JP4908926B2 - Granule sorter - Google Patents

Description

  The present invention relates to a granular material sorter that discriminates and sorts the granular material of different kinds of plant seeds mixed in the granular material as a raw material.

  Conventionally, the granular material of the raw material transported from the harvesting land, for example, the raw material wheat, contains impurities mixed at the time of harvesting or transporting. The impurities are grass, stones, seeds and stems of different plants, and these impurities are removed prior to the wheat processing step. As a method for removing impurities, there are a method for sorting by size difference, a method for sorting by difference in specific gravity, a method for sorting by shape difference, a method for sorting by color difference, etc. A suitable method is used accordingly. As a sorter corresponding to various methods, for example, there are a shape sorter such as a sieving machine, a specific gravity sorter, and a diverter, a color sorter, and the like.

  Here, among the foreign substances, there are seeds of different kinds of plants that are similar in size and specific gravity to wheat, and it may be difficult to perform sufficient selection using only a sieving machine or a specific gravity sorter. . In addition, wheat and seeds of different plants may be crushed during transportation, and in this case, it becomes difficult to select seeds of different plants due to the difference in shape. In particular, dicotyledonous plant seeds such as peanuts, peanuts, wisteria beans, azuki beans, kidney beans, peas, peppercorns, broad beans, mungbeans, cowpeas, chickpeas, lentils, kintokimame, quail beans, tiger beans, purple beans, soybeans, coffee beans, etc. It is easy to break into two or more, and the original size is a grain that is larger than wheat so that it can be separated from wheat due to the difference in size. Therefore, it may be difficult to separate from wheat due to the difference in size. Therefore, in order to select wheat and different kinds of plant seeds with high accuracy, a color sorter that sorts granular materials based on the difference in color is effective.

As the color sorter, in Patent Document 1, the granular material of the raw material is a non-defective product, the granular material mixed in the granular material of the raw material is a defective product, and a spherical threshold value and a planar threshold value are used to select from among the good products A granular color sorter for sorting defective products is described.
JP 2005-230703 A

  By the way, when sorting the granular material other than the raw material and the granular material other than the raw material by color, it is possible to select with sufficient accuracy only by setting the data indicating the color of the granular material as a spherical or planar threshold value. It can be difficult to do. For example, in the granular material color sorter described in Patent Document 1 described above, the granular material which is a raw material that is substantially spherical in the uniform color space of the L * a * b color system or the L * u * v color system. Sorting is performed by setting a color as a threshold value and excluding a portion indicating the color of a granular material other than the raw material included in the spherical threshold value with a planar threshold value. In this case, it is difficult to select a granular material other than the raw material that exhibits a color very similar to the granular material of the raw material from the granular material of the raw material. In some cases, the granular material other than the raw material partially shows substantially the same color as the granular material of the raw material. In such a case, it is difficult to select the granular material other than the raw material.

  The subject of this invention is providing the granular material sorter which can discriminate | determine and classify | classify granular materials other than the raw material which shows the color similar to the granular material of a raw material with high precision.

The granular material sorting machine according to claim 1 is disposed at a position opposite to the granular material transferred by the conveying device and the granular material transferred by the transfer device, and irradiates the granular material to be transferred with light, Using a photographing device for photographing an image of a granular material, a reference table storing granular material color information indicating a color specific to a predetermined granular material using hue and saturation as parameters, and an image photographed in the photographing device, Conversion means for converting the color indicated by the pixels constituting the granular material into discrimination information using hue and saturation as parameters, and the discrimination information of each pixel constituting the granular material is stored in the reference table. The granular material color information is used to determine whether or not the color of the predetermined granular material is indicated, and the color of the predetermined granular material is indicated out of the number of pixels constituting the pixel area of the granular material. Discrimination information determined to be If the number of the predetermined ratio or more, characterized by comprising a determination means for granulate is determined to the a predetermined granules, and an output unit for outputting a discrimination result by the discrimination means.

