is the end point (upper limit) of the vertical axis r, and these are values defined in advance as parameters.

Next, the projection data calculation unit 5 creates projection data in which the density values are tabulated for each declination of the polar coordinate conversion image F _POL ^{T (k, l)} (S9).

  Here, step S8 and step S9 will be described with reference to FIG. Although binarization is not considered in the diagram shown in FIG. 6, FIGS. 6A and 6B are actually binarized images.

When the binarized image F _BIN ^{T (k, l)} is, for example, FIG. _6A , the coordinate conversion unit 4 converts the binarized image F _BIN ^{T (k, l)} into the center of gravity (center) of the tablet. _Is an image F _POL ^{T (k, l)} converted into a coordinate system with the distance r from the center of gravity as the vertical axis (y direction) and the deflection angle θ as the horizontal axis (x direction) (FIG. 6 ⁾ . (See (B)).

Projection data calculating unit 5, respectively counts the density value of the polar coordinate conversion image ^{_{F POL T (k, l)}} , for each deflection angle theta (horizontal axis component of the polar coordinate conversion image ^{_{F POL T (k, l)}} ) ( addition) Then, projection data is created with the vertical axis representing the total value and the horizontal axis representing the value of the horizontal axis of the polar coordinate conversion image F _POL ^{T (k, l)} (see FIG. 6C). Here, on the data created by the projection data calculation unit 5, the total value of the chipped portion (and the dividing line portion) is characterized in comparison with other values (in the example of FIG. 6C, the total value). Note that this is low). The projection data calculation unit 5 stores the total values in P _j ^{T (k, l)} (x) (x is a horizontal axis component in FIG. 6C), respectively.

Returning to the flowchart of FIG. The projection data calculation unit 5 performs a smoothing process on the projection data obtained as described above (S10). As shown in FIG. 7A, the projection data obtained as described above has a minute noise or a special point of only a minute section. Therefore, the projection data calculation unit 5 smoothes the noise component and generates a noise component. A smoothing process is performed to reduce this. The smoothing process is performed using the following equation.
Pj ^{mT (k, l)} (x) = (ΣPj ^{T (k, l)} (x + p)) / (2 · Sm + 1)
Note that the range of the sum Σ of the sequence is from p = x−Sm to p = x + Sm.
Sm is a predefined parameter, and (2 · Sm + 1) is the smoothing width in the x direction of Pj ^{mT (k, l)} (x). When the smoothing width increases, it acts as a low-pass filter in which fine fluctuations are attenuated by smoothing.

Next, the determination unit 7 determines chipping (S11). Here, the process by the determination part 7 is demonstrated. The determination unit 7 calculates an average value using the following formula.
P _AVE = (ΣPj ^{T (k, l)} (x)) / w
The range of the sum Σ of the sequence is from x = 0 to x = w−1, and w is the width of F _POL ^{T (k, l)} .

Thereafter, the determination unit 7 performs a process of removing the average value calculated using the following expression from P _j ^{T (k, l)} (x) and Pj ^{mT (k, l)} (x).
Pj ^{T (k, l)} F (x) = Pj ^{T (k, l)} (x) −P _AVE
Pj ^{mT (k, l)} F (x) = Pj ^{mT (k, l)} (x) −P _AVE

The determination unit 7 has predetermined values (T _chip , T _chip ^m ) in which Pj ^{T (k, l)} F (x) and Pj ^{mT (k, l)} F (x) calculated as described above are defined in advance. The presence / absence of chipping is determined by determining whether it is below the threshold. That is, when at least one of the following conditions is satisfied, the determination unit 7 determines that the tablet has chipping.
Pj ^{T (k, l)} F (x) <T _chip
Pj ^{mT (k, l)} F (x) <T _chip ^m
Note that x is 0 to w-1 (w is the width of F _POL ^{T (k, l)} ).

In the processing of the determination unit 7 described above, P _AVE is dynamically calculated. However, P _AVE may be previously added to T _chip and T _chip ^m as a fixed value.

  The image processing unit 90 is provided with an execution control unit realized by cooperation of the CPU, the memory, and the program. The execution control unit performs the determination in step S5 and the loop control from step S6 to step S11. It may be done.

  Next, the arrangement of lighting fixtures when taking an image with the CCD camera 80 will be described with reference to FIG. In the present embodiment, not only the lighting fixture 400 that applies transmitted light obliquely from above to the tablet stored in the recess, but also a lighting fixture 400A that applies transmitted light from diagonally downward. Since chipping is present in the vicinity of the edge of the tablet, the contrast of the defect can be emphasized by applying light transmitted obliquely from below to the edge. Moreover, it is divided into a bright defect and a dark defect depending on the chipping shape.

  In addition, by arranging the luminaire so that the transmitted light is applied from obliquely below, the intensity of illumination from above can be suppressed, and the spotted pattern on the tablet surface can be suppressed. In addition, the feature value of the secant becomes easy to appear. In this embodiment, the board 301 is made of a transparent member so that the transmitted light can be applied from an obliquely downward direction.

(Embodiment 2)
In the first embodiment, an apparatus and method effective for inspecting the presence or absence of chipping have been described. In the second embodiment, an apparatus and method effective in inspecting adhesion of foreign matter will be described.

  FIG. 9 shows a functional block diagram of the image processing unit according to the present embodiment. The image processing unit 90A includes a center-of-gravity position calculation unit 1, a tablet cutout unit 2, a coordinate conversion unit 4A, a projection data calculation unit 5 (aggregation unit), a mask processing unit 6, and a determination unit 7A.

  The coordinate conversion unit 4A is an image cut out by the tablet cutout unit 2, that is, an image obtained by taking an image of one tablet, and converted into polar coordinates with the tablet barycentric position as a pole, and the distance from the barycentric position, An image is created with the vertical axis and the horizontal axis as the vertical axis and the horizontal axis, respectively.

  The mask processing unit 6 determines a value (hereinafter referred to as a mask threshold value) determined based on the concentration value of the secant line contained in the tablet among the total values of the concentration values calculated for each declination created by the projection data calculation unit 5. A value lower than that) is masked. As the mask process, for example, a process of changing the total value that is below the mask threshold value to a value that cannot occur can be considered.

  The determination unit 7A determines whether there is a value lower than the threshold value among the total values for each declination by the projection data calculation unit 5 and the values other than the total value masked by the mask processing unit 6. It is determined whether or not foreign substances are attached to the tablet.

  Since the functional blocks other than the coordinate conversion unit 4A, the mask processing unit 6, and the determination unit 7A, the configuration of the tablet packaging device, and the arrangement of the lighting fixtures are the same as those in the first embodiment, description thereof is omitted here.

  Next, the operation of the image processing unit 90A according to the present embodiment will be described with reference to FIG. In FIG. 10, the same reference numerals as those in FIG. 3 are the same as those described in the first embodiment, and the description thereof is omitted here.

The coordinate conversion unit 4A obtains the center (x _c ^T (k, l) −x ₁ , y _c ^T (k, k) from the image cut out by the tablet cutout unit 2 in step S6 (corresponding to the ROI in the first embodiment). 1) Create an image (F _POL ^{T (k, l)} ) that has been subjected to polar coordinate conversion by -y ₁ ) (S8A). Note that ((x _c ^T (k, l) −x ₁ ), (y _c ^T (k, l) −y ₁ )) is a coordinate system based on the upper left corner of the input image F _IN (x, y). It is a notation. The processing of the coordinate conversion unit 4A is the same as that of the coordinate conversion unit 4 of the first embodiment except that the input image becomes an image cut out by the tablet cutout unit 2, and thus detailed description thereof is omitted.

  The mask processing unit 6 performs mask processing on the projection data created by the projection data calculation unit 5 at a location that is lower than the mask threshold (S101). In the case of the secant, the mask threshold value is a value defined by the total value obtained by examining in advance how much the total value will be.

  For example, referring to FIG. 11, there are two places where the total value is particularly low (see FIG. 11A). In order to mask these two locations, the mask processing unit 6 determines whether the value is below the mask threshold, and performs mask processing on the locations below (see FIG. 11B). Note that the inspection accuracy is further improved by slightly increasing the masking width (x direction) so as to cover only all the dividing lines. The value previously investigated is used for this fine amplification.

  The determination unit 7A determines whether foreign matter is attached (S11A). The determination unit 7A does not set a portion masked by the mask processing unit 6, that is, a total value masked by the mask processing unit 6, as a determination target. Except for this point, the second embodiment is the same as the first embodiment, and a detailed description thereof will be omitted.

  The image processing unit 90A is provided with an execution control unit realized by the cooperation of the CPU, the memory, and the program. The execution control unit performs the determination in step S5 and the loop control from step S6 to step S11A. It may be done.

  Although the first embodiment is an effective means for inspection of chipping and the second embodiment is an effective means for inspection of foreign matter adhesion, the determination of foreign matter adhesion may be performed in the form of the first embodiment. The chipping determination may be performed in the mode of the second embodiment. In addition, check whether the tablet is printed, check the tablet color (inspect whether a different color tablet is included, check whether different types of tablets are mixed), the tablet enters the recess The first embodiment and the second embodiment can be applied to the inspection relating to tablets other than the inspection of chipping and the inspection of adhesion of foreign matter, such as inspection of whether or not there is.

  The CCD camera 80, the image processing unit 90 described in the first embodiment, and the state image processing unit 90A described in the second embodiment can be used as one tablet packaging inspection apparatus. In this case, the tablet packaging inspection apparatus may perform the tablet inspection in the order of the foreign matter inspection described in the second embodiment after the chipping inspection described in the first embodiment, or vice versa. Moreover, you may mix the step demonstrated in Embodiment 1 and Embodiment 2 as needed.

  The present invention can be implemented in various other forms without departing from the spirit or main features thereof. Therefore, the above-described embodiment is merely an example in all respects and should not be interpreted in a limited manner. The scope of the present invention is shown by the scope of claims, and is not restricted by the text of the specification. Moreover, all modifications, various improvements, substitutions and modifications belonging to the equivalent scope of the claims are all within the scope of the present invention.

  DESCRIPTION OF SYMBOLS 1 Center-of-gravity position calculation part, 2 Tablet cutting part, 3 Binary part, 4,4A Coordinate conversion part, 5 Projection data calculation part, 6 Mask processing part, 7, 7A Judgment part, 10 Plastic sheet conveyance part, 20 Recess preparation Part, 30 tablet supply part, 40 PTP sheet conveying part, 50 sheet combining part, 60 packaging sheet cutting part, 70 taking part, 80 CCD camera, 90, 90A image processing part, 100 tablet packaging device, 301 board, 400, 400A lighting equipment.

Claims (4)

An imaging unit for imaging a plastic sheet in which one or more tablets are supplied to the recess;
From the image captured by the imaging unit, a tablet cutout unit that cuts out each region where one tablet is imaged,
A binarization unit that binarizes an image obtained by imaging one tablet cut out by the tablet cutout unit;
An image obtained by converting the image binarized by the binarization unit into polar coordinates with the center of the tablet as a pole, and creating an image with the distance from the center and the declination as axes. A coordinate transformation unit;
From the image created by the coordinate conversion unit, a totaling unit for totalizing density values for each declination,
A determination unit that determines whether there is a value that falls below a threshold value among the total values that are totaled for each declination by the total unit;
An execution control unit that controls the binarization unit, the coordinate conversion unit, the counting unit, and the determination unit so that each of the images cut out by the tablet cutting unit is processed;
A tablet packaging inspection apparatus. In the tablet packaging inspection apparatus according to claim 1,
The binarization unit selects a maximum density value on an image of the tablet and calculates a threshold value for binarization based on the maximum density value. The imaging device
Image a plastic sheet with one or more tablets in the recess,
Computer
Obtain images captured by the imaging device, and cut out each region where one tablet is imaged from the image,
Binarize the image of one clipped tablet,
For the binarized image, an image converted into polar coordinates with the center of the tablet as a pole, creating an image with the distance from the center and the declination as axes,
From the created image, the density values are totaled for each declination,
The tablet packaging inspection method which performs the process which determines whether there exists a value which is less than a threshold value among the total values totaled for every declination to each of the said cut-out image. In the tablet packaging inspection method according to claim 3,
In the binarization, a maximum density value on the image of the tablet is selected, and a threshold value for binarization is calculated based on the maximum density value.

Images