JPH08210962A - Image analyzing method and granulation control method in granulating device - Google Patents

Abstract

PURPOSE: To precisely measure a particle characteristic and precisely form a required granule by an efficient granulating operation by binarizing the image of a particle in granulation, and separating the binarized particle image to analyze it. CONSTITUTION: The photographed crude image of a particle in granulation is binarized by smoothing and variable-density treatment, and the mutually adjacent or superposed particles are separated by separating means such as contraction, circulation and wedge. For example, in the contraction separation effective for individually separating a plurality of mutually adjacent particles, or removing a fine powder adhered to the surface of the particle, the binarized image is contracted at a fixed contraction coefficient to separate the contact part, and when the fine powder is removed, the image is contracted until a powder with an extremely small diameter can not be observed. At this time, in the relatively large particle image, the image is left while only the diameter is contracted, and only the powder is removed. A measurement is performed for the separated image, and the measured value is expanded at the same ratio as the contracting coefficient to determine the characteristic of each particle.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention captures particles during granulation or coating in a granulator, etc., and analyzes the captured images to obtain particle characteristics such as particle shape and particle size. The present invention relates to a method and a control method for controlling granulation and coating based on the obtained particle characteristics.

[0002]

2. Description of the Related Art A conventional granulating apparatus includes a fluidized bed granulating apparatus,
Various granulators such as a rolling granulator are known. In these granulators, the granulators mainly used differ depending on the purpose of granulation. For example, in a fluidized bed granulator, the shape is often irregular and the particle size is relatively large, and in the rolling granulator, the shape is relatively spherical. However, it is known that even if the granulation method is constant, the shape, particle size, etc. are different or change within a certain range due to the difference in the granulation conditions.

On the other hand, the desired particle shape, the desired particle size, etc. are different depending on the purpose of use of the granulated product.
A granulator is used. However, in order to obtain a granulated product that meets the purpose as much as possible, even if the granulation is performed by a desirable granulation method, setting the granulation conditions becomes a problem.
Moreover, it is difficult to directly confirm whether or not the desired granulation product is obtained.

For the above-mentioned reasons, there has been a demand for a granulating apparatus capable of observing the shape and particle size of particles during granulation.
However, in the granulating tank during granulation, the shape and particle size of individual particles during granulation cannot be directly observed by any of the granulating apparatuses.

Therefore, there is a photographing device adapted to photograph the shape of the granular material during granulation or coating. This conventional example is described in JP-A-54-92389 and JP-A-59-
81535, JP-A-63-266339, JP-A-6
Those described in 0-15541, JP-A-57-59143, JP-A-58-73730, JP-A-4-265142 and the like are known.

Among these conventional examples, JP-A-54-923
89, JP-A-59-81535, JP-A-63-26.
6339 and JP-A-4-265142 both drop a granular material sent by a belt conveyor or the like, and photograph from the side opposite to the illumination direction or at a right angle. Since these particles cannot be separated completely because they fall naturally, the particles cannot be accurately determined because they overlap each other. Therefore, it must be separated in advance when it is transferred by a belt conveyor. Furthermore, the present invention is not applicable to such an apparatus, instead of photographing the granulated material during granulation as in the present invention.

Further, in Japanese Patent Laid-Open No. 57-59143, the shape of a granular material sent by a belt conveyor is detected by a linear sensor camera, and the particle size is measured by frequency analysis, and no photograph is taken. . Further, this conventional example is not for photographing an image of a moving granular material such as during granulation.

In Japanese Patent Laid-Open No. 58-73730, a stroboscopic image is taken during mixed granulation, and similarly, the image is not taken in a dispersed state.

Further, Japanese Patent Application Laid-Open No. 4-265142 relates to photographing of particles during granulation, and is one in which a part of the granular material is taken out from the granular material take-out tube and attached to an adhesive tape. is there.

Further, as a photographing device for photographing particles in granulation or coating, there is a photographing device developed by the present applicant. The structure in the vicinity of the tip of this photographing device is
As shown in.

As shown in this figure, the front end portion 41 of the photographing device is shown.
Is a lens barrel 4 of a photographing system in which a photographing optical system and others are built-in.
A slight gap 44 is formed between the lens barrel 42 and the cover 43. 45
At the tip of the illumination system, for example, an optical fiber bundle 46 whose other end is located at a high-speed strobe light emission source is arranged, and a slit 47 oriented in the lateral direction is formed at the tip. The cover 48 at the front end of the illumination system is located in front of the cover 43 of the optical system, so that the slit 47 is covered by the cover 4.
3 is located slightly forward of the tip surface 43a. In this cover 48, the optical fiber bundle 46 is arranged as described above, and the cover 48 and the optical fiber bundle 46 are arranged.
A slight space 49 is formed between and.

Further, air supply ports 51 and 52 are formed at the root of the tip of the photographing device, and air for air purge is sent to these 51 and 52. Each tube is connected.