The granular material sorter according to claim 1 stores granular material color information indicating the color of a predetermined granular material using the hue and saturation as parameters, and the hue and saturation as parameters of the color of the granular material being transferred. And the stored granular material color information are used to determine whether or not the transferred granular material is a predetermined granular material, and the determination result is output. That is, by using the hue and saturation that accurately indicate the color characteristics of the granular material as parameters, the granular color information indicating the specific color of the granular material is stored, and the stored granular color information and discrimination information are stored. Are compared with each other. Therefore, for example, a granular material other than the predetermined granular material mixed in the predetermined granular material can be discriminated with high accuracy by the color difference.
Further, for each discrimination information of each pixel constituting the pixel area of the granular material, it is determined whether or not the color of the predetermined granular material is indicated, and the determination information determined to indicate the color of the predetermined granular material Is equal to or greater than a predetermined ratio of the number of pixels constituting the pixel area of the granular material, it is determined that the granular material is the predetermined granular material. That is, by discriminating on the basis of the ratio of the number of pixels indicating the color specific to the predetermined granular material, even if the granular material has a color partially similar to the predetermined granular material, the color difference can be obtained with high accuracy. It is possible to discriminate and sort the granular material with high accuracy.

The granular material sorting machine according to claim 2 is arranged at a position opposite to the transfer device for transferring the granular material and the granular material transferred by the transfer device, and irradiates the transferred granular material with light. An imaging device that captures an image of the granular material, a reference table that stores granular material color information indicating colors specific to a predetermined granular material using hue and saturation as parameters, and an image captured by the imaging device. Conversion means for converting the color indicated by the pixels constituting the granular material into discrimination information using hue and saturation as parameters, and the discrimination information of each pixel constituting the granular material is stored in the reference table. The granular material color information is used to determine whether the color of the predetermined granular material is indicated, and the color of the predetermined granular material is selected from the number of pixels constituting the pixel area of the granular material. Determined to show If the number of different information is a predetermined ratio or more, excluding the discriminating means granulate is determined to the a predetermined granules, the granules is determined that the non-predetermined particulates by said discrimination means And a sorting means.

The granular material sorter according to claim 2 stores the granular material color information indicating the color of a predetermined granular material using the hue and saturation as parameters, and the hue and saturation as parameters of the color of the granular material being transferred. Using the discriminating information shown as above and the stored granule color information, it is discriminated whether or not the transferred granular material is a predetermined granular material, and the granular material determined not to be the predetermined granular material Are excluded and sorted. That is, by using the hue and saturation that accurately indicate the color characteristics of the granular material as parameters, the granular color information indicating the specific color of the granular material is stored, and the stored granular color information and discrimination information are stored. Are compared with each other to discriminate the particulate matter, and the particulate matter other than the predetermined particulate matter is excluded. Therefore, it is possible to discriminate the color difference with high accuracy and perform high-accuracy sorting, and since the color of the granular material is indicated by two parameters of hue and saturation, the color of the granular material is discriminated. The amount of calculation at the time of conversion to can be reduced, and a quick sorting process can be performed.
Further, for each discrimination information of each pixel constituting the pixel area of the granular material, it is determined whether or not the color of the predetermined granular material is indicated, and the determination information determined to indicate the color of the predetermined granular material Is equal to or greater than a predetermined ratio of the number of pixels constituting the pixel area of the granular material, it is determined that the granular material is the predetermined granular material. That is, by discriminating on the basis of the ratio of the number of pixels indicating the color specific to the predetermined granular material, even if the granular material has a color partially similar to the predetermined granular material, the color difference can be obtained with high accuracy. It is possible to discriminate and sort the granular material with high accuracy.

  The granular material sorting machine according to claim 3, wherein the sorting means excludes the granular material that is determined not to be the predetermined granular material from the falling trajectory of the granular material discharged from the terminal portion of the transfer device. It is characterized by that. According to the particulate matter sorting machine according to the third aspect, since the particulate matter other than the predetermined particulate matter is excluded from the falling trajectory of the particulate matter, the particulate matter is quickly sorted in the process of transferring the particulate matter. be able to.

