JP2006514737A - Method and apparatus for measuring the amount of non-aggregating particles in a mixture - Google Patents

Abstract

The present invention relates to a method and apparatus for dynamically measuring the relative amount of particles having different optical properties and / or shapes in a mixture, wherein the particles are moved down, A feeder for delivering particles, an illumination source for illuminating the particles on the inclined passage, an image receiver for recording a reflected light image of the particles, and a composition calculator for determining the relative amount of particles based on the reflected light image Comprising.

Description

  The present invention relates to a method and apparatus that enables the dynamic determination of the proportion of different types of non-aggregating particles in a sample of a mixture of particles.

  Various devices are known for measuring the size distribution and composition of granular materials. For example, U.S. Patent No. 6,057,049 describes an apparatus for determining particle size distribution and characterizing particle shape in a particle mixture. The particles are weighed between the light source and the image acquisition device and roll down vertically under the influence of gravity. A digital record of the silhouette projected area of the particle is used to analyze the particle. A similar device is described in US Pat. There, a silhouette projection method is used. In this device, a collimated laser beam directed to penetrate a sample of descending particles and a means for recording the light that has passed through the sample are used. A digitized image of the silhouette image is used to analyze the particles according to the criteria for surface area, square root of surface area, maximum width, and maximum height. Patent Document 3 describes a particle descending device that is subjected to a silhouette projection method to measure the particle size distribution of particles.

  U.S. Pat. No. 6,057,059 teaches how to generate a falling single layer curtain of particles. In this case, they can then be illuminated with a conventional light source. As in the case of the above-mentioned Patent Document 2, silhouette projection is used. A recording unit for receiving the projected light allows the collection of information about the silhouette of the descending particles and can be used to identify the various colors or gray tones of the particles. The means for generating the particle curtain in a single layer form consists of supplying the particles on a conical vibrating plate that makes an entry angle with the horizon so that the particle residence time on the plate is controlled.

  Patent Document 5 describes a detection device for removing non-standard particles, particularly extremely colored rice kernels. The particles are illuminated as they fall free from the end of the tilt plate and their color is measured. Particles having a measured color that exceeds a threshold level are ejected using an air nozzle.

  According to computer translation, US Pat. No. 6,057,017 illuminates a stationary sample of a mixture of different types of particles and collects an image of the sample using a digital camera, which is then analyzed and identified, for example, by color Means are described that can determine the proportions of the particle types that are produced.

  In Patent Document 7, particles having a constant quality can be produced by using means for continuously measuring the size of the particles and using feedback to adjust the ratio of water and powder supplied to the granulator. A granulation apparatus is described. In the particle sizing method, the particles appear to be deflected from a descending vertical stream of particles onto a tilted plate illuminated by a light source. The reflected light is detected by a digital camera that captures information about the sample, and this data is later converted into size and distribution information that is used to adjust the control of the aforementioned granulator.

  It is desirable to obtain information about non-aggregating particles in a mixture of different particles. In this case, the information not only describes the particle size and shape of the particles, but also for each type of particle according to other observable properties such as color or gray tone that may be indicative of chemical inclusions. The fraction is also measured. In the technical fields cited above, researchers have collected data on particles using a variety of methods, all of which have significant limitations in determining composition.

  In the systems described in Patent Literature 1, Patent Literature 3, Patent Literature 2, and Patent Literature 4, when a particle descends vertically, a projected image or silhouette image of the particle is collected. In this configuration, the particles roll down as they descend, so the collected image is a two-dimensional representation of the three-dimensional particles represented in random directions. Because of the random orientation, the shape and size information that can be determined for individual particles is limited. Furthermore, since a projection image or a silhouette image is used, it becomes difficult to identify a property (for example, color) that requires upper illumination. In US Pat. No. 6,057,086, the ability to determine the color or gray tone of particles using a contrast plate is emphasized, but how to place the contrast plate relative to the cameras and light sources shown on both sides of the curtain of falling particles. In the only example relating to color or gray tones, reference is made to the determination of the number of black and light particles that differ greatly in tone. Patent Document 5 provides a mechanism for identifying defective granules (in this case rice) and then eliminating them. This publication does not indicate an intention to improve the measurement of the composition of the rice sample and provides no mechanism for it.

  In Patent Document 6, since the sample particles are held in a stationary state, one dimension can be presented to the camera while being held in a fixed direction. With such a system, one color or gray tone particle can be identified in contrast to other particles of different color or gray tone. Since it is also possible to determine the area of one type or the other type of particles, a measurement of the ratio of one type of particle to the other type can be made. However, the static configuration of this system can be difficult to incorporate enough particles that do not contact each other to allow imaging of a representative sample. This constraint may require unrealistic time requirements and laborious operations.

  In patent document 7, the particles produced by the granulating device are guided on an inclined plate. The plate is illuminated upward and an image of the particles is recorded by a digital camera as the particles descend. The angle of the plate is adjusted so that the particles move down the plate at “appropriate speed”. The granule shape is not defined, but in the examples spheres are mentioned. The method of delivering particles to the plate is not specified. According to this publication, the particle size and particle size distribution are determined. Without careful control of how the granules are introduced into the tilted plate, it is not guaranteed to align with the plane of the plate, otherwise only spherical granules will be accurately measured. Only. For example, granules having an elongated shape will bounce and the top view image will be distorted, resulting in inaccurate measurement of their true dimensions. The method and apparatus disclosed in this publication is limited to determining the particle size distribution of spherical particles and is not intended to determine the distribution based on particle shape or optical properties.

US Pat. No. 6,061,130 U.S. Pat. No. 4,497,576 UK patent application GB2012948A International Publication No. 89/05971 Pamphlet European Patent Application No. 1083007-A2 JP-A-10-332567 Japanese Patent Laid-Open No. 8 [1996] -89780

The present invention relates to a method for determining the ratio of different particle types in a mixture. This method
(I) non-aggregating particles of at least two particle types in a channel inclined at an angle sufficient to lower the particles along the channel, wherein each particle type is at least one different from the other particle types Providing a mixture comprising optical properties and / or shapes;
(Ii) illuminate the particles along the inclined path;
(Iii) collecting a reflected light image of the illuminated particles;
(Iv) calculating a ratio of at least one particle type based on data of the reflected light image exhibiting at least one different optical property and / or shape. The at least one optical property is preferably at least one of reflectance, luminescence, and variations thereof at visible, ultraviolet, or infrared wavelengths. The inclined passage is more particularly provided by a surface. Preferably, the ratio is based on at least one of a number fraction, a weight fraction, and a volume fraction.

A further aspect of the invention is an apparatus for determining the ratio of particles of different particle types in a mixture. This device
(I) a particle feeder having an exit end;
(Ii) an inclined passage having an upper inlet end located adjacent to and below the outlet end of the feeder so that non-aggregating particles can descend down the inclined passage;
(Iii) an illumination source oriented with respect to the inclined passage so that the particle can be illuminated upward as the particle descends down the inclined passage;
(Iv) an image receiver oriented with respect to the inclined passage so that a reflected light image of the particle can be collected as the particle descends down the inclined passage;
(V) a composition calculator for converting the reflected light image signal received from the image receiver into data indicative of at least one ratio of particle types in the mixture based on at least one optical property and / or shape of the particles; Comprising.

  The present invention can be more fully understood with reference to the accompanying drawings described as follows.

  As used in this disclosure, the following terms shall have the definitions specified as follows:

  “Particle” means a relatively small individual part or amount of solid material, and more particularly a relatively small solid object.

  By “particle type” is meant a class of particles that share a common optical property and / or shape that can be distinguished from other classes of particles having different optical properties and / or shapes. A mixture of particles may comprise more than one particle type. Examples of mixtures comprising two or more particle types include a mixture of red and green particles (where red and green particles each constitute a distinguishable particle type); cylindrical particles And spherical particles (in this case, cylindrical and spherical particles each constitute an identifiable particle type); a mixture of broken and non-destructive seeds (in this case, broken and non-destructive seeds are , Each of which constitutes an identifiable particle type), but is not limited thereto.

  By “non-aggregating” is meant a particle property that suggests that the tendency of movement to be constrained by sticking or sticking to each other can be ignored.

  A “particle type standard” is one that is associated with a specified optical property and / or shape that indicates a particular particle type and distinguishes this particular particle type from other particle types. Particle type standards may be based on dimensional relationships and / or other properties that suggest shape and / or optical properties (eg, color, color distribution, etc.). For example, a white particle type standard identified as a dark particle type may be based on a gray scale threshold (particles above this value are classified as a white particle type). Yellow particle type color standards may specify red, green, and blue intensities. Does the particle have an optical property and / or shape that falls within an acceptable deviation (ie tolerance) from the particle type standard or meets or exceeds the threshold specified for the particle type standard? Can be assigned to a particle type. As another option, particles can be assigned to a particle type based on a determination of which particle type standard is closest in optical properties and / or shape.

  A “particle type composition” is a composition comprising particles of at least two particle types. When describing such particle type compositions by the method and apparatus of the present invention, reference is made to at least one ratio, for example, a relevant fraction such as a number fraction or a volume fraction or a weight fraction. The rate includes the relative amounts of at least two particle types.

  “Number fraction” means the proportion of particles belonging to one particle type, the total number of particles belonging to that particle type and at least one other particle type in the composition of particles of at least two particle types. Is the reference for comparison. The number fraction can be calculated by dividing the sum of the number of particles of one particle type by the sum of the number of particles of that particle type and at least one other particle type. In addition, if the individual particle volume (or a dimension that can be used to calculate the particle volume) is determined, the volume fraction can be derived using the number fraction. Also, if the individual particle weight (or volume available for particle weight calculation) is determined, the weight fraction can be derived from the number fraction. The “number percentage” is a value obtained by multiplying the number fraction by 100%.

  “Volume fraction” divides the sum of the volume of particles of one particle type by the sum of the volume of particles of that particle type and at least one other particle type in the composition of particles of at least two particle types. Is calculated. The “volume percent” is a value obtained by multiplying the volume fraction by 100%.