In the image pickup apparatus having such a structure, when air is supplied through the air supply tube, air gaps 4 around the lens barrel 42 of the image pickup system from the air supply ports 51 and 52, respectively.
4 and the air gap 49 around the optical fiber bundle 46 of the illumination system
One is jetted in the axial direction of the lens barrel 42 and the other is jetted in a direction perpendicular to the axial line.

Further, light emitted from a stroboscopic light source (not shown) passes through the optical fiber bundle 46 and exits from the slit 47, and at the same time, particles on the object plane of the optical system are photographed. Here, by the above-mentioned air purging with air,
It is damp and is filled with powder or fine particles with a small diameter.It is exposed in the tank, and the powder adhering to the tip of the camera is blown away. It will not be possible. In this way, no powder or granular material is attached to the tip of the optical system or near the slit.

At the same time, an air flow is generated by the air in front of the tip, and the particles in the vicinity of the air flow. This flow consists of a forward flow in the direction along the axis and a flow out of the slit 47 in a direction perpendicular to the axis, and is a combined flow of the two flows.

In this case, most of the close-packed particles at the tip of the photographing apparatus are separated from the tip by the air flow in the axial direction from the photographing system, and some of the particles remaining near the tip are: It is dispersed by the air flow through the slits of the illumination system and is in a state of being separated into individual parts.

As described above, according to the above-described conventional photographing apparatus, the particles in the vicinity of the tip end thereof are sufficiently separated so that they can be individually photographed, and photographing is performed in this state.
It is possible to obtain an image in a state of being individually separated. Further, since it is illuminated by the illumination light from the lateral direction from the slit, it is possible to capture a brightly illuminated particle image in a dark field of view. In addition, it is possible to carry out the measurement while keeping a certain distance.

Further, as a modified example of the photographing apparatus shown in FIG. 16, one shown in FIGS. 17 to 21 can be considered. In these figures, 61 is the tip of the photographing device, 61a is the tip opening, 62 is the lens barrel that also serves as the air supply tube, and 62a and 62b.
Is an opening provided in the lens barrel, 63 is an optical system (imaging system) arranged in the lens barrel, 64 is a camera head, 65 is an optical fiber for illumination connected to a light source such as a stroboscopic light emitting device, and 65a is Fiber optic tip, 66, 66a
Is an air supply path, 66b is an air supply port, and 67 is a cover glass.

[0019]

DISCLOSURE OF THE INVENTION As described above, the values required to measure the particles necessary in the granulation operation include the particle size, the number of particles, the average particle size, the particle size distribution, the σg value, the shape factor, the surface condition, etc. There are particle characteristics represented by By accurately measuring these, it is possible to efficiently perform the granulation operation and to accurately produce a desired granulated product. In particular, if these particle characteristics can be accurately obtained during granulation, it is more desirable because desired granules can be obtained with good yield by changing the granulation conditions based on the obtained values. .

However, conventionally, after the granulation, a part of the granulated product was taken out, and a required value among the respective values showing the above-mentioned particle characteristics was obtained.

By equipping the granulating tank with the above-mentioned photographing device developed by the applicant, it becomes possible to obtain a particle image during granulation, and by performing image processing based on this, the particle characteristics showing It becomes possible to measure the value of, and it becomes possible to perform granulation operation under optimum conditions.

However, even if the above-mentioned image pickup device is used, it is not sufficient to completely prevent a part of particles from being photographed by overlapping. As a result, although it is possible to perform a good granulation operation using the above-mentioned photographing device, it cannot be said that a very good effect is obtained.

An object of the present invention is to perform an image processing by a computer on the basis of a particle image in a granulating layer in operation photographed by using a photographing device to prepare an image in a state in which all particles are separated. Then, it is to provide an image analysis method in which various values are obtained and analyzed based on this.

A further object of the present invention is to carry out an analysis based on an image of the above-mentioned separated particles, and based on the analysis, granulation conditions (the liquid addition rate during granulation) , The amount of air, etc.).

[0025]

According to the image analysis method of the present invention, a raw image obtained by photographing a particle being granulated by a photographing device is binarized, and the binarized image is sequentially contracted at a constant ratio. Then, the separated image is formed, and the particle characteristics such as the number of particles and the particle shape are measured by this image.

Another method of the present invention binarizes a raw image obtained by photographing particles being granulated by a photographing device, applies an inscribed circle to the binarized particle image, and based on the characteristics. This is to separate each overlapping particle, and to measure the diameter of each circle corresponding to each separated particle to measure the particle characteristics such as particle diameter.

Still another method of the present invention is to binarize a raw image obtained by photographing a particle being granulated by a photographing device, and based on this binarized image, a mass of particles in the image, or a superposed particle. The angle of the concave portion (wedge-shaped portion) of the group is obtained, and the divided image is formed at the portion where the angle is within the predetermined range, and the particle characteristics such as the number and shape of particles are obtained from this image. It is a thing.