  According to a fourth aspect of the present invention, there is provided the particulate matter sorter, wherein the photographing device shoots an image of the particulate matter irradiated with light by the irradiating means for irradiating the particulate matter to be transported with the light. And an element.

  According to the particulate matter sorting machine according to claim 4, since the particulate matter is irradiated with light to photograph the particulate matter image, it is possible to photograph a clear image appropriately showing the color of the particulate matter. it can. Accordingly, it is possible to acquire information appropriately indicating the color of the granular material from the photographed image, and to determine the granular material with high accuracy.

  Further, in the granular material sorter according to claim 5, in the image photographed by the photographing element, the conversion means determines the individual granular material from the difference between the color of the background material of the granular material and the color of the granular material. The pixel area is specified, and the color indicated by each pixel constituting the pixel area is converted into discrimination information.

  The granular material sorter according to claim 5 specifies the pixel region of each granular material from the difference between the color of the background member of the granular material and the color of the granular material, and the color indicated by each pixel constituting the pixel region. It is converted into discrimination information. That is, by using a color completely different from the color of the granular material as the color of the background member, the pixel area of the granular material can be appropriately specified. In addition, since the color of each pixel constituting the pixel area is converted into discrimination information, the color difference can be discriminated with higher accuracy compared to the case where the color of the granular material is shown in a predetermined color system. Can do.

  In the present invention, whether or not the transferred granular material is the predetermined granular material using the information indicating the color of the predetermined granular material using the hue and saturation that accurately indicate the color characteristics of the granular material as parameters. Is being determined. Therefore, it is possible to discriminate with high accuracy, for example, granular materials other than wheat mixed in wheat, which is a raw granular material, by performing discrimination using granular material color information and discrimination information that accurately shows the color unique to the granular material. can do.

  Further, in the present invention, the transferred granular material is discriminated using information indicating the color of the predetermined granular material using the hue and saturation that accurately indicate the color characteristics of the granular material as parameters, and the predetermined granularity is determined. Excludes particulate matter that is determined not to be. Therefore, even if the color of the raw material granular material, for example, the granular material other than wheat mixed in wheat is similar to that of wheat, the hue and saturation that accurately indicate the characteristics of the color are used as parameters. By using the information indicating the color, it is possible to determine whether or not the granular material is wheat with high accuracy, and to accurately exclude the granular material other than wheat with high accuracy.

  Hereinafter, a granular material sorter according to an embodiment of the present invention will be described with reference to the drawings. Drawing 1 is a longitudinal section showing the outline of granular material sorter 2 concerning an embodiment. The granular material sorter 2 includes an endless belt conveyor 4 which is a transfer device for transferring raw material granular material, and a raw material supply unit 6 is disposed on the transfer start end side of the endless belt conveyor 4. The endless belt conveyor 4 is driven in a direction indicated by an arrow in FIG. 1 at a speed of 64 cm / second. The raw material supply unit 6 includes a raw material hopper 8 and a vibration feeder 10. And the granular material supplied from this raw material supply part 6 falls on the endless belt conveyor 4, and is transferred in the direction shown by the arrow in FIG.

  Further, a color image of the granular material (hereinafter simply referred to as “image”) is obtained by irradiating the granular material to be transferred with illumination light at a position facing the surface (transfer surface) on which the granular material is placed and transferred on the endless belt conveyor 4. The photographing device 12 for photographing the above is disposed. The imaging device 12 includes an irradiation unit 14 that irradiates light to the granular material that is transferred by the endless belt conveyor 4, and a CCD camera 16 that images the granular material irradiated with light by the irradiation unit 14.