  “Weight fraction” means the weight ratio of particles of one particle type, based on the weight of that particle type and at least one other particle type in the composition of at least two particle types. become. Determining the weight of an individual particle from its volume and envelope density, and then dividing the sum of the weight of particles of one particle type by the sum of the weight of particles of that particle type and at least one other particle type; The weight fraction can be calculated. The “weight percent” is a value obtained by multiplying the weight fraction by 100%.

  “Envelope density” refers to the weight-to-volume ratio of individual particles and is substantially uniform within each particle type.

  “Bulk density” means the weight to volume ratio of a bulk sample composed of many particles. The volume of the bulk sample includes not only the volume of the individual particles but also the volume of the gap between the particles.

  “Bounce angle” means the deviation angle of a vector through the longest dimension of a particle, and the particle is in equilibrium (ie, stationary) on an inclined path or surface that is described for use in the method and apparatus of the present invention. The position of the vector at More specifically, the bounce angle is a declination component in a direction perpendicular to the inclined passage or the inclined surface (that is, a direction away from or toward the inclined passage or the inclined surface), and a direction parallel to the passage or the surface. Is not included.

  “Luminescence” refers to the emission of light at one or more electromagnetic frequencies caused by illumination with light at other electromagnetic frequencies; luminescence includes, for example, fluorescence and phosphorescence.

  “Reflectivity” means the ratio of the amount of light in a particular electromagnetic frequency or electromagnetic frequency range reflected by a surface to the amount of light at one or more frequencies incident on the surface.

  The “reflected light image” means an image of an object illuminated from the same overall direction as when the object is imaged. Most commonly, the reflected light is caused by the reflection of light in the same overall spectral region as the illumination. However, the reflected light may include light due to fluorescence and phosphorescence. For example, when ultraviolet light illumination is performed on an object having a fluorescent or phosphorescent surface, visible light may be emitted and collected to form a reflected light image.

  “Grayscale” means a numerical (typically integer) scale that specifies the relative amount of light collected from a video camera onto pixels of a digitized image. Gray scale can be used to assign a numerical value to each pixel of the digitized image. This value is related to the amount of reflected light from the portion of the imaging object captured by the pixel. If essentially no light is collected on the pixel (eg, imaging a non-reflective black background), the grayscale level is typically assigned a value of zero. A particular high level or higher level of light collected on a pixel (eg, light when imaging a sharp white object) is typically assigned an upper grayscale value. This value is 255 for 16-bit binary grayscale.

  “Shape” means an identifiable three-dimensional form that can be described by a geometry-based or empirically determined relationship between its defined dimensions. The simplest shape based on geometry is a sphere. This can be fully described by specifying only its diameter. All projections of the sphere show the diameter. A cylinder with a circular cross-section can be described by specifying its length and diameter. The length and diameter of the cylinder can generally be determined from the length and width of its rectangular top view image. In this case, if the cylinder is placed on its side, the length (longest dimension) of the rectangular image corresponds to the length of the cylinder, and the width (shortest dimension) of the rectangular image corresponds to the diameter of the cylinder. become. Using this information, the volume of the cylinder can be calculated. Other regular shapes include ellipsoids (shapes similar to Australian football or American football). This can be described based on its longest and shortest dimensions. This information can also be obtained from the top view. For irregularly shaped particles, it may be necessary to collect a variety of empirical information prior to analysis in order to identify the particles by shape. However, if there is a sufficient shape difference, such as flake shaped particles versus bead-like objects (eg, corn flakes vs raisins), it is not difficult to distinguish between different shaped particles. Examples of different shapes include spheres, cylinders, ellipsoids, cubes, rhomboids, disks, flakes, corn seed shapes, and the like. A particular shape can take various dimensions. For example, the cylinders can differ from each other in that they have different diameters and / or lengths. Different sized cylinders may share the same proportionality and have the same length-to-width ratio, while cylinders may also differ in that their length-to-width ratios are different.

  “Spherical” means a rounded shape similar to a sphere. All planar cross-sections that pass through the spherical shape are substantially circular. The three vertical axes that pass through the center of the spherical shape have approximately the same length.

  “Ellipsoidal” means that some planar cross-sections are not circular but elliptical (ie, oval), and the ellipsoidal shape can be described by three vertical axes that penetrate the center of the ellipsoid, Means a rounded shape that differs from a sphere in that at least one of the axes is not substantially equal to the other axis (ie, combined with another axis to define a non-circular ellipse) To do.

  The details provided herein are given by way of example only and are intended to exemplify the embodiments of the invention and what is considered most useful. Are provided to facilitate understanding of the principles and conceptual aspects of the present invention. In this regard, it is not intended that the details of the invention be set out in more detail than is necessary for a basic understanding of the invention, but how some forms of the invention may actually be implemented is described in this book. The description will become apparent to those skilled in the art.

  By using the method and apparatus according to the present invention, extremely accurate particle analysis can be easily performed. By using such an apparatus and method, particles in a particle sample having different optical properties and / or shapes can be identified based on a number of optical and / or dimensional characteristics. For example, such optical and / or dimensional characteristics include, but are not limited to, color, color variation, destructive versus non-destructive shape, size, roughness, and brightness. The apparatus and method of the present invention solves the problems encountered with previous devices previously described in the background section of this specification.

  The method and apparatus of the present invention is used to calculate the ratio of different particle types in a particle mixture comprising non-aggregating particles of at least two particle types identified based on different optical properties and / or shapes. Useful. The method and apparatus of the present invention can be used not only to determine the ratio of particle types in a particle type composition, but with this method and apparatus, the longest dimension, the shortest dimension, It is also possible to calculate dimensions such as area by determining dimensions such as perimeter. In addition to determining the proportion of at least one particle type in the particle type composition, the method according to the invention is particularly useful when the reflected light image is also used to calculate at least one dimensional property of the particles. . Preferably, the at least one dimensional characteristic is at least one of a longest dimension, a shortest dimension, an area, and a circumference. If the thickness is known or can be inferred from the top view, the particle volume and volume fraction can be calculated. (If the particle volume is known by other methods, the volume fraction can be calculated similarly.) If the envelope density is also known, the particle weight and weight fraction can also be calculated. (If the particle weight is known by other methods, the weight fraction can be calculated similarly.)

  The method and apparatus of the present invention can be used with a wide variety of particles that are non-aggregating. When electrostatic interactions work, very small particles can adhere and integrate, and this effect is more pronounced when particles have low density. Accordingly, the methods and apparatus of the present invention typically have a surface that can accommodate an inscribed sphere of at least about 0.1 mm in diameter, more preferably at least about 0.5 mm in diameter, and most preferably at least about 1 mm. Used for having particles. The method and apparatus of the present invention is most conveniently used for particles having a longest dimension of about 5 cm or less, but larger particles can be used if the device components, including inclined surfaces, are appropriately scaled up. Can do. Examples of particles useful in the present invention include, but are not limited to, granules, chips, pebbles, candy, jewelry, beads, metal shots, bullets, coins, seeds, processed foods, minerals, pills, and the like. is not. In a preferred embodiment, the method and apparatus of the present invention is used with a mixture of non-aggregating particles comprising particles that are substantially cylindrical in shape. In a more preferred embodiment, the method and apparatus of the present invention is used for a mixture of non-aggregating particles comprising particles having a substantially cylindrical shape with a circular cross section.

  Preferably, the method and apparatus of the present invention is used for a particle mixture comprising seeds and / or particles comprising at least one agriculturally active substance. Agroactive substance shall mean any kind of biologically active substance used in agriculture. Agricultural active substances include, for example, crop protection agents (eg herbicides, fungicides, fungicides, bactericides, invertebrate pesticides), plant growth regulators (eg flowering or fruiting promoters or Suppressors, harvest enhancers or organ detachment agents, growth and regrowth inhibitors), leaf drying agents, chemical hybrid formation agents (eg, pollination inhibitors or sterilants), and plant nutrients (eg, fertilizers) Can be mentioned. In a preferred embodiment, the method and apparatus of the present invention is used for a mixture of non-aggregating particles comprising particles comprising at least one crop protection agent. Crop protection agents are most commonly chemicals, but may also be biological agents such as Bacillus thuringiensis and entomopathogenic bacteria, viruses, and fungi. A mixture of non-agglomerated particles comprising an agriculturally active substance comprises non-aggregating particles that do not contain an agriculturally active substance, but contain other useful ingredients such as surfactants. Also good. In a preferred embodiment, the method and apparatus of the present invention comprises a particle comprising at least one agriculturally active substance and has a substantially cylindrical shape, typically a substantially circular cross-sectional shape. It can be used in a mixture of non-aggregating particles. The substantially cylindrical non-agglomerated particles comprising at least one agriculturally active substance are, for example, wet compositions (eg in the case of paste extruded granules) or heating compositions comprising a heat activated binder (eg , In the case of hot extruded granules), or more generally by extrusion (extruded granules are provided).

  Agroactive substances can also be used for non-crop cultivation applications. For example, invertebrate pest control agents can be used in other horticultural applications (eg, forests, greenhouses, nursery plants, or ornamental plants that are not cultivated in the field), general (human) and animal health applications, residential structures applications and It may be useful for commercial construction applications, household products applications and storage applications.

  When carrying out the method according to the invention, the non-agglomerated particles are fed into an inclined passage (typically provided by an inclined surface). The particle descends below it by rotation and / or sliding substantially under the control of gravity. Thus, as the particle descends below it, it will generally remain in contact with the surface and the longest axis of the non-spherical particle will tend to be aligned parallel to the surface. The particles are illuminated when descending down the inclined passage. In this configuration, it is possible to record using the individual reflected top view image of the particle, preferably by a video camera or the like oriented perpendicular to the inclined path. Image analysis can identify optical properties and / or shapes and determine dimensional characteristics such as longest dimension, shortest dimension, area, and perimeter. If other parameters are available, particle volume, volume fraction, particle weight, and weight fraction can be calculated.