According to the method of the present invention, a plurality of images (screens) at different photographing times are stored in one screen and the plurality of images are processed as one screen, thereby significantly increasing the speed of image measurement and analysis. Is possible.

The method of the present invention is used for the analysis of particle characteristics such as particle separation at each point in time during the process of growing from powder to granules in the granulation step and the particle size distribution based on the separated particle image. It is possible to change the settings of the parameters used at this time, which enables optimal analysis according to the growth state of the grain during the growth process.

As described above, when the method of the present invention is used, the growth of particles from powder to granules is changed by changing the set value of each parameter when carrying out the method of the present invention at each time of granulation. It is possible to capture an image that is in the particle state at each point in the process and separate an appropriate particle image that matches the captured particle image. Therefore, based on the separated desired particle image of the particles during granulation, the particle size at each time of granulation,
It is possible to accurately grasp the particle size distribution, shape, etc. as numerical values. As a result, it is possible to accurately control the water addition rate, the amount of air, etc. based on these particle sizes and the like.

That is, the present invention is also a control method for controlling the granulation conditions during granulation based on the above method.

The control method of the present invention is a granulating apparatus equipped with a moisture meter so as to control the water content of particles, and based on the above-mentioned method, the particle diameter, particle size distribution and shape factor during granulation. Granulate under appropriate granulation conditions while measuring etc., and after the particle size reaches a desired value, by controlling the water content to keep the desired particle size and continuing granulation,
A spherical granulated product having a desired particle size, a large density, and an efficient yield can be obtained.

Further, according to the present invention, in various granulations such as fluidized bed granulation, stirring granulation, tumbling granulation and coating,
Very good control can be performed based on the particle characteristics obtained by image analysis.

Next, a more detailed description of the computer-aided image analysis method and granulation control method of the present invention will be given.

First, a means for separating a captured raw image, which is the basis of image measurement, will be described. As the separating means, there are mainly three separating means such as contraction separation, circular separation, and wedge separation. These separating means will be described in detail.

The contraction / separation means obtains a binary image by binarizing the photographed raw image by smoothing and shading processing.

The separating means separates the particles which are in contact with each other or overlap each other among the images subjected to these processes.

Shrinkage separation is (C) → (D) when two particles or two or more particles in contact with each other are individually separated as shown in FIG. 1 (A) → (B). As shown in the figure, it is an effective means for removing powder and the like when minute powder and the like adhere to the surface of the particles.

This separating means contracts the binarized image with a certain degree of contraction, and for example, as shown in FIG. 1A, two particles on the left side are in contact with each other and photographed. The image is a contracted image as shown on the right, which separates the contact areas of A. Similarly, when three particles are in contact with each other as shown in FIG. 1B, they can be individually separated. Further, when there is an adhered substance as shown in FIG. 1C, the image on the left side is contracted to form the image on the right side. In this case, the particles themselves, which are photographed in a relatively large size, remain as shown on the right side only when their diameters are reduced, but powders and the like having extremely small diameters as compared with these particles are removed and not observed. This allows unwanted images to be erased and removed. Measurement is performed based on the separated image, and the above-mentioned particle characteristics are obtained based on the obtained value expanded by the same degree as the degree of contraction.

In this contraction separation, the degree of contraction for obtaining a preferable separated image differs depending on the difference between the captured raw images.
Therefore, the contraction / separation means uses the degree of contraction as a variable parameter. The above-mentioned characteristics of each particle are calculated by a computer based on the image in which each particle is separated in this way. However, the value obtained here is the value in the contracted state, and the actual value is calculated based on the value expanded by the degree of contraction.

Since this contraction / separation means need only contract the image, it is possible to perform the processing at a high speed with extremely simple processing. However, it is not preferable if the particles are largely overlapped with each other, because the particles cannot be separated no matter how the particles shrink (no matter how large the shrinkage is).

In the circular separation, the raw image is first binarized in the same manner as in the contraction separation to create a binary image. A circle corresponding to each particle is created as shown in FIG. 2 based on this binary image. That is, the center of gravity and the area of the particle are obtained based on the mass of particles. A circle centered on the obtained center of gravity and having the same area as the obtained area is formed. Thus, as shown in FIG. 2B, a circle centered on the center of gravity of each particle is formed for each particle. This circle is superimposed on the corresponding position of the raw image, and the raw image is separated by this circle to form a separated image for each particle. There is also a circular separation method such as drawing an inscribed circle and separating the raw image based on this circle.

The circular separating means determines whether proper separation is possible or not depending on the size of the circle used for the separation, and the preferable size of the circle varies depending on the size of the particles of the photographed image. Also, as shown in the figure, whether or not correct separation is possible depends on the degree of overlap between adjacent circles. Therefore, the minimum circle and the separation rate are required as variable parameters in this circular separation means.

This separation means requires a slightly complicated process and is inferior to other means in terms of processing speed. However, it is extremely effective when separating based on a raw image with a large degree of overlapping of particles. In particular, it is desirable to apply each particle having a relatively spherical shape such as a coating.