  Here, as the CCD camera 16, the granular material transferred by the endless belt conveyor 4 driven at 64 cm / second is a predetermined pixel unit, for example, the traveling direction of the endless belt conveyor 4 (direction indicated by the arrow in FIG. 1). Any camera can be used as long as it can shoot 32 times / sec in units of length x width = 480 x 640 pixels when the direction perpendicular to the traveling direction is horizontal. Good. Since the endless belt conveyor 4 is driven at 64 cm / second, in order to reliably capture the granular material to be transferred, the image capturing by the CCD camera 16 is performed so as to partially overlap the previously imaged range. Shooting is performed. That is, when the first shooting is performed and the second shooting is performed, the second shooting is performed so that the predetermined range overlaps with the first shooting range. Similarly, the third shooting is performed so that the predetermined range is overlapped with the second captured range, and the fourth shooting is performed so that the predetermined range is overlapped with the third captured range. In this way, by capturing the image of the granular material so that the predetermined range sequentially overlaps, the captured granular material can be reliably imaged.

  In addition, as the irradiation part 14 with which the imaging device 12 is provided, the white fluorescent lamp which has each wavelength of RGB light is used. Further, the color of the transfer surface (corresponding to “background member”) of the endless belt conveyor 4 is the color of the granular material so that the granular material can be clearly identified in the image taken by the CCD camera 16. A completely different color, such as black, is used.

  On the transfer end side of the endless belt conveyor 4, a granular material other than a predetermined granular material, for example, a granular material determined to be a granular material other than the raw material granular material, A blast device 18 is arranged as a sorting means for sorting and removing particles from the fall trajectory G of the granular material discharged from the end of the belt conveyor 4. A blast device drive unit 30 (see FIG. 2) for driving the blast device 18 is connected to the blast device 18, and the drive of the blast device 18 is controlled by the blast device drive unit 30. That is, a plurality of air nozzles are provided in a lattice shape at a position facing the drop trajectory G of the blast device 18, and among the granular materials falling along the drop trajectory G, the granular material to be excluded The particulate matter is selected by blowing high-pressure air only from the nozzles at positions corresponding to the matter (exclusion subject particulate matter). Here, in this embodiment, the particulate matter is selected using a blast device that blows high-pressure air, but any other particulate matter can be excluded from the fall trajectory G. An apparatus may be used. In the air blast direction of the blast device 18, an accommodating portion 19 that accommodates the particulate matter that is excluded and selected by the blast is disposed. Although not shown in FIG. 1, the granular material sorter 2 is provided with a control device 20 that controls processing in the granular material sorter 2.

  FIG. 2 is a block diagram showing an outline of the control device 20 according to the embodiment of the present invention. The control device 20 includes a preprocessing unit 22 connected to the CCD camera 16, a data storage unit 24 that stores an image of the granular material input to the control device 20 through the preprocessing unit 22, and each image from the granular image. A control unit 26 that performs processing such as converting the color indicated by the pixel of the granular material into discrimination information and a reference table 28 that stores the granular material color information are provided.

  The preprocessing unit 22 uses the image data from the CCD camera 16 to execute processing for specifying the pixel area of each granular material. Data indicating the color of each pixel constituting the pixel area of each granular material specified by the preprocessing unit 22 is stored in the data storage unit 24.

  The reference table 28 stores granular material color information, that is, information indicating a color unique to a predetermined granular material with hue and saturation as parameters. For example, wheat color information indicating a color unique to wheat, which is a raw material granular material, using hue and saturation as parameters is stored. In addition, when the granular material other than the raw material granular material (exclusion target granular material) mixed in the raw material granular material is clarified, the specific color of the excluded granular material, the hue and the saturation are parameters. The granular material color information of the granular material to be excluded shown as may be stored. Further, when a plurality of exclusion target granular materials are mixed in the granular material of the raw material, the granular material color information corresponding to each of the mixed exclusion target granular materials may be stored in the reference table 28.

  Here, the granular material color information will be described by taking as an example the case where the raw material granular material is wheat and the excluded granular material is mylo (red milo and white milo). It is known that color can be expressed by three attributes: hue, saturation, and lightness, hue represents the difference in hues of red, blue, and green, saturation represents the vividness of the color, and lightness represents the color Represents the brightness of In the granular material sorter according to the embodiment of the present invention, hue and saturation are used as parameters, but hue, saturation and brightness may be used as parameters.