  To facilitate image analysis, preferably the particles are also substantially more substantial so that particle contact (eg, contact between one particle and the other) is minimized when descending along the inclined path. To be held in the plane of the passageway. Preferably, less than 25% of the particles on the slant path are in contact with other particles in the reflected light image, more preferably less than 10%, most preferably less than 2% of the particles in the reflected light image. It is in contact with other particles. In order to minimize particle contact while maximizing throughput, the particles are preferably fed so that they are uniformly deposited along the inlet end of the inclined surface providing the passage.

  The supply according to the invention comprises arranging the particles in the vicinity of the outlet end of the feeder where the particles are supplied to the inclined passage, and the inclined passage is adjacent to the supply outlet end and located below it Comprising an end. In order to cope with supply movement, a gap is typically defined between the supply outlet end and the inlet end of the inclined passage. In order to place the particles substantially in the plane of the surface and minimize the bounce angle of the non-spherical particles, the gap is preferably below the shortest dimension of the particles to be measured. When the bounce angle of the particles is 10 degrees or less, the dimensional measurement of the top view image of the non-spherical particles is more accurate. In an embodiment of the invention in which the particles are non-spherical, Applicant has determined that if the gap between the feed outlet end and the inclined channel inlet end is less than or equal to the shortest dimension of the particle, the particles on the inclined channel It has been found that at least 80% of these have a bounce angle of 10 degrees or less. As the gap is narrowed, the percentage of particles having a bounce angle of 10 degrees or less may decrease. Preferably at least 90%, more preferably at least 95% of the particles have a bounce angle of 10 degrees or less.

  The particles are supplied to the inclined passage by a particle feeder. A wide variety of feeder designs are suitable for the method and apparatus of the present invention. The particles can be supplied using vibration, or mechanically conveyed, such as on a conveyor belt. In one embodiment, a sample of a physical mixture of two or more particle types is fed to the feeder by a hopper. The sample in the hopper contains enough particles to comprise a representative sample of the mixture to be measured. It is also within the scope of the present invention for the hopper and feeder to be a single integrated unit. Preferably, the feeder is arranged at a distance not higher than the shortest dimension of the analyte particle upward from the upper end of the supplied inclined surface, preferably closer to the upper edge of the inclined surface. A horizontal vibration feeder having Preferably, the edge of the feeder is aligned parallel to the inclined surface and its center is positioned above the center of the inclined surface. The width of the outlet end of the feeder can be equal to or less than the width of the inclined surface.

  The particle feed rate can be set by adjusting the particle feeder design and / or the configuration of the elements in the particle feeder. As another option, the particle feed rate and particle feeder can be controlled by a feed controller that can vary, for example, the frequency of a horizontal vibratory feeder or the speed of a feeder with a conveyor belt. The feed controller can be adjusted manually or based on feedback from a composition calculator or related computing system.

  The tilted surface need only be tilted to a degree sufficient to lower the particles down under gravity control, for example by rotation or sliding. The minimum tilt angle (tilt angle relative to horizontal) sufficient to impart motion to the particles depends on factors such as the size, density, and coefficient of friction with the surface of the particles. If the inclined surface is vibrated, the minimum inclination angle can be reduced to a very small value, so it is within the scope of the invention to vibrate the inclined surface, but excessive vibration causes excessive bounce. In an exemplary method and apparatus of the present invention, the inclined surface is not significantly oscillated, as it can potentially cause artificial enlargement or reduction of the particle image. Only very little experimentation is required to determine the minimum tilt angle for a particular tilted surface and particle mixture. Preferably, the steeper angle is significantly greater than the minimum tilt angle, since steeper tilt angles promote faster particle movement and improve throughput. Typically, the tilt angle is at least 30 ° with respect to the horizon. However, if the particle separation occurs as a result of faster particle movement caused by the increase in tilt angle and cannot be compensated for by the feeder, there may be too few particles in each image. Furthermore, if the inclination is too steep, the particles roll down and the bounce angle may exceed 10 degrees. Accordingly, the inclined surface preferably makes an angle of about 60 ° or less with the horizon. Typically, an inclined surface that makes an angle of about 45 ° with the horizon will work well with the method and apparatus of the present invention.

  The inclined surface may be planar, convex, or have one of several “V” shaped cross-sections that direct particles to parallel passages. “Convex” means a slightly curved surface that slopes downwards from the center line toward both sides, the purpose of which is to increase the separation of particles from one another as they descend the slant path. It is. A planar surface is preferred. This is because the design is simple and it can be kept at a relatively constant focal length from the image receiver (eg, camera).

  The inclined surface is preferably coated or painted to be reasonably non-reflective and is made to provide a contrast that allows the recording of an accurate top view image of the particles. For example, the surface comprising a smooth non-reflective glass coated with a matte black paint on the back side works well when using visible light, black and white cameras, and dark gray particles from gray to dark brown To do. This configuration provides a smooth surface of the glass where the particles descend and a dark background that provides good contrast suitable for image analysis. As long as the surface is smooth enough to lower the particles without causing a reduction in speed, the specific coefficient of friction between the inclined surface and the particles is not critical.

  Preferably, the inclined surface is easily blown and does not contain accumulated dust. In one particularly preferred embodiment of the invention, even when the non-agglomerated particles contain a small amount (eg 0.2% by weight) of dust, they are generally parallel to the inclined surface but slightly downwards. Blow an inert gas stream to remove dust particles that may accumulate. Such accumulation can result in the formation of images that do not represent actual particles or the generation of static fragments that can be counted more than once. By “inert gas” is meant any gas that does not react with the particles or apparatus of the present invention. Typically, the inert gas is air. Since gas flow is not necessary for particle transport, the gas flow velocity and the angle at which the gas flow is directed in a plane should not be so great as to cause particle turbulence or bounce. The gas flow may be continuous or periodic as convenient or necessary. The gas flow can be provided by one or more nozzles intended to blow the gas stream substantially parallel to the inclined surface, but slightly downward. The gas stream can be delivered most simply by a single nozzle with a slit-shaped orifice arranged parallel to the inclined surface and having a width similar to the width of the inclined surface.

  The terminal velocity of a particle on an inclined surface is the lower limit velocity when the particle velocity naturally decreases (toward the terminal velocity) as the particle moves down the inclined surface. Terminal velocity will depend on the size and density of the particles as well as the tilt angle and the coefficient of friction between the particles and the tilted surface. In order to prevent the particles from aggregating on the inclined surface due to the reduced speed, the velocity of the particles from the feeder must be less than the terminal velocity of the particles descending the inclined surface. Typically, the velocity of the particles from the feeder is much slower than the terminal velocity, and the particles are accelerated as they move down the inclined surface, but are relatively common in the methods and apparatus of the present invention. With a short surface length, the terminal velocity will not be reached. The feed rate of the particles to the inclined surface can be adjusted to minimize contact with each other as the particles move down the surface.

  Particles delivered from the bottom edge of the inclined surface can be handled as needed and preferred. Typically, the particles are collected in catch trays for disposal or recirculation. If the number of particles imaged and analyzed in the first pass of particles from the feeder is not sufficient as a representative sample, the collected particles can be returned to the feeder for further passage through the device.

  The apparatus of the present invention can also be coupled to the output of the mixing system so that the proportions of different particle types in the output mixture can be determined in parallel. A representative sample of the output mixture can be removed periodically or continuously and fed to the apparatus. The analyzed particles can then be returned to the output stream of the mixing system. The mixture composition variation over time can then be shown by recording and displaying the ratio of the different particle types in the output mixture determined by the method of the present invention as a moving average or for a fixed time.

  The particles are sufficiently illuminated so that they can collect a reflected light image of the top view of the particles as they descend down the inclined passage. Typically, the illumination is performed using visible light normally provided by one or more artificial light sources, although sunlight can also be used. If the image receiver (for example, a receiver including a camera) is sensitive to the respective electromagnetic wave length, an infrared light source and an ultraviolet light source can be used. Also, if the particles to be imaged are fluorescent or phosphorescent (when conversion from ultraviolet light to visible light occurs), an ultraviolet light source can be used with an image receiver that is sensitive to visible light. Suitable light sources are, for example, incandescent lamps (eg tungsten filaments in nitrogen and / or argon; tungsten filaments in halogen: quartz-halogen lamps, tungsten halogen lamps), discharge lamps (eg fluorescent lamps, mercury vapor lamps, Sodium vapor lamps, metal-halide lamps, carbon-arc lamps, xenon flash tubes), semiconductors (eg, light emitting diodes), and flame heating (eg, Welsbach mantle and limelite). An incandescent lamp is the preferred visible light source because it is simple. A mercury vapor lamp and a fluorescent lamp having an ultraviolet wavelength emitting phosphor are preferred ultraviolet light sources.

  In order to adequately illuminate the part of the particle from which the top view image is collected, it is necessary to arrange one or more light sources. In particular, where the light source provides narrow illumination rather than diffusely spreading over a large area, the light source is preferably located near the image receiver and oriented in approximately the same direction as the image receiver is directed. It is done. Typically, the illumination provided by one or more light sources is continuous, but when using a specific lamp design (eg, a xenon flash tube), illumination is provided only when the image receiver receives an image. Is done.

  A top view image of the particles is collected using reflected light captured by an image receiver directed in the direction of the inclined path. Preferably, the reflected light image of the particles is collected in a direction substantially perpendicular to the inclined passage, so that it is necessary to point the image receiver substantially perpendicular to the inclined passage. The image receiver comprises a camera with a lens system that forms an image on the photosensitive element. The photosensitive element can be, for example, an orthicon tube or a vidicon tube or a charge coupled device (CCD). A monochrome camera (ie, a “black and white” camera that can only record all light levels) may be sufficient, or a color camera may be used if the particle type is identified according to the optical properties involving different colors. . In a preferred embodiment of the device according to the invention, the image receiver is a color image receiver and comprises a color camera.