Further, this circular separating means is not suitable for grasping the actual shape of each particle, because the separated images are all observed as circular particles, but the average particle diameter during granulation, etc. Is extremely effective for obtaining Therefore, it becomes possible to measure the average particle diameter with high accuracy.

Next, the wedge separation will be described.

In FIG. 3, (A) is an image binarized based on the raw image, and the contour line image shown in (B) is formed based on this image. From this contour image of (B), the state of the unevenness at each point on this image is obtained in terms of angles. Among the angles representing the obtained concavo-convex state, for example, the angle is −5.
A portion of 0 ° or less, that is, a concave portion (sign minus) and an angle of 50 ° or less is extracted. This point corresponds to the point where the particles partially overlap each other. Points that meet this condition join each other (partially overlapped)
Corresponds to the deep part of the recess that occurs at the point. The many wedges extracted in this way are shown in FIG. 3 (D).
Is. Although it appears as a small dot in this figure, FIG. 3 (G) shows it in an enlarged manner. When the extracted wedge is expanded, an image in which the wedges are fused is obtained as shown in FIG. These (D) of FIG.
FIG. 4 is an enlarged view of (E). That is, FIG.
The wedges a ₁ and a ₂ of FIG. 4A are the b ₁ and b ₂ of FIG.
The wedges a ₃ and a _{4 in} (A) are b ₃ and b _{4 in} FIG.
・ It becomes like this. These expanded images b ₁ , b ₂ , b ₃ , b ₄ ...
.. is contracted, two wedges a ₁ , a ₂ , a ₃ , a ₄ ... Which face each other as shown in FIG. 3 (H) are connected by a line c ₁₂ , c ₃₄ ,. You will get an image like. By superimposing the linear images c ₁₂ , c ₃₄ , ... And the contour image shown in (B), the image on the left side of (F) is obtained. This is further overlaid with the image shown in (A), which is the original binarized image,
When the image on the left side of (F) is erased, a separated image in which each particle is completely separated is obtained as in the right side.

In the wedge separating means described above, a preferable separated image can be obtained by selecting an appropriate value for the separation angle, the contour line length, and the degree of expansion and contraction. Therefore, in the wedge separation means, the separation angle, contour line length, expansion and contraction are important parameters.

Although this wedge separation means has a slower processing speed than the contraction separation means, it is possible to sufficiently process an image having a large degree of overlap between particles. or,
Since each separated particle image is separated in a state very close to the shape of the particle itself, unlike the circular separation, it is possible to accurately grasp the shape of the particle during operation.

As described above, when the particles in the process such as granulation are simply photographed, particle images separated from each other cannot be obtained. It was impossible to obtain it as data.

According to the present invention, by converting the raw image photographed by each of the above-mentioned separating means into an image in which each particle is separated, it is possible to obtain each value indicating the above-mentioned particle characteristics as digitized data. enable.

A further important feature of the method of the invention is that the particles are separated using one or more of the above separating means,
By applying the data obtained by image analysis based on the separated particle image to granulation and other steps, the most efficient and accurate control is possible.

Specifically, for example, when the optimum requirements for the operation are determined in the raw material to be granulated and various granulation conditions, accurate data are obtained by applying the above-mentioned particle separation means, and this data is used. The operation is controlled by controlling the amount of liquid added, the amount of air, etc. based on the calculated particle size, particle size distribution, etc., and it is possible to obtain an extremely efficient and desired granulated product.

In order to realize the above control, an appropriate one of the three separating means is used, and appropriate parameter setting and setting are performed at each stage according to the progress of the operation, that is, according to the growth of particles. Appropriate operation control becomes possible by making changes.

Hereinafter, general operation control according to each separating means will be described.

First, as parameters common to all of them, there are a minimum maximum diameter, a photographing magnification and the like. For example, in a granulation process using powder as a raw material, at the start of granulation, all the raw materials are so-called fine particles, and as the granulation progresses, the particles grow and become large particles. Therefore, at the beginning of the granulation, it is impossible to photograph the powder (fine particles) individually separated unless the photographing magnification is increased.
Further, when granulation progresses and most of the particles have a large diameter, when photographed at a high photographing magnification, only a small number of particles are enlarged and observed, which is not preferable for measuring the number of particles.
Therefore, in consideration of the growth rate of particles based on experience, it is necessary to increase the photographing magnification at the start of granulation and sequentially decrease the magnification at predetermined time intervals.

Next, in the granulation process, the growing particles have variations such as rapidly growing particles and conversely hardly growing particles. Therefore, in the data created by performing image processing based on all the particles, there are particles with extremely small particle diameters and particles with extremely large particle diameters compared with many other particles, which may cause an error. is there. Therefore, it is important to set the minimum and maximum particle size. As a result, it is necessary to remove particles having a minimum particle size or less and particles having a maximum particle size or more, which cause an error. In this case, the average particle diameter increases as the operation progresses, so it is necessary to change the setting to an appropriate value according to the operation progress.