  FIG. 3 is a diagram showing an example of the granular material color information of the raw material granular material, that is, information (wheat color information) indicating the color of wheat with hue and saturation as parameters. As shown in FIG. 3, the wheat color information has a hue on the horizontal axis and a saturation on the vertical axis, a hue between 0 and 80, and a saturation between 0 and 252 in four increments. In the table showing the color of each pixel constituting the image of wheat, it is information indicated by a designated area (area shown by oblique lines).

  Here, the wheat color information shown in FIG. 3 is created by using the photographed image by transferring only the wheat in advance in the granular material sorter according to the embodiment and photographing the wheat by the CCD camera 16. Is remembered. That is, before sorting the granular material, only wheat is experimentally transferred and an image of wheat irradiated with light by the irradiation unit 14 is taken by the CCD camera 16, and the wheat pixel in the taken wheat image is taken. Identify the area. And the color which each pixel which comprises the specified pixel area | region shows is converted into the information which used the hue and the saturation as parameters. Specifically, the color indicated by the RGB information which each of the pixels constituting the pixel region of wheat has is converted into information indicating hue and saturation as parameters. Next, the number of pixels showing the same hue and saturation is totaled, and a tabular result shown in FIG. 3 is created. Here, the numbers indicated by the tabulation results in FIG. 3 indicate the number of pixels that indicate a color using a predetermined hue and saturation as parameters among the pixels constituting the wheat image. That is, in FIG. 3, for example, the number of pixels indicating hue and saturation (28, 40) is “8”, and the number of pixels indicating hue and saturation (28, 72) is “94”. ing. And in the tabulated results, for example, the administrator who manages the sorting process in the granular material sorter 2 specifies the range showing the unique color of wheat, for example, as shown by the hatched area in FIG. Information indicated by the designated area is stored in the reference table 28 as wheat color information. It should be noted that the designation of the area that is the wheat color information may be automatically performed by the control unit 26. In other words, for example, the control unit 26 may designate an area to be wheat color information depending on whether the total number of pixels is greater than or less than a preset threshold value.

  In addition, after shooting a plurality of wheat and creating wheat color information for each wheat, the average value of the created wheat color information, that is, the average value of the number of pixels showing the same hue and saturation is calculated. One wheat color information may be created and stored in the reference table 28. In the table shown in FIG. 3, the area indicating the wheat color information can be specified in an arbitrary shape. For example, the area is not limited to a continuous area, and a plurality of areas representing colors unique to wheat are specified and specified. It is good also considering all the performed area | regions as wheat color information. Further, for example, as shown in FIG. 4, “◯” is shown when the unique color of wheat is shown, and “X” is shown when the unique color of wheat is not shown, corresponding to the hue and saturation values. The wheat color information may be stored in the reference table 28 in such a format. That is, the wheat color information may be stored as a table format in which “o” or “x” is associated with all of hue and saturation (0, 0) to (80, 252).

  FIG. 5 is a diagram illustrating an example of granular color information of the granular material to be excluded, that is, red mylo color information in which a color unique to red mylo is shown using hue and saturation as parameters. The red mylo color information is information indicated by the hatched area in the table of FIG. That is, similarly to the case shown in FIG. 3, the numbers in the hatched area in FIG. 5 indicate the number of pixels indicating the color unique to red milo. Specifically, in FIG. 5, for example, the number of pixels indicating hue and saturation (24, 116) is “64”, and the number of pixels indicating hue and saturation (20, 128) is “43”. It is shown that. Note that the red miro color information shown in FIG. 5 is also taken in the granular material sorter according to the embodiment by transferring only the red miro in advance and photographing the red milo by the CCD camera 16 in the same manner as the wheat color information shown in FIG. The image is created using the photographed image and stored in the reference table 28. Further, the red miro color information may also be stored in the reference table 28 in the format shown in FIG.