  Although a single monochrome or color camera can be used to analyze most mixtures of particle types, the invention is not limited to the use of a single camera. As an example, it may be advantageous to use two cameras, preferably synchronized. For example, if one particle type is a type that causes strong reflection in the infrared region of the spectrum, an image is provided using a single camera with sensitivity in that region, and particles that cause strong reflection in the infrared region are detected. Can be identified. The visible light camera can then be synchronized to the first camera to collect images used to identify particles that do not cause strong reflections in the infrared. Either camera can be used to collect dimensional information.

  The camera can generate an analog signal or a digital representation of the image. When the camera produces an analog signal, the signal is represented by a digital representation of the image (referred to herein as a digitized image) by a signal processing element in the image receiver, a unit for calculating composition, or other unit Convert to Such digitization techniques are very common in the art. A digitized image means that if the image is represented as a matrix of pixels (ie, pixels) and the image is black and white (ie, monochrome), each is assigned a grayscale value. If the image is color, each pixel is assigned several values according to one of several color image representation schemes (for example, in the RGB system, the intensity value is for each of the red, green, and blue components). Assigned). The digitized image can then be processed by a composition calculator. In a composition calculator, the data collected from the digitized image is compared to a threshold or other particle type standard defined for a particular particle type, thereby at least one optical property of the particle and / or At least one ratio describing the particle type composition based on the shape is determined. A composition calculator typically comprises a digital processing unit and memory (eg, a digital computer), and typically one or more output devices of the type commonly used in computers (eg, , Inkjet paper, impact paper, thermal paper or photosensitive paper, laser, toner, thermal wax transfer, and / or dye sublimation technology, light emitting diode (LED) array, cathode ray tube display (CRT), liquid crystal display (LCD) ), CRT technology, LCD technology, or a projector based on mirror (DMD, DLP) technology) to output output composition data (eg, report). Composition data is also removable storage such as magnetic disk drives (eg floppy diskettes, Zip® disks) and optical disk drives (eg CD writers and DVD writers) for storage or transfer to other computers. It can also be output to the device in electronic form. The composition data can also be transmitted to other computers via a network based on, for example, conductive wires, optical fibers, or radio frequency transmission.

  The number of images processed per second by the system is typically such that the particles are preferably 2 before exiting the camera field of view, such that more than 10% of the particles in the sample to be analyzed are imaged. It is selected according to the moving speed of the particles and the field of view of the camera so that they are not imaged twice.

  Preferably, the shutter speed of the camera is rapid enough to capture a particle image with an artificially enlarged motion that is no more than 1-2% of the true length in the direction of motion. In the present invention, the particle velocity is generated primarily by gravitational acceleration constrained by frictional forces with the inclined surface and ambient air. The enlargement of the image size results from the distance that the particles move between opening and closing of the shutter. A typical shutter speed of a commercially available camera is 1/8000 second. If the particle moving speed is kept below 100 mm / sec, any dimension will not result in an enlargement of greater than about 0.01 mm. Thus, an image of a 3 mm long moving particle moving down an inclined surface will not appear to be magnified beyond 0.06 mm. If higher measurement accuracy is required for a particular analysis, this can be done using a camera with a faster shutter speed, or by providing a means to reduce particle speed, for example 15 It can be achieved by using a surface angle of between -30 °. Also, since image enlargement is predictable and systematic, it is possible to correct the area, width, and length of the image by incorporating correction factors into the software (this is sufficient Within the scope of technology).

  Preferably, the camera is positioned or the camera zoom lens is adjusted so that the full width of the tilted surface is captured in the field of view, i.e., all particles in the sample pass through the field of view of the camera. Ensuring that all particles pass through the field of view helps ensure that a representative sample is imaged. Preferably, the downward length of the inclined surface is at least as long as the field of view of the camera. For economic and convenience purposes, the downhill length of the inclined surface is typically not much longer than the field of view of the camera.

  The camera, the light used, the illumination method, the light absorption characteristics of the inclined surface, and the software that analyzes the data from the camera images are selected according to the particle characteristics so that particles of different particle types can be distinguished from each other. For example, if the particle type difference is based on different shapes or different grayscale values, a simple black and white camera may be appropriate. If there are color differences between the different particle types, a color camera will be preferred or required. If the difference in infrared or ultraviolet light reflectance magnitude is the basis for particle type identification, then a light source and camera that is useful for the electromagnetic frequency of the appropriate infrared or ultraviolet light is needed. Particle types with different compositions (eg, different crop protection actives) otherwise do not have sufficient shape, color difference, or color difference to distinguish them according to the method of the present invention It would be possible to enhance the color or tone of one particle type as part of its manufacture so that it can be identified. For example, additional pigments can be added to one particle type to enhance the color or gray level contrast identified with other particle types. An inert agent such as titanium dioxide or carbon black can be added to the composition to lighten or darken the color, respectively. Another possible method is to add a fluorescent component to a composition of one particle type so that visible light is emitted when irradiated with ultraviolet light. This method is preferred only if a simpler solution cannot be used because it requires the exclusion of ambient visible light.

  In an exemplary embodiment of the invention shown in FIG. 1, a particle mixture containing at least two particle types to be analyzed is placed in a hopper 1. Particles 3 are supplied from a particle feeder 2 (shown as a horizontal vibratory feeder) to an inclined surface 4 that is inclined at an angle α (shown near 45 °) with a horizontal line. The particle feed rate is controlled using the feed controller 11 in manual mode or in feedback mode based on feedback from the composition calculator 8. The feed rate is controlled so that the particles are hardly in contact with each other as they descend below the surface, and the same particles are preferably not imaged twice in successive images. As the particles descend below the inclined surface 4, the illumination source 5 (shown as multiple units) illuminates the particles 3, while the image receiver 6 (shown as a video camera) moves. The image of the particles to be captured The image is analyzed using a composition calculator 8 (for example, a digital computer) that outputs a composition report using the output device 9. The output device 9 typically comprises at least one computer monitor. In this exemplary embodiment, the particles 3 delivered from the inclined surface 4 are collected in a catch tray 7 and using a nozzle 10 to keep the inclined surface 4 free of dust. A stream of inert gas is blown.

  FIG. 2 shows how a typical monochrome camera image of cylindrical particles with two different gray scale values on a black colored inclined surface looks.

  3A and 3B show how particle bounce occurs. If the particles to be measured are spherical, bounce along the inclined surface is not a problem with respect to measurement accuracy. However, in the case of non-spherical particles, this is not the case because substantial accuracy bounces along the inclined surface will have a substantial impact on measurement accuracy. FIG. 3A shows an enlarged schematic view of the relationship between the particle feeder 2 and the inclined surface 4 when it includes cylindrical particles 3 having a circular cross section with the shortest dimension being the diameter d. The exit end of the particle feeder 2 is arranged at a distance h above the front edge of the inclined surface 4. The particles 3 are tilted slightly when delivered from the particle feeder 2 because there is a distance that the leading edge of each particle must descend before contacting the inclined surface 4. As a result, a part of the particles 3 may slightly bounce.

FIG. 3B shows that when a particle is displaced from the plane by a bounce angle β in the direction of particle length l, the corresponding image length l _s of the particle in the top view image created perpendicular to the inclined surface may be reduced. It shows that there is. The relationship between the ratio of image length l _s and particle length l as a function of bounce angle β is given by equation 1.
Formula 1
l _s / l = cos (β)
If the bounce angle β does not exceed 10 degrees, the image length l _s will be less than 2% less than the particle length l, which will provide more than 98% accuracy in the top view image.

  Depending on other particle dimensions (eg, thickness), the bounce angle β may also artificially increase the image length due to the side of the particle being face up and incorporated into the top view image. It can be a cause. Since neither artificial image reduction nor enlargement is desirable, the bounce angle β is preferably minimized.

  As shown in FIG. 3A, the bounce tends to be greatest immediately after the particle 3 is delivered from the particle feeder 2 to the top of the inclined surface 4 and decreases with subsequent movement down the inclined surface 4. Preferably, at least 80% of the non-spherical particles on the inclined surface 4 have a bounce angle of 10 degrees or less. More preferably, at least 90% of the non-spherical particles have a bounce angle of 10 degrees or less. Most preferably, at least 95% of the non-spherical particles have a bounce angle of 10 degrees or less. For spherical particles, the bounce angle is not a problem. Applicants have found that the bounce angle β of the particles 3 on the inclined surface 4 is typically less than 10 degrees when h is less than or equal to d.

  The ratio of different particle types within the particle type composition is calculated using a composition calculator that analyzes a digitized image of the particles on the inclined surface. Based on the analysis of digitized images that identify different properties based on optical properties and / or shape, the compositional calculator will assign each analyzed particle to one of two or more particle types Identify as Identifying a particular analyzed particle as belonging to one of two or more particle types compares the data collected from the digitized image of the particle and puts it into a particular particle type This can be done by comparing to a threshold defined for other or other particle type standards. The number of particles assigned to each particle type can then be summed. The number fraction of a particular particle type is the sum of particles having that particular type divided by the sum of particles of that particle type and at least one other particle type.

  The computational method required to identify the particle type based on optical properties and / or shape will depend on the particular optical difference and / or shape difference identified. While the method can vary greatly, suitable methods will be apparent to those skilled in the art in view of the particular characteristics identified. Incorporating such methods into computer code is well within the ability of those skilled in the art.

  FIG. 4 illustrates a method for particle type identification based on the optical properties of the digitized top view image of the particles. The centroid 21 of the top view of the image is positioned at the intersection of the rectangular diagonals 20a and 20b. Next, the image data of 25 pixels in the kernel 22 around the centroid 21 is analyzed. For example, when identifying particles using grayscale, select a “threshold” grayscale value so that the light colored particle pixel value is greater than or equal to that value and the dark colored particle pixel value is less than that value. To do. Each measured particle can then be identified and tagged as a member of either the dark or light colored group. All particles in the data set are evaluated and assigned to one group or the other. If the color is an identifying optical property, determine the average intensity of several wavelengths of light collected by the pixels in the kernel 22 and compare to the standard color value to identify the most likely particle color . This is repeated for all collected images and then each imaged particle is assigned to a particle type defined by color.