As described above, according to the progress of granulation, images are taken at an appropriate magnification, and only particles required as data are left in the image, so that the coefficient required for operation can be accurately calculated and obtained. It will be possible.

Next, each separation means is applied to separate the raw image, and the above-mentioned data is obtained based on the separated particle image,
The average particle diameter, circularity, etc. are measured from the values, and control can be performed so that normal granulation operation is performed based on these values.

In this case, for example, when the number of particles is simply obtained, contraction separation is effective, when the value related to the particle diameter is obtained, circular separation is preferable, and when the shape factor is obtained, wedge separation is preferable. . Therefore, it is preferable to appropriately select and control these.

In the above control, in addition to the above-mentioned maximum and minimum particle diameters, the degree of shrinkage in the case of contraction separation, the maximum and minimum particle diameters and the separation ratio in the case of circular separation, the contour line length in the case of wedge separation, Appropriate control is possible by appropriately changing the parameters of separation angle expansion and contraction during granulation.

Further, the present invention is a granulation control method which is a combination of the above control method and water content control. That is, when the granulation control is performed to control the water content so that the water content becomes constant when the particles grow by a normal granulation method from the start of granulation and grow to a certain particle diameter, the image analysis method described above. Is applied to measure the particle size and its change accurately, measure the amount of water when the particle size reaches a desired value, and then control the water content to keep the water velocity constant at that time. By doing so, granulation is performed so that a granulated product having a constant particle size and a high density can be obtained with good yield.

Further, in the control method in which the image analysis and the water content control are combined, after the desired particle size is reached, the addition amount and the addition amount are adjusted so that the particle size does not change and the particle size does not change. The particle size of the granulated product can be made more constant by controlling the air flow rate.

Further, in the granulation operation control method combining these image analysis and water content control, the particle size distribution or shape coefficient is measured by image analysis, and when the measured value reaches a desired value or the value becomes stable. If the end point of the granulation is set to such a time, it becomes possible to obtain a granulated product having a constant particle size and a high density with a higher yield.

[0065]

EXAMPLES Next, examples of an image analysis method and a granulation control method based on particle images separated based on photographed particle images of the present invention will be described.

For example, a granulating apparatus in which the above-mentioned photographing apparatus developed by the present applicant is installed in the fluidized bed granulating apparatus is used.

FIG. 6 shows a granulating apparatus equipped with a moisture meter and a photographing device, which is capable of moisture control and image analysis. In the figure, 1 is a granulation tank, 2 is a slit plate rotatable by an appropriate driving means, 3 is a stirring blade rotated by an appropriate driving means, 4 is a blowing port, and 5 is a spray gun for spraying a binder or a coating liquid. , 6 is a product discharge port, 7 is a moisture meter, and 10 is a photographing device.

An example of image analysis using the fluidized bed granulator shown in this figure is shown.

First, the raw material powder is put into the granulating tank 1 in the same manner as in ordinary granulation. Next, the binder is sprayed from the spray gun, and warm air is blown from the blower port 4 to start granulation. On the other hand, shooting by the shooting device is started, and shooting is performed continuously, for example, every 3 seconds. At the start of granulation, since the raw material is fine particles, photographing is performed at a relatively high magnification, and as the particles grow, the photographing is switched to a lower magnification. Here, the photographing magnification may be changed continuously or intermittently by changing the magnification of the optical system.

As described above, photographing is continuously performed from the start of granulation to the end of granulation. Of the images thus photographed, a certain time interval is set, image information is photographed at each interval, and image information is output, and image analysis is performed based on the method of the present invention.

In this embodiment, smoothing can be selected for the input image. Next, the minimum number of pixels and the maximum number of pixels for binarization are set from the output image information. Thus, for example, if even one pixel does not reach the threshold value for binarization within the minimum number of strokes, the area in the minimum number of strokes is removed. That is, the area is removed. This removes dust and particles significantly smaller than the average particle size to obtain a binary image. Similarly, the area is removed even when the number of strokes exceeds the threshold over the maximum number of strokes.
This also removes particles that are significantly larger than the average particle size.

As described above, binarization is performed within the range of the area removal parameter value by setting the maximum pixel number and the minimum pixel number. The setting range of the maximum and minimum values of this area removal is 1 pixel to 262144 pixels in this embodiment.

Next, at the time of binarization, an appropriate threshold level is set. For example, in the case of an image with a relatively low density, the threshold value is set low, and in the case of a high density, it is set high so that a good binary image can be obtained. In this embodiment, the threshold value for binarization can be set and changed in the range of 0 to 255. As a method for obtaining the threshold value, there are a p-Talle method, a mode method, a differential histogram method, a discriminant analysis method and the like.