  FIG. 6 is a diagram showing granular color information of other granular objects to be excluded, that is, white mylo color information in which a color unique to white miro is used as a parameter of hue and saturation. The white mylo color information is information indicated by a hatched area in the table of FIG. 6 as with the above-described wheat color information and red mylo color information. That is, as in the case shown in FIGS. 3 and 5, the numbers in the hatched area in FIG. 6 indicate the number of pixels indicating the color unique to white milo. Specifically, in FIG. 6, for example, the number of pixels whose hue and saturation indicate (36, 92) is “28”, and the number of pixels whose hue and saturation indicates (36, 52) is “34”. It is shown that. In addition, white milo color information shown in FIG. 6 is also transferred in advance in the granular material sorter according to the embodiment in the same manner as the wheat color information shown in FIG. 3 and the red mylo color information shown in FIG. A white miro is photographed by the CCD camera 16, created using the photographed image, and stored in the reference table 28. Also, the white mylo color information may be stored in the reference table 28 in the format shown in FIG.

  For example, when the seeds of the different plants mixed in the raw material granular material are unknown, only the granular material color information indicating the color unique to the raw material granular material is stored in the reference table 28. For example, when it is necessary to discriminate the seeds of a specific different plant mixed in the raw material granular material, the granular material color information indicating the color unique to the seed of the different plant is stored.

  2 is connected to a blast device drive unit 30 that controls the drive of the blast device 18. The blast device driving unit 30 drives the blast device 18 to blast high-pressure air only from a predetermined air nozzle of the blast device 18 and includes a predetermined delay unit (not shown). That is, the control unit 26 specifies the position of the granular material determined to be a granular material other than the raw material granular material using the image captured by the CCD camera 16 on the image, and includes information indicating the specified position. A control signal is transmitted to the blast device drive unit 30. Therefore, the blast device driving unit 30 adjusts the time from when the CCD camera 16 is photographed until the granular material reaches the position facing the air nozzle of the blast device 18 in the fall locus G by the delay unit. The driving of the blowing device 18 is controlled so as to blow high-pressure air only from a predetermined air nozzle.

  Next, the particulate matter sorting process in the particulate matter sorter according to the embodiment of the present invention will be described. In the following, a case where the raw material granular material is wheat and the granular material other than wheat is selected using wheat color information stored in the reference table 28 will be described as an example.

  In the granular material sorter 2 shown in FIG. 1, the raw material granular material (wheat mixed with seeds of different plants) fed from the vibrating feeder 10 of the raw material supply unit 6 is transferred by the endless belt conveyor 4. At this time, light is irradiated by the irradiation unit 14 to the raw material granular material placed and transferred on the transfer surface of the endless belt conveyor 4, and an image of the raw material granular material is given by the CCD camera 16 in a predetermined pixel unit. Taken. Then, an image of the granular material photographed by the CCD camera 16 is input from the CCD camera 16 to the preprocessing unit 22. Since a plurality of raw material granules are placed on the endless belt conveyor 4, the image captured by the CCD camera 16 includes a plurality of raw material granular images.

  In the pre-processing unit 22, first, the pixel area of each granular object in the granular image captured by the CCD camera 16 is specified. Specifically, an area showing the color of the background member of the granular material from the image of the granular material photographed by the CCD camera 16, that is, the color of the transfer surface of the endless belt conveyor 4 which is transferring the granular material, for example, black. And an area showing a color other than black is specified as a pixel area of each granular material. Note that data indicating the color of each pixel constituting the pixel area of each granular material specified in the preprocessing unit 22 is stored in the data storage unit 24.

  Next, using the data indicating the color of each pixel constituting the pixel area of each granular material stored in the data storage unit 24, the color indicated by each of the pixels constituting the pixel area is indicated with hue and saturation as parameters. Converted into discriminant information. That is, the control unit 26 takes out RGB information indicating the color of the pixels constituting the pixel region of the granular material from the data storage unit 24. Next, using the extracted RGB information, the color indicated by the pixel is converted into information using hue and saturation as parameters, thereby converting the color indicated by the pixel into discrimination information. Similarly, the color indicated by the pixel is converted into discrimination information for all the pixels constituting the pixel area of the granular material. Here, the process of converting the color indicated by the RGB information into discrimination information having hue and saturation as parameters is executed using a known method. Here, the preprocessing unit 22 and the control unit 26 constitute conversion means.