  Distinguishing one type of particle from the other is the reflection and luminescence in the visible, infrared (including near infrared), and ultraviolet regions of the electromagnetic spectrum and their variations (variations with electromagnetic frequency). , And even spatial variations such as stripes, banding, mottling, and uniformity). In such a case, analyzing the 25 pixel kernel around the centroid of the particle image may not be a proper calculation method. For example, for striped particles, it is possible to examine sections along the length of the particle image, thus checking for grayscale or wavelength variations, and selecting methods that suggest different patterns. In the visible light region, different optical properties typically include light that is not equally reflected at all frequencies, resulting in color (eg, described by hue and saturation) and different overall lightness. Infrared and ultraviolet light is also typically not reflected equally at all frequencies, so this variation can also be used for analysis. Agents that emphasize differences in optical properties to allow accurate identification of particles during analysis (eg, carbon black for darkening, titanium dioxide for lightening, phosphors to induce fluorescence or phosphorescence) The particles of one particle type in the mixture can be formulated to contain.

  Calculate the ratio of at least one particle type in the particle type composition and use various calculation procedures or algorithms, all well within the ordinary skill in the art, to determine the longest dimension, shortest dimension, area, and The dimensional characteristics of the top view image, such as the perimeter, can be determined. For example, commercially available software such as LabView IMAQ from National Instruments Corporation (Austin, Texas) can be used to generate a digitized top view of a particle two-dimensional image. Provides information such as area and perimeter.

The top view area of a particle can be easily determined by calculation by first counting the pixels in the digitized image. FIG. 5 shows a digitized form of a representative particle image obtained from FIG. 2 (ie, a rectangular top view image shape obtained from imaging a cylindrical particle or the like). Next, in order to give a top view area in a quantity unit such as mm ² , the reciprocal of the square of the digitized image calibration coefficient (hereinafter simply referred to as “calibration coefficient”) is added to the number of pixels in the digitized particle image. Multiply. The calibration factor relates the digital measurement of the digitized particle image to the actual particle size and is expressed by the pixel width in the digitized particle image corresponding to the physical unit distance (eg, mm) of the imaged particle. . For example, by placing a thin circular matted white colored disc of known diameter on an inclined surface and dividing the number of pixel widths covering the disc image diameter by the actual disc diameter (eg, diameter in mm) The calibration factor can be determined.

  The determination of top view particle area based on pixel count is subject to systematic errors. “Systematic error” means a regular and predictable error with respect to the particular configuration of system elements and the measurement method used. Systematic errors in area measurement are primarily caused by the resolution of the digitized image (which produces a finite calibration factor) and the limitations of the procedures and algorithms that define the boundaries or edges of the particle image by pixels. The actual boundary of the particle image will often cross the middle of the pixel. Counting the crossed pixels as part of the image tends to estimate the top view area of the particle slightly larger, whereas if you do not count the crossed pixels, you will estimate the top view area of the particle slightly less. Let's go.

  Figure 5 shows the measurement method used counts the pixels crossed by the particle boundary as part of the particle image? A particle image is shown. In FIG. 5, each square represents a pixel. The hatched portion shown in FIG. 5 represents the area determined by the measurement method, and in this case, an area of 456 pixels is generated. However, if the pixels that the particle image intersects can be treated as decimals to give a true top view area, it will be appreciated that the area only corresponds to 405 square pixels. Therefore, the relative error is (456-405) /405=12.6%. Smaller particles will generally exhibit a larger positive relative error, and larger particles will exhibit a smaller positive relative error. Not only increasing the particle area, but also increasing the calibration factor reduces the relative error. If this systematic error can be predicted, the measured particle area can be adjusted to more accurately represent the actual area. Also, in the analysis of the amount and frequency of light captured by each pixel (eg grayscale, color depth), the pixels where only the secondary part is occupied by the particle image are excluded and the main part is occupied by the particle image It is possible to count the number of processed pixels.

  Another measurement that can be made using a digitized particle image is the shortest distance (ie, its width) across the image. There are various ways of determining this distance, all within the skill of the art, each having an error factor associated with it. By way of illustration, a method of determining a specific width that is useful when the particle image is rectangular (eg, obtained from cylindrical particles) will now be discussed. In FIG. 5, the centroid 31 of the top view of the image is positioned at the intersection of the rectangular diagonals 30a and 30b. A horizontal line 32 and a vertical line 33 are drawn from the centroid 31 of the digitized particle image pixel to the edge. Line 34 is drawn between horizontal line 32 and vertical line 33 at an angle of 45 °. Next, three additional lines 35, 36, and 37 are drawn between the two shortest lines of lines 32, 33, and 34 (in this case, lines 32 and 34). Next, the position of the width is defined by the shortest of the lines 32-37 (in this case, line 34), and this line is extended from the digitized particle image pixel to the background pixel. The width is then specified as the distance between the two pixels at the ends of the line where the digitized particle image is in the background. After calculating the width and area, the length of the rectangular digitized particle image can be calculated by dividing the area by the width.

  FIG. 5 also shows how errors can occur when measuring the width. The exact width measurement is the hypotenuse of a triangle with sides a 'and b'. What is actually measured is the hypotenuse of a triangle with sides a and b. In this particular method, the relative error will always be positive because the line is drawn to the outer boundary of the pixel that may only be partially occupied by the particle image. In some cases, as in the case of lines 32 and 33 in FIG. 5, the true value and the measured value are essentially the same. This occurs because the line is drawn at 0 ° and 90 °. Both of these lines are parallel to the pixel boundary grid lines and in this case coincide with the pixel boundary grid lines. None of the lines 33, 34, 35, 36, and 37 may represent the shortest line that can be drawn from the centroid 31 to the particle image transition to the background. Cause an error factor. By incorporating corrections into the software using methods well within the skill in the art, these systematic errors can be reduced, but additional computer processing time or speed may be required. Also, for the area calculation, the amount and frequency of light captured by each pixel (so that only the sub-portion excludes pixels occupied by the particle image and counts pixels where the main portion is occupied by the particle image ( For example, by performing analysis of gray scale and color depth, it is possible to reduce errors when measuring the width.

  If the particle volume of a particle type is a constant known value, the volume fraction can be calculated using the known particle volume after determining the number of particles of each particle type. If the particle volume of the particle type is not constant, to calculate the volume fraction, it is necessary to determine the volume of each particle.

Depending on the availability of dimensional information (eg, thickness) that cannot be measured directly but can be inferred from top view particle images for some shapes, the volume of the particles can be calculated. 6A-6G illustrate various particle shapes that can be evaluated using the methods and processes of the present invention. FIG. 6A shows a top view 40, a side view 41, and a front view 42 of a cylindrical particle that typically has rounded edges. FIG. 6B shows a top view 43, a side view 44, and a front view 45 of an ellipsoidal particle having two equal minor axes and one longer axis. FIG. 6C shows a top view 46, a side view 47, and a front view 48 of the round disc particle. FIG. 6D shows a top view 49, a side view 50, and a front view 51 of the hexagonal disk particles. FIG. 6E shows a top view 52, a side view 53, and a front view 54 of the round oval particle. FIG. 6F shows a top view 55, a side view 56, and a front view 57 of the spherical particles. FIG. 6G shows a top view 58, a side view 59, and a front view 60 of a particle that is somewhat irregular but has substantial symmetry due to two mirror planes and one axis of rotation (C _2v symmetry group). Show.

  Some of the shapes shown in FIGS. 6A-6G are particularly important in applying the method and apparatus of the present invention to agriculture and nutrition. The granules extruded through a circular die have a cylindrical shape as shown in FIG. 6A. Granules and some crop seeds made using bread granulation have an ellipsoid shape as shown in FIG. 6B. Some dietary supplement tablets have a round disc shape as shown in FIG. 6C. Some candies have a round oval shape as shown in FIG. 6E. Granules made using fluid bed granulation and some crop seeds such as soybeans have a spherical shape as shown in FIG. 6F. Corn seeds have the shape shown in FIG. 6G.

In the cylindrical shape shown in FIG. 6A, the ellipsoidal shape shown in FIG. 6B, and the spherical shape shown in FIG. 6F, the hidden (thickness) dimension is assumed to be the same as the width of the top view particle image. be able to. The volume can be calculated using standard mathematical formulas. For example, the cylindrical volume V of FIG.
Formula 2
V (πd ² l) / 4
Given by. In the formula, d is the particle diameter (particle diameter estimated from the width of the particle image), and l is the particle length (particle length obtained from the length of the particle image). Specific formulas for calculating volumes of various shapes are well known in the geometric art. More generally, if the cross-sectional area of a particle can be specified as a function f (x) (where x specifies a distance along an axis perpendicular to the cross-section), the volume V is the particle's It can be obtained by analytical integration or numerical integration along the axis from one end to the other end.

  For the shapes shown in FIGS. 6C, 6D, 6E, and 6G, the thickness cannot be inferred from the top view particle images. However, if the thicknesses of the shapes shown in FIGS. 6C, 6D, and 6E are known in other ways (eg, if the thickness is uniform between particles), the top view with relatively high accuracy. The particle volume can be easily calculated from the image. In some examples, geometrically regular shapes such as cylinders, spheres, ellipsoids, and disks provide the maximum dimensional accuracy measured using the method and apparatus of the present invention. As shape regularity or predictability decreases, so does the particle size measurement accuracy. Even for some somewhat irregular shapes, such as the corn seed shown in FIG. 6G, the shape regularity is sufficiently consistent that approximate volume calculations can be made with some knowledge of the thickness dimension. If thickness dimensional information is known or can be inferred from top view particle images, the volume can be calculated with higher or lower accuracy for almost all shape types. However, in many applications of the method and apparatus of the present invention (for example, an application that determines the proportion of red candy in a mixture of candy having different colors or a ratio of weed seeds compared to crop seeds), It would be the only calculation desired to calculate the number fraction based on the counts of particles of different particle types.