In the image thus obtained, there may be an unfavorable image analysis image such as an image cut in the middle of particles in the periphery thereof. In order to prevent this, removal by a frame is necessary. In this embodiment, a frame in the range of 0 to 511 pixels in the x direction and 0 to 479 pixels in the y direction can be set, and pixels outside the frame can be set. The image of is removed.

Further, the binarized image is separated by selecting one of the above-mentioned image separating means. Here again, the setting of parameters becomes important although it depends on each separation means.

First, when contraction separation is selected, contraction is set as a parameter. In this embodiment, 1
Each value of up to 99 pixels can be set, and an appropriate value among them can be selected.

In the case of circular separation, the maximum and minimum radii and the separation rate are parameters that can be set. With respect to these parameters, in the present embodiment, the minimum and maximum radii are 2 to 256.
The pixel and the separation rate can be selected in the range of 0 to 1.0.

Further, in the wedge separation, the separation means, the contour line length, the expansion and the contraction are provided as parameters, and in the present embodiment, the setting range thereof is the separation angle 0 ° to 180.
The contour length is 0 to 9 pixels, and the expansion and contraction are 1 to 99 pixels.

By selecting a desired separation means from the above three separation means and selecting each parameter,
A well separated image is obtained. Image analysis is performed on the basis of the separated image obtained in this way, and the particle shape and other particle characteristics described above can be accurately grasped as numerical values.
Moreover, the particle characteristics during the granulation operation are obtained. Therefore, based on the particle characteristics during this granulation, air flow rate, air temperature,
The desired granulation product can be granulated with good yield by controlling the liquid addition rate and other granulation conditions.

In the image analysis described above, when a photographed image is input, it is possible to divide one screen into a plurality of sections and process a plurality of images as one screen. This makes it possible to significantly increase the processing time. For example, as shown in FIG. 5, one screen is divided into four sections, and images at the time of shooting t ₁ , t _2, ... Since it can be processed at once, you can speed up.

As in the above-described embodiments, one of the features of the present invention is to obtain accurate particle characteristics by making the captured images into well-separated images by image processing by a computer. It is also a feature that control is performed based on this accurate particle characteristic.

First, the measurement items in this example regarding the particle characteristics obtained by the image analysis method of the present invention are shown below. (1) Particle diameter (a) Ferret diameter A (longitudinal direction) (b) Ferret diameter B (horizontal direction) (c) Circle equivalent diameter Of these (a), (b) and (c), The ferret diameter A is the length in the vertical direction of the particle, that is, the vertical length, and the ferret diameter B in (b) is the lateral length of the particle, that is, the horizontal length, equivalent to the circle in (c). The diameter is obtained by calculating the area of a particle and using the diameter of a circle having an area equal to this area as the diameter of the particle. (2) Number of particles (3) Criteria for average particle size and particle size distribution (a) Number standard → Average particle size is calculated from particle size and number.

(B) Volume basis → From the volume of particles, the diameter of the volume corresponding to this is taken as the particle diameter, and the average particle diameter is determined from this and the number of particles.

(C) Weight basis → 50% of lognormal distribution
The particle size and the 84.13% particle size are calculated, and the particles are converted into weight values by the following formula.

LogWD (50% diameter) = logND (5
0% diameter) +0.69078 log ² σg In the above formula, WD (50% diameter) is the weight-based geometric average diameter, and ND (50% diameter) is the number-based geometric average diameter. Further, σg is a geometric standard deviation and is given by the following equation.

Σg = (50% particle size) / (84.13% particle size) (4) Particle size distribution (a) Log normal distribution (b) Frequency distribution (5) Average particle size (a) 50% cumulative geometric mean (B) Arithmetic mean diameter (6) σg → Indicates the width and narrowness of the particle size distribution with a value obtained by the following formula.

Σg = (50% particle diameter) / (84.13% particle diameter) (7) Circularity (a) Shape factor (peripheral length) ² / 4π × particle area (b) Ratio of major diameter to minor diameter In the examples, the above-mentioned measurement items can be calculated, and the particle characteristics can be obtained.

What is important in obtaining these particle characteristics is the selection of the most appropriate one of the three separation means, and the selection of each parameter, depending on the granulation method, the growth of particles during granulation, and the like. Select, change, etc.

Selection of these separation means and parameters is as follows.
It may be set based on many years of experience, but in addition, the experiment should be performed first, the analysis should be performed based on the images at that time, and the separation means and parameters should be set based on this before manufacturing. If so, more appropriate settings can be made, which is desirable.

FIG. 9 shows a graph comparing the measured values with those prepared based on the results of image analysis using the method of the present invention.

In FIG. 9, the experimental value is a value obtained by taking out particles during operation and measuring with a microscope. With separation treatment, a value obtained by performing image analysis after separating particles by the method of the present invention. That is, the value measured by the method of the present invention and the value without separation processing is the value obtained by image analysis of the photographed particle image as it is (without using the separation means of the present invention).

As is clear from this figure, according to the method of the present invention (when the separation process is performed), a value extremely close to the experimental value is obtained, whereas when the separation process is not performed, the experimental value is obtained. The value deviates from the value.