  Next, the control unit 26 refers to the wheat color information stored in the reference table 28 and determines whether or not the determination information indicates a color unique to wheat. For example, as shown in FIG. 3, if the hue and saturation of the discrimination information is (28, 40), it is judged that this discrimination information indicates the color of wheat, and the hue and saturation of the discrimination information is ( 28, 116), it is determined that this determination information does not indicate the color of wheat. Similarly, with respect to all the discrimination information constituting the pixel area of one grain, it is discriminated whether or not each of them indicates the color of wheat by referring to the wheat color information.

  Next, the number of discriminating information determined to indicate the color of wheat for one grain, that is, the number of pixels determined to indicate the color of wheat, is totalized, and the total number of pixels is The ratio occupied in the pixel area is calculated. And when the ratio of the discrimination | determination information discriminate | determined which shows the color of wheat is more than predetermined, for example, 80% or more etc., it will discriminate | determine that this granular material is wheat. On the other hand, when the ratio of the discrimination information is less than a predetermined value, for example, less than 80%, the granule is discriminated as a non-wheat granule, that is, a seed of a heterogeneous plant mixed in wheat. Is done. In the control unit 26, the ratio of the discrimination information in the pixel area of the granular material when determining whether the granular material is the raw material granular material can be arbitrarily set.

  Next, the exclusion target particulate matter determined not to be wheat is excluded from the fall track G and selected when it falls along the fall track G from the transfer end side of the endless belt conveyor 4 by the blast device 18. In other words, the control unit 26 uses the image taken by the CCD camera 16 to specify the position of the exclusion target granular material in the image, and a control signal including information indicating the position of the exclusion target granular material is driven by the blast device. Transmitted to the unit 30. In response to a control signal from the control unit 26, the blast device driving unit 30 determines an air nozzle that blasts high-pressure air among the air nozzles provided in the blast device 18. Then, the timing is adjusted by a delay unit (not shown), and when the granular material photographed by the CCD camera 16 reaches the fall locus G, the blowing device 18 is driven to blow high-pressure air from a predetermined air nozzle. Thus, the exclusion target particulate matter is excluded from the fall trajectory G, and the excluded exclusion subject particulate matter is accommodated in the accommodating portion 19.

  According to the granular material sorter according to this embodiment, the granular material color information accurately indicating the specific color of the granular material is stored in the reference table using the hue and saturation that accurately indicate the color characteristics of the granular material as parameters. Therefore, the granular material can be selected with high accuracy. That is, when the raw material granular material is wheat, the wheat color is indicated by hue and saturation as parameters (see FIG. 3), the region indicating the unique color of wheat (hue and saturation range) is specified and specified. The stored range is stored in the reference table as wheat color information (granular material color information) that accurately shows the color unique to wheat, and the granular material is selected with reference to the stored wheat color information. For example, even when the colors of wheat and granular materials mixed in wheat are similar, wheat and granular materials other than wheat can be selected with high accuracy.

  Further, in the granular material sorter according to the above-described embodiment, the color of the captured granular material is converted into discrimination information using the hue and saturation as parameters, and the granular material color information is referenced with reference to the stored granular material color information. Sorting things. That is, since the color of the granular material is converted into the discrimination information indicated by using two parameters, the number of parameters to be used is smaller than in the case of indicating the color in the spatial domain, and the color of the granular material is used as the discrimination information. The amount of calculation at the time of conversion can be reduced, and a quick sorting process can be performed. In addition, since it is possible to perform a quick sorting process, it is possible to improve the processing capacity of the sorting process in the granular material sorter.