  If the particle weight of a particle type is a constant known value, the weight fraction can be calculated using the known particle weight after determining the number of particles of each particle type. If the particle weight of the particle type is not constant, it is necessary to determine the weight of the particles to calculate the weight fraction. The weight of the particles can be determined by multiplying the envelope density of the particles by their volume. The envelope density can be measured by immersing a measured weight of a particular particle type in a liquid in which the particles are not soluble. Dividing the weight of the particles by the volume of displaced liquid gives the envelope density.

  It may be desirable to determine the total weight percent of chemical components in a mixture of particle types having different amounts of chemical components. For example, an agrochemical product may comprise a mixture of particles of two or more particle types, each particle type containing a different agriculturally active substance (ie, active ingredient). If the active ingredient itself does not provide sufficient optical property differences to easily identify the particle type, other components can be added to provide distinct chemistry and / or shape of the particles to provide different chemistry. It can be used as a marker to identify the composition, for example, particle types with different active ingredients. Typically, the percentage of an ingredient (eg, active ingredient) in a particle type is available prior to mixing from the amount of ingredient used to make the particle or from a chemical analysis of the made particle. With this information, the total percentage of a particular component in the mixture can be easily calculated.

  Without further effort, those skilled in the art will be able to utilize the present invention to its fullest extent using the above description. Accordingly, the following examples are considered merely illustrative and are not intended to limit the present disclosure in any way. The various calculation options presented in the following examples are merely presented for illustration.

Example 1
Exemplary Apparatus and its Operation The following example utilizes the apparatus and method of the present invention described in this first example. Referring to FIG. 1 as a schematic, the device consists of the following parts:

  The hopper 1 that provides the supply of particles 3 to the particle feeder 2 is a stainless steel funnel (Cole-Parmer Model SR-07265-00 (Cole-Parmer Model SR-) with an upper end diameter of about 5 inches (12.7 cm). 07265-00)), which is narrowed from the upper end to the opening of 0.5 inches (1.3 cm) or more from the lower end. The particle feeder 2 is a Syntron (registered trademark) (model FL-TO-C) light capacity electromagnetic vibrating feeder (FM Corporation) (Syntron (registered trademark) (Model FL-TO-C). Light Capacity Electromagnetic Vibrating Feeder (FMC Corp.), supported in parallel by a table or bench top (not shown) on which the device is seated. The funnel is located at the back end of the vibratory feeder and is fixed so that the funnel outlet is 3/8 to 5/8 inch (1.0 to 1.6 cm) above the feeder.

  The inclined surface 4 is a flat, non-reflective glass piece having dimensions of 2.9 inches (7.4 cm) wide and 4.0 inches (10 cm) long. The back side of the glass is coated with a matte black paint. The glass piece is cushioned and seated on a piece (not shown) of a polyethylene foam sheet having the same dimensions as the glass and having a thickness of 1/8 inch (0.3 cm). Glass and foam are supported at an inclination angle α of 45 ° with the horizon. The front edge (that is, the outlet end) of the particle feeder 2 is disposed at a position of 1 mm or less upward from the upper end (that is, the upper inlet end) of the inclined surface 4. The glass of the particle feeder 2 and the inclined surface 4 is not in contact. Feeder frequency (that is, feed speed) is controlled by an FMC power controller (model CC-1A) (FMC Power Controller (Model CC-1A)) as the feed controller 11.

  The image receiver 6 is a video camera (Prunix Black and White Progressive Scan Analog Model 9701), whose zoom lens is approximately from the inclined surface 4 It is located at 9-10 inches (23-25 cm) and points in a direction perpendicular thereto. The zoom lens of the camera is equivalent to Cole Palmer model # P-48901-04 (Cole-Parmer model # P-48901-04). The camera shutter speed is 1/8000 sec and the camera resolution is 640 × 480 pixels.

  W / A / C Long Life 10,000 hours #EXN 50W MR16 floodlight (W / A / C Long Life 10,000 hours #EXN 50W MR16 floodlights) (11530 Garden City, New York) Two tungsten-halogen lamps equivalent to the W.A.C Lighting Company (available from W.A.C Lighting Co., 615 South Street, Garden City, New York 11530) as the illumination source 5 Has been placed. Lamps powered by Astec LPS 253, 12V DC, 20 amp power supply (not shown) (not shown) were directed to illuminate the inclined surface 4 uniformly. It is housed in the W • A • C Lighting # LP007 Black Surface (W • A • C Lighting # LP007 Black Surface) which is a low voltage mount.

  The air stream from the nozzle 10 is used to sweep the inclined surface 4 to remove dust. The nozzle 10 comprises a slit having a dimension of 1.75 inches × 0.02 inches (4.44 cm × 0.05 cm) and arranged with a long dimension parallel to the inclined surface 4. The nozzle 10 is arranged downward above the upper end of the inclined surface 4 and at an angle of 50 ° with the horizontal so that the air stream sweeps the inclined surface 4 at an oblique angle. The air supply to the nozzle 10 is 5-10 psi (34.5-68.9 kPa), which is 0-30 psi (0-207 kPa) pressure connected to a 70 psig (483 kPa) air source (not shown). Provided by SMC NAR 2000-N01, a regulator (not shown).

  An appropriately sized receiver is used for the catch tray 7 that collects the particles as they descend below the surface.

  A computer equipped with an Intel Pentium II 200 megahertz processor (Intel Pentium II 200-megahertz processor) and Microsoft Windows 95 operating system is used as the composition calculator 8. National Instruments' LabView IMAQ software is used to process the camera images. A National Instruments “Frame Grabber” board # 1408 (National Instrument's “Framegrabber” Board # 1408) is installed in the computer to digitize analog images. Software, including Microsoft Excel, is used to create reports comprising tables and / or graphs from data obtained from camera image processing. The output device 9 includes both a computer monitor and a printer.

  The system is calibrated as follows. A piece of white paper is placed flat against the glass plate and the light is adjusted so that the light is evenly distributed on the paper. Next, place another paper piece with a precisely known black circle with a diameter of about 10 mm on the white paper piece, and after fully adjusting the camera focus, determine the number of pixel sides per millimeter. Determine and determine the calibration factor in pixels / mm for use in subsequent sample analysis.

  A 20-40 g particle sample is placed in the funnel, the air supply is started, the glass surface is swept, and the camera zoom lens is adjusted to capture an image in the field of view through which the particles pass. The vibratory feeder is activated and the feed rate is set so that 5-10 particles are captured per image. This corresponds to a feed rate of about 100-150 particles per second. Capture approximately 2-3 images per second and continue analysis until all particles placed in the funnel have passed through the field of view. The captured image is analyzed as described in the examples below.

Example 2
Analysis of multiple samples of particle mixture Pinnacle (R) 25DF (Pinnacle (R) 25DF) herbicide paste extruded particles (DuPont, 126.4g) containing 25 wt% thifensulfuron-methyl as active ingredient ) And 75% by weight of sulfentrazone as the active ingredient Authority® 75DF (Authenticity® 75DF) herbicide paste extruded particles (FMC Corporation, 126.3 g). Used to make a mixture. Both particle types were shaped into columns by setting the diameter to about 1 mm and changing the length in the range of about 1 mm to 5 mm. The color of Pinnacle (R) 25DF (Pinnacle (R) 25DF) was light tan, and the measured envelope density was 1.85 g / cm < ³ >. The color of Authority (registered trademark) 75DF (Authority (registered trademark) 75DF) was medium brown, and the measured envelope density was 1.43 g / cm ³ . The particles were tumbled together to produce a nearly uniform mixture (although uniformity was not a specific goal of this experiment). Next, 252.7 g of the mixture was divided into aliquots to provide 10 samples having weights ranging from 17.31 to 27.60 g.

  Ten samples were analyzed under visible light using the method of the present invention. The appearance of the top view image on the black slanted surface was similar to that illustrated by FIG. Using the method described for FIG. 4 utilizing the average grayscale value of the 25 pixel kernel surrounding each particle's centroid and a threshold of 190 based on the 16-bit grayscale, the particles are colored (ie, Pinnacle®). (Pinnacle (R))) or dark (i.e. Authority (R)).

The particle image width was determined by the method described for FIG. 5 using computer software. The area of the top view particle image was determined using LabView IMAQ software. The particle length was calculated by dividing the particle image area by the particle image width. Since the particles are cylindrical, the particle volume was calculated according to Equation 2, assuming that the particle image width accurately represents the particle diameter. This allowed the volume percent to be calculated. Equation 4 shows a formula for calculating the volume percent VP ^L of light colored particles in a mixture of n ^L light colored particles and n ^D dark colored particles. In the formula, ΣV ^L _i is the sum of the volumes of the light colored particles, and ΣV ^D _j is the sum of the volumes of the dark colored particles.

  The volume percentages of light colored particles (ie Pinnacle® (Pincle®)) and dark colored particles (ie Authority®) were analyzed based on particle length. For example, Table 1 lists the percent volume for particles of various length ranges in Sample 2. In sample 2, about 1690 light colored particle images and 3110 dark colored particle images were analyzed.

  Table 1 shows that the particles of Pinnacle® (Pinnacle®) and Authority® (Authority®) had similar length distributions. This distribution is desirable to make a uniform mixture of particles.

Weight percent was calculated based on volume percent and envelope density. Since the envelope density of the light colored particles and dark colored particles are 1.85 and 1.43 g / cm ³ , respectively, from equation 5, it contains light colored particles of volume percent VP ^L and dark colored particles of volume percent VP ^D. wt% WP ^L of bright colored particles in a sample can be obtained.

Formula 5
WP ^L = 100% · (1.85 · VP ^L ) / (1.85 · VP ^L + 1.43 · VP ^D )

  Summary data for all 10 samples is listed in Table 2.

  The bottom of columns F and I in Table 2 is the sum of the weights of light and dark particles calculated for all samples. These sums are essentially equal to the initial measured weight of the individual components. This result shows the remarkable potential of this analytical method. Applicants have been able to use this method to break down the composition of a well-mixed particle mixture into its original components.

Example 3
Analysis of Particle Mixture Samples with Known Composition Ranges As shown in Table 3, a measured weight of Light Colored Classic® 25DF (Classic) containing 25% by weight of chlorimuron-ethyl as the active ingredient (ai) (R) 25DF) Herbicide Paste Extruded Particles (DuPont) with a measured weight of dark colored Authority® 75DF (Authority®) containing 75% by weight of sulfentrazone as active ingredient 75DF) Herbicide paste extruded particles (FMC Corp.) were mixed to produce 7 particle mixture samples of approximately 20 g each. Columns B and C list the amount of each particle type. Columns D and F show the corresponding percentage of each particle type in the sample, and columns E and G represent the active ingredients chlorimuron-ethyl (CE) and sulfene in the sample calculated based on their known composition. Shows the percentage of trazones.

  Each sample was analyzed using the method of the present invention. The sample was passed through the apparatus more than once to image 4000 particles. Data was processed in the same manner as described in Example 2 above. Next, an estimate of the percentage of each active ingredient in each sample was made based on the volume percent of each particle type, their measured envelope density, and the percentage of known active ingredient contained in each. The results are shown in Table 4.

Columns A and E are Classic® 25DF (Classic® 25DF) particles and Authority® 75DF (Authority® 75DF) contained in each mixed sample determined by the method of the present invention. It lists the volume percent of particles. Rows B and F show the envelope density of Classic® 25DF (Classic® 25DF) of 1.43 g / cm ³ and Authority® 75DF (Authority® 75DF) of 1.53 g. Table 1 lists the weight percent of the formulation calculated based on / cm ³ . Columns C and G show the measured weight percent (measured weight%) of the chlorimuron-ethyl active ingredient and sulfentrazone active ingredient calculated from the calculated weight percentage of the formulation and the known concentration of the active ingredient in each particle type, respectively. It is enumerated. These measurements were taken from the Classic® 25DF (Classic® 25DF) and Authority® 75DF (Authority® 75DF) used to make the samples as shown in columns D and H. ) And the feed weight percent calculated based on the weight of (). From the comparison of columns C and D with columns G and H, Classic® 25DF (Classic® 25DF) and Authority® 75DF (Authority® 75DF) used to make the samples were used. In contrast to the corresponding value calculated from the weight of chlorimuron-ethyl, the measured weight percent of chlorimuron-ethyl is close to that value, but consistently slightly less than that, and the measured weight percent of sulfentrazone is It was shown to be close to the value but consistently slightly larger than that.

Example 4
Analysis of a mixture of seven different colored particle types Colored coated candy packages were purchased. A sample consisting of 51 candies was thoroughly mixed. This sample contained uniform particles of round oval shape (similar to that shown in FIG. 6E) having a diameter of about 12 mm and a thickness of about 5 mm at the maximum thickness. Each particle had a distinct color. Seven different colors were present in the mixture. In this example, the color-based particle distribution was analyzed by the method of the present invention. The apparatus described in Example 1 was used except that a color Sony DFW-X700 Digital Video Camera was used as the image receiver 6 (in FIG. 1) instead of the monochrome camera. . The sample was fed to the inclined surface 4 of the device (FIG. 1), and when all 51 particles in the sample had passed down the inclined surface 4, the same sample was again fed. This was repeated several times until 41 particle images were collected. These particle images were digitized and the information from each pixel was fed simultaneously into two independently configured electronic buffers for storing the color camera output. By configuring the first buffer to store the total intensity of light on each pixel, the black background of the inclined surface can be identified from the falling particle profile to determine the area in pixels of each particle image. I made it. In addition, this data was evaluated to determine the perimeter of the particle image, from which the particle diameter was directly calculated. The second buffer was configured to store the intensity of light of red, green, and blue wavelengths. Next, the intensity (excitation level) of each of the 25 pixels surrounding the centroid of each particle image is evaluated on a scale of 0 to 255 units of red (R), green (G), and blue (B) light. The average red, green, and blue intensities in the 25 pixel area were then calculated for each particle. Color standard calibration values were defined by selecting particles representing each particle type of seven different colors. The representative particles were kept fixed during the color measurement. The results are listed in Table 5.

  An example of the algorithm used to determine the color of each pixel in the particle image is shown in Equation 6 for the yellow standard.

Equation 6
| R _measurement- R _{calculation (y)} | + | G _measurement- G _{calculation (y)} | + | B _measurement- B _{calculation (y)} | = yellow color match deviation equation 6, R _measurement , G _measurement , and B _measurement Is the measured intensity of red, green and blue light for each particle. Substituting the red, green, and blue intensity values listed in Table 5 for the yellow standard into the yellow calibration value R _{calculation (y)} , G _{calculation (y)} , and B _{calculation (y)} , Equation 6a is obtained. It is done.

Equation 6a
| R _measurement- 255 | + | G _measurement- 255 | + | B _measurement- 129 | = Yellow color match deviation In equations 6 and 6a, the yellow color match deviation is the red, green, blue measured intensity and the yellow standard It is the sum of absolute deviations from the corresponding calibration strength.

  For each particle image, the process was repeated for all seven colors, using an equation similar to Equations 6 and 6a. For example, if the particles are yellow, the measured excitation value should be close to the yellow standard values listed in Table 5 and the sum of the absolute values of the differences shown in Equations 6 and 6a must be small. (Actually about 0). Next, when replaced with excitation levels for other color standards, the calculated color match deviation must be greater than the yellow color match deviation.

  Table 6 shows the color match deviation calculated from Equations 6 and 6a and their similar equations for each of the 41 captured particle images. Based on the minimum color match deviation value shown in bold, the data was classified by the most likely color of the particles (shown in the last column).

  In a 41 particle image collected from a sample of 51 particles, some statistical errors due to sampling are expected. For comparison, Table 7 shows the actual particle color distribution and the particle color distribution measured using the method and apparatus of the present invention.

  Columns B and D in Table 7 show the actual number of each color of the original sample versus the number measured. Columns C and E provide a more direct comparison based on the percentage of particles of each color in the sample versus measurement. In view of the small number of particles in the analyzed samples, the comparison is remarkably similar.

  In determining the particle diameter based on the particle image, an average of 11.7 mm was obtained compared to 12.7-14 mm by direct measurement using a caliper. Again, there was good agreement considering the preliminary nature of the experiment.

Example 5
Analysis of particle mixtures in which a distinctive color is assigned to one particle type. In many cases, different particle types have similar gray scale values when viewed under visible light, so the previous Examples 2 and 3 Cannot be identified with a black and white video camera such as Dye can be incorporated into the composition of one of the particle types to provide different spectral properties. In this example, about 1% by weight of FD & C Blue Dye # 1 (CAS No: 2650-18-2) is incorporated into the Kaolin clay based paste extruded particle formulation. The resulting particles are blue in color and reflect 450-500 nm light. In other kaolin clay based extruded particle formulations, no dye is added so that the color of the particles is kept light tan.

  The apparatus described in Example 1 is used except that the image receiver 6 (in FIG. 1) is a color video camera. A 40 g sample is made by thoroughly mixing 20 g tan colored particles and 20 g blue colored particles. Each particle type is shaped into a cylinder so as to have a diameter of about 1 mm and a particle length in the range of 1-4 mm. A mixture of light tan and blue particles is analyzed by the method of the present invention. The image is digitized and the information from each pixel is fed into two independently configured buffers for storing the color camera output. By configuring the first buffer to store the total intensity of light on each pixel, the black background of the inclined surface can be identified from the falling particle profile to determine the area in pixels of each particle image. To. Further, this data is processed according to the method described in Example 2 to determine the width, length, and volume of the particles. The second buffer is configured to simultaneously store the intensities of red, green, and blue wavelengths of light. Next, the intensity of the blue light received by each of the 25 pixels surrounding the centroid of each particle image is examined and an average is calculated for each particle. Using this information, the imaged particles are identified as belonging to the “blue” particle type or “non-blue” particle type by the general method described in Example 4. The calculated particle volume and measured envelope density are then used to calculate the volume percent and weight percent of each particle type in the mixture by the general method described in Example 2.

Example 6
Analysis of particle mixtures where the particle type is identified using UV light If the particle formulation contains a dark colored component such as a lignin sulfonate surfactant, the optical properties can be added even if the colored component is added to one formulation. Cannot be sufficiently distinguished from the other component. When mixing two dark colored particle types together, another way to identify the particle type is to combine them with different levels of components that fluoresce under ultraviolet light, such as fluorescein. In this example, kaolin-based paste extruded particles containing lignin sulfonate are compounded to contain 0.5% by weight of fluorescein. Other kaolin-based paste extruded particles containing lignin sulfonate are compounded to contain 1% by weight of fluorescein. If fluorescence is used during image formation, ambient visible light is best excluded, so the apparatus described in Example 1 with reference to FIG. 1 is modified as shown in FIG. The elements of the apparatus shown in FIG. 7 are as described in Example 1 with respect to FIG. 1 but have the differences shown below. The chamber 12 surrounds the elements of the apparatus including the inclined surface 4 and the image receiver 6 to prevent them from entering ambient visible light. By providing an opening in the chamber 12, the particles 3 to be evaluated can be introduced into the hopper 1, the analyzed sample can be taken out by the catch tray 7, and the illumination source 5 a irradiates the falling particles 3 to receive the image receiver. 6 makes it possible to record the image of the falling particles 3. The illumination source 5a includes an ultraviolet light source such as a mercury vapor discharge lamp. The image receiver 6 includes a black and white camera.

  Fluorescein emits orange colored light when illuminated with ultraviolet light. By setting the fluorescein content in each particle type to a different level, different levels of luminescence are captured by the black and white camera. Particle images showing higher levels of luminescence correspond to particles containing 1% by weight fluorescein, and particle images showing lower levels of luminescence correspond to particles containing 0.5% by weight fluorescein To do. The data collected for the sample can then be analyzed based on the general method described in Example 2.

Example 7
Quality analysis of rice products using visible light Rice foods may contain processed rice grains from different sources. These combinations produce a flavor and aroma blend that will please customers. In this example, a rice product is produced that contains two processed rice types: white colored long grain rice and dark colored wild rice seed. The product is a mixture of two types and the composition targets a ratio of 75 weight percent white rice to 25 weight percent wild rice. A quality assurance system is devised that includes randomly extracting a package of rice products from a production line and analyzing the ratio of white rice to wild rice using the present invention. The envelope density of the two rice types was measured and found to be essentially the same. Using the general method described in Example 2, the total amount of each 8 ounce (227 g) sample is evaluated. For analysis of this type of product, a relative accuracy of ± 10% is sufficient. Therefore, it is possible to process the shape of the rice grain as a cylinder, but it would be easy to use an alternative algorithm that regards the shape of the grain as an ellipsoid. The entire contents of the product package are supplied to the apparatus described in Embodiment 1 with respect to FIG. According to the general method described in Example 2, the volume of each imaged kernel is calculated from the corresponding particle image. Each image is also processed to identify the corresponding grain as a member of the white rice type or wild rice type so that the volume percent of each type can be calculated. Since the envelope densities of the two rice types are equal, the weight percent for both components is equal to the volume percent.

Example 8
Analysis of a mixture of particles having two different shapes This example shows how a mixture of cylindrical and ellipsoidal particles can be analytically separated using the method of the present invention.

  Three samples were made by mixing the cylindrical particles made by the paste extrusion method with the nearly ellipsoidal particles made by the pan granulation method. Three mixtures were made as shown in Table 8.

  Paste extruded particles were cylinders with a diameter of about 1 mm and a length in the range of 0.5 mm to 5 mm. The bread granulated particles were shaped in the form of almost ellipsoidal pellets having a length that is longer than 50% of the overall width. The size variation of these ellipsoidal particles generally included proportional elongation in both length and width.

  Samples were analyzed using the apparatus described in Example 1. Individual samples of 912, 472, and 551 were obtained for samples 1, 2, and 3, respectively, by repeatedly passing each sample through the instrument and accumulating data. The data was analyzed with the intention of separating the particle images and their dimensional parameters into cylindrical and ellipsoidal particle groups.

  The length, width, perimeter and area of the particles were determined using LabView IMAQ software and the general method described in Example 2. As shown in FIG. 8, the area / circumference ratio (A / P) was plotted against the length / width ratio (L / W) for each particle image. Based on the clustering of visually distinct data points, a diagonal line 70 (having a slope of 0.6417 and a y-axis intercept of 5.3) is drawn to create a cluster 71 of ellipsoidal particle data points and cylindrical particle data points. The cluster 72 was separated. This method allowed each of the three sample data sets to be divided into two subsets, one containing ellipsoidal data in one subset and cylindrical data in the other. From this particle image analysis, Sample 1 has 817 possible cylindrical particles and 95 possible ellipsoidal particles, Sample 2 has 329 possible cylindrical particles and 143 possible Suggested 3 ellipsoidal particles, sample 3 showed 337 possible columnar particles and 214 possible ellipsoidal particles.

To determine the weight percent of each particle type in the mixture, the total volume of each type of particle was first calculated. In order to calculate the particle length required to calculate the L / W ratio plotted in FIG. 8, the particle image area was simply divided by the particle image length. This method of calculating length is reasonably accurate in calculating the volume of cylindrical particles that give a rectangular image. However, when calculating the volume of an ellipsoidal particle that gives an elliptical image, it is described by Equation 7 where _Le is the length, A is the image area, and W is the image width. The exact method used was used.

Equation 7
L _e = (4 · A) / (π · W)

Next, the individual volume of each particle was calculated. For cylindrical particles, the volume was calculated according to Equation 8 (where V _c is the volume, W is the image width, and L _c is the length).

Equation 8
V _c = π · (W / 2) ² · L _c

For ellipsoidal particles, the volume was calculated according to Equation 9 (where V _e is the volume, W is the image width, and _Le is the length).

Equation 9
V _e = (4/3) · π · (W / 2) ² · (L _e / 2)

The volume of the cylindrical and ellipsoidal particles was summed separately and then the volume fraction of each particle type was calculated. The total weight of each particle type was calculated by multiplying the total volume by the envelope density. Envelope density is the paste extrusion cylinder 1.33 g / cm ^3, a pan granulation ellipsoid was 1.21 g / cm ^3. The total weight was used to calculate the weight fraction of each particle type in each sample. Table 9 lists the weight fractions measured according to the present invention along with the actual weight fractions for comparison.

  The results obtained from the method and apparatus of the present invention are fairly close to the actual sample composition, taking into account the small sample sizes and empirical methods used to distinguish between cylindrical and ellipsoidal particles Indicated by. The most accurate results were obtained with sample 1, which had the largest sample size or smallest statistical sampling error. In addition to statistical sampling errors, in the case of particles having an A / P ratio and an L / P ratio in the area where the clusters contact, an image of cylindrical particles is ellipsoidal using the A / P vs. L / P cluster method. Some errors may have been caused by the inability to completely distinguish from the particle-like image. Other image analysis, such as identifying a rectangular outline in a particle image as an elliptical outline, could identify the particle type more accurately. However, the results obtained from this example are sufficiently specific to show that the method of the present invention has considerable value in determining the composition of a particle mixture whose distinguishing characteristic is particle shape. .

1 shows an overall view of one embodiment of the apparatus of the present invention and its major components. The example of the image which can be image | photographed with the video camera of FIG. 1 is shown. The relationship between the inclined surface and the feeder and the “particle bounce” effect are illustrated. 3 illustrates a method for determining the optical properties of particles from image information. A method for calculating the area and the shortest distance of a particle cross section from image information is shown, and illustrates how systematic errors can occur when measuring area and width. Illustrates some typical particle shapes suitable for the method and apparatus of the present invention. Example 6 relates to Example 6 and shows an embodiment of the device modified to use ultraviolet light illumination. The graph relates to Example 8 and shows a graph in which particle image data is clustered into an ellipsoid set and a cylinder set.

Claims (23)

A method for determining the ratio of different particle types in a mixture comprising:
(I) non-aggregating particles of at least two particle types in a channel inclined at an angle sufficient to lower the particles along the channel, wherein each particle type is at least one different from the other particle types Providing a mixture comprising optical properties and / or shapes;
(Ii) illuminate the particles along the inclined path;
(Iii) collecting a reflected light image of the illuminated particles;
(Iv) A method comprising calculating a ratio of at least one particle type based on data of a reflected light image exhibiting at least one different optical property and / or shape.   The method of claim 1, wherein the at least one optical property is at least one of reflectance, luminescence, and variations thereof at visible, ultraviolet, and infrared wavelengths.   The method of claim 2, further comprising calculating at least one dimensional characteristic.   The method of claim 3, wherein the at least one dimensional characteristic is at least one of a longest dimension, a shortest dimension, an area, and a circumference.   The method of claim 1, wherein the inclined passage makes an angle of less than about 60 ° with the horizontal.   The method of claim 1, wherein the reflected light image is collected in a direction substantially perpendicular to the inclined path.   Feeding comprises positioning the particles in the vicinity of the outlet end of the feeder where the particles are fed into the inclined passage, and the inclined passage has an inlet end located below and adjacent to the feeding outlet end. The method of claim 1 comprising.   The method of claim 1, wherein less than 25% of the particles on the inclined path are in contact with other particles in the reflected light image.   9. The method of claim 8, wherein less than 10% of the particles on the inclined path are in contact with other particles in the reflected light image.   10. The method of claim 9, wherein less than 2% of the particles on the inclined path are in contact with other particles in the reflected light image.   4. The method of claim 3, wherein the non-aggregating particles are non-spherical and at least 80% of the particles have a bounce angle of 10 degrees or less.   The method of claim 11, wherein at least 90% of the particles have a bounce angle of 10 degrees or less.   The method of claim 11, wherein at least 95% of the particles have a bounce angle of 10 degrees or less.   Feeding comprises positioning the particles in the vicinity of the outlet end of the feeder where the particles are fed into the inclined passage, and the inclined passage has an inlet end located below and adjacent to the feeding outlet end. 12. The method of claim 11, comprising comprising defining a gap between the outlet end and the inlet end.   15. A method according to claim 14, wherein the gap has a length that is less than or equal to the shortest dimension of the [measuring] particle.   The method of claim 1, wherein the mixture of non-aggregating particles comprises substantially cylindrically shaped particles.   The method of claim 1, wherein the mixture of non-aggregating particles comprises particles having a substantially cylindrical shape with a circular cross section.   The method of claim 1 wherein the mixture of non-aggregating particles comprises seeds.   The method of claim 1, wherein the mixture of non-aggregating particles comprises particles comprising at least one agriculturally active substance.   The method of claim 1, wherein the mixture of non-aggregating particles comprises particles comprising at least one crop protection agent.   The method of claim 1 wherein the inert gas is blown across the surface of the inclined passage. An apparatus for determining a ratio of particles of different particle types in a mixture,
(I) a particle feeder having an exit end;
(Ii) an inclined passage having an upper inlet end located adjacent to and below the outlet end of the feeder so that non-aggregating particles can descend down the inclined passage;
(Iii) an illumination source oriented with respect to the inclined passage so that the particle can be illuminated upward as the particle descends down the inclined passage;
(Iv) an image receiver oriented with respect to the inclined passage so that a reflected light image of the particle can be collected as the particle descends down the inclined passage;
(V) a composition calculator for converting the reflected light image signal received from the image receiver into data indicative of at least one ratio of particle types in the mixture based on at least one optical property and / or shape of the particles; A device comprising: The apparatus of claim 22, wherein the image receiver comprises a color camera.

Images