From this, it is understood that the method of the present invention can obtain accurate particle characteristics without extracting the particles from the particle image during granulation.

Next, an example of a control method in which the image analysis method of the present invention and the control of the water content by the water content meter are combined will be shown.

In the granulating apparatus shown in FIG. 6, the raw material powder is put into the granulating tank 1, air is blown from the blowing port 4 and the binder is sprayed from the spray gun to carry out the granulation. On the other hand, the granular material being granulated is photographed by the photographing device 10 at predetermined time intervals, and accurate image analysis is performed based on the images separated by the separating means. After the granulation proceeds in this way and the average particle size reaches a desired value, the water content of the particles at the desired average particle size is detected by the moisture meter 7. That is, the average particle size is measured by image analysis using the method of the present invention on the basis of the image captured by the image capturing device 10, and at the same time, the water content at the same point is determined from the detection amount of the water content meter 7, The amount of water at the desired average particle size is measured. After that, granulation is continued while controlling the amount of air from the blower port 4 and the rate of addition of liquid from the spray gun 5 so as to keep the same amount of water constant. After that, the operation is stopped when the particle size distribution obtained from the measurement value by the image analysis by the above method reaches a predetermined value. Thereby, a granulated product having a desired particle size and a desired density can be obtained with good yield.

In this experiment, as raw materials, lactose, corn starch and microcrystalline cellulose were mixed at a ratio of 6: 3: 1,
Further, it is granulated using a powder to which a small amount of acetaminophen is added as a main drug.

Further, when the predetermined particle size distribution is reached, the rotational speed of the rotary plate (slit plate 2) is gradually increased, and at the same time the shape factor is measured by image analysis. When the shape factor reaches the target value, the operation is stopped. To do. This makes it possible to obtain a granulated product having a desired particle size, high sphericity, and high density.

In the above examples, desired particles can be obtained with high accuracy and high yield, as compared with granulation by only controlling the water content.

Further, in the above-mentioned Examples, more accurate granulation can be obtained not only by controlling the water content after the desired particle diameter but also by adopting the image analysis method of the present invention. . That is, after the desired particle size is reached, granulation proceeds while keeping the particle size at a desired value, whereby a high-density granulated product can be obtained.

The above granulation is subjected to image analysis based on the particle image photographed by the photographing device 10, whereby the water content and the air flow are controlled while measuring the accurate particle diameter so that the particle diameter is always constant. Control to be.

FIGS. 7 and 8 are graphs showing measured values when control is performed by the above method. FIG. 7 shows an experiment in which the water content is constant, and FIG. 8 shows an experiment result in which the particle diameter is constant. Is. As is clear from these figures, the particle size is kept exactly constant.

The following table shows another experimental example when granulation was performed by controlling the water content.

10 to 15 are graphs prepared based on the above experimental data.

This experiment was conducted with a desired particle size of 300 μm.
m was selected, and the amount of water when the target particle size was reached was measured by an infrared moisture meter, and granulation was carried out while keeping the amount of water unchanged. The experiment was stopped when the circularity did not change and reached a substantially constant value.

As can be seen from this table, the ferret diameter B increased from the start of the experiment, and At 14, the ferret diameter B reached the target. Instead, control was performed so that the water content at that time did not change, and values such as the Feret diameter were calculated by image analysis. As shown in the table, the ferret diameter is almost constant, and No. The circularity is larger than 19 and is almost the target value of 0.83
Further, the particle size distribution also reaches the target value of 95. Looking at these, the circularity and the particle size distribution are at the target values and are stable at No. The operation was stopped at 28. As a result, particles with a high circularity target value could be obtained with good yield.

Of the various granulation methods, for example, in the case of fluidized bed granulation, if the amount of air fed in is large when the particle size is small, most of the particles are blown up, which is not preferable. Conversely, when particles grow and the number of particles with a large particle size increases, if the air volume is small compared to the liquid addition rate, the particles stick to each other and the particle size becomes larger. However, it becomes difficult to obtain particles having a desired particle size with good yield.

Further, in the case of stirring granulation or tumbling granulation, the number of stirring blades or rotating plates can be changed according to the particle diameter to efficiently granulate high density particles with high yield.

In the fluidized bed granulation, the control method of the present invention is based on the separated image obtained by separating each particle by any of the above-mentioned three separating means based on the image taken by the photographing device. The particle characteristics obtained by the analysis, especially the particle size, are determined.When the average particle size is small, the amount of blast air is small, and when the average particle size is large, the amount of blast air is controlled to be large to achieve good production. Control so that granulation is performed.

The control method of the present invention is to use the above-mentioned image analysis method in the stirring granulator to determine the particle characteristics, especially the average particle size, and when the average particle size is small, decrease the rotation speed of the stirring blade, and When the particle size is large, the number of revolutions of the stirring blade is controlled to be large, so that a granulated product having a large particle density and a desired particle size can be obtained with good yield.

In the rolling granulation, the control method of the present invention is
For the granulated product during granulation, obtain the particle characteristics obtained by the above-mentioned image analysis method, for example, the particle size and the granularity coefficient.When the particle size is small, the rotation speed of the rotating plate is small, and the particle size is large. When it is, the rotation speed of the rotating plate is increased and the granulation is advanced,
By stopping the granulation operation when the desired constant sphericity is reached by measuring the granularity coefficient, for example, sphericity,
It is possible to improve the yield of a granulated product having a desired constant particle size, a high density, and a large sphericity.

[0111]

According to the image analysis method and the granulation control method of the present invention, it is possible to obtain a particle image well separated from a photographed image, and thereby to obtain accurate particle characteristics. Further, desirable granulation control can be performed based on the obtained particle characteristics, and desired granulation products can be obtained with good yield.

[Brief description of drawings]

FIG. 1 is a diagram showing a contraction / separation unit.

FIG. 2 is a diagram showing a circular separating means.

FIG. 3 is a diagram showing a wedge separating means.

FIG. 4 is an enlarged view showing a part of the wedge separating means.

FIG. 5 is a diagram showing a plurality of images on one screen.

FIG. 6 is a diagram showing an example of a granulating apparatus equipped with a projection means and a moisture meter.

FIG. 7 is a diagram showing experimental results when the method of the present invention is applied to granulation by controlling the water content.

FIG. 8 is a diagram showing another experimental result when the method of the present invention is applied to granulation by controlling the water content.

FIG. 9 is a graph comparing the method of the invention (separated particles), as-imaged (non-separated particles) and microscopic measurements.

FIG. 10 is a graph of the ferret diameter B of another experimental result.

11 is a diagram showing a particle size distribution in the case of FIG.

FIG. 12 is a graph of the ferret diameter B during granulation of the above experimental results.

FIG. 13 is a diagram showing a particle size distribution in the case of FIG.

FIG. 14 is a graph of the ferret diameter B when the granulation is stopped according to the above experimental results.

FIG. 15 is a diagram showing a particle size distribution in the case of FIG.

FIG. 16 is a view showing the arrangement of an image capturing apparatus used in an embodiment of the present invention.

FIG. 17 is a diagram showing a modified example of the photographing apparatus.

FIG. 18 is a diagram showing another modification of the image capturing apparatus.

FIG. 19 is a diagram showing still another modified example of the photographing apparatus.

FIG. 20 is a view showing still another modified example of the photographing apparatus.

FIG. 21 is a view showing still another modified example of the photographing apparatus.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Tetsuro Kamada 3-10-18 Kitamachi, Nerima-ku, Tokyo Fuji Paudal Co., Ltd.

Claims (10)

[Claims] 1. A separating means for separating a particle image based on a photographed particle image during granulation, a means for obtaining data of each particle from the particle image separated by the separating means, and based on the obtained data. An image analysis method for particle characteristics, in which the operation conditions are controlled to perform granulation operation. 2. The image analysis method in granulation according to claim 1, wherein the separating unit is a contraction separating unit that contracts a captured raw image to create an image in which each particle image is separated. 3. The circular separating means for obtaining a large number of circles based on the center of gravity and area of each part in the image and separating the particles to form an image in which each particle is separated. Image analysis method in the granulation of milk. 4. The separating means obtains the deepest point of the concave portion where the angle of the concave portion in the image is equal to or less than a predetermined value, and separates the obtained points by a line connecting two adjacent points. The image analysis method in granulation according to claim 1, which is a wedge separation means for separating each particle. 5. The image analysis method according to claim 2, wherein the image analysis is performed by changing the set value of each parameter by the separating means at every predetermined interval from the start of granulation. 6. Granulation while changing various granulation conditions based on measured values such as particle size, particle size distribution, and shape factor at each time point during granulation obtained by the image analysis method of claim 5. A granulation control method in which granulation is performed and granulation is stopped. 7. Granulation in which water content is controlled based on measurement of water content by a water content meter, (average) at each time point during granulation obtained by the image analysis method of claim 5.
A granulation control method, wherein after the particle diameter reaches a predetermined value, the water content is controlled so as to keep the water content constant when the particle diameter reaches a desired value in the water content control. 8. In the granulation in which the water content is controlled based on the measurement of the water content by a water content meter, the particle diameter at each time point during the granulation, which is obtained by the image analysis method according to claim 5, reaches a predetermined value. Then, a granulation control method in which the granulation conditions such as the water content are changed so as to maintain the predetermined particle size. 9. After the particle diameter reaches a predetermined value,
9. The granulation control method according to claim 7, wherein the particle size distribution is obtained by the image analysis method, and the granulation is stopped when the particle size distribution reaches a desired value. 10. The shape factor is determined by the image analysis method after the particle diameter reaches a predetermined value, and the granulation is stopped when the coefficient reaches the predetermined value.
Or the granulation control method of 8.