  In the above-described embodiment, the case where wheat color information is stored as the granular material color information is taken as an example, but the granular material color information of the exclusion target granular material may be stored. For example, when it is clear that red milo is mixed in wheat and it is necessary to sort out the red miro, the red miro color information indicating the color of the red miro that is the object of exclusion is stored in the reference table. Let Then, it is determined whether or not the discrimination information indicating the color of the granular material indicates the color of red milo with reference to the red miro color information, and the selection is performed only when it is determined that the granular material is red milo. It may be possible to exclude only red milo without fail. Further, without selecting the red milo, for example, by storing the number of particulate matter determined to be red miro in the data storage unit and outputting the determination result stored in the data storage unit via an output unit (not shown) or the like. The ratio of red milo mixed in wheat may be indicated with high accuracy.

  Moreover, you may make it memorize | store several granular material color information as granular material color information of an exclusion object granular material. That is, it is clear that red milo and white milo are mixed in wheat, and when it is necessary to distinguish or sort red milo and white milo, red miro color information is used as granular material color information of the granular material to be excluded. And white mylo color information are stored in the reference table. Then, referring to the red milo color information and the white milo color information, it is discriminated whether the discrimination information indicates the color of red milo or the color of white milo, and it is discriminated whether the granular material is red milo or white milo. However, the determination result may be stored or selected only when it is determined as red milo or white milo. In this case, for example, it is possible to appropriately check the mixing situation of a specific granular material from among a plurality of granular materials mixed in wheat. Further, for example, it is possible to exclude a granular material that needs to be excluded from a plurality of granular materials mixed in wheat with high accuracy.

  Further, the granular material color information of the raw material granular material and the granular material color information of the exclusion target granular material may be stored in the reference table, respectively. For example, in the reference table, the wheat color information may be stored as the granular material color information of the raw material granular material, and the red mylo color information and the white mylo color information may be stored as the granular material color information of the exclusion target granular material. And it may be made to discriminate | determine whether the discrimination | determination information of a granular material shows the color of wheat, the color of red milo, or the color of white miro, and you may make it discriminate | determine whether a granular material is wheat.

It is a longitudinal section of a granular material sorter concerning an embodiment. It is a block block diagram of the control apparatus of the granular material sorter which concerns on embodiment. It is a figure which shows an example of the granular material color information of the predetermined granular material which concerns on embodiment. It is a figure which shows the other example of the granular material color information of the predetermined granular material which concerns on embodiment. It is a figure which shows an example of the granular material color information of the exclusion object granular material which concerns on embodiment. It is a figure which shows the other example of the granular material color information of the exclusion object granular material which concerns on embodiment.

Explanation of symbols

2 ... granular material sorter, 4 ... endless belt conveyor, 12 ... photographing unit, 16 ... CCD camera, 18 ... blast device, 20 ... control device, 22 ... Pre-processing unit, 26... Control unit, 28... Reference table, 30.

Claims (5)

A transfer device for transferring particulate matter;
An imaging device that is disposed at a position opposite to the granular material transferred by the transfer device, and irradiates the transferred granular material with light to shoot an image of the granular material;
A reference table storing granular color information indicating colors specific to a predetermined granular object using hue and saturation as parameters;
Conversion means for converting the color indicated by the pixels constituting the granular object into discrimination information using hue and saturation as parameters using an image photographed by the photographing device;
For each of the discrimination information of each pixel constituting the granular material, the granular material color information stored in the reference table is used to determine whether or not the color of the predetermined granular material is indicated. When the number of discriminating information determined to indicate the color of the predetermined granular object out of the number of pixels constituting the pixel area of the object is equal to or greater than a predetermined ratio, the granular object is the predetermined granular object and discriminating means for discriminating that there is,
An output unit for outputting a discrimination result by the discrimination means;
A granular material sorter characterized by comprising: A transfer device for transferring particulate matter;
An imaging device that is disposed at a position opposite to the granular material transferred by the transfer device, and irradiates the transferred granular material with light to shoot an image of the granular material;
A reference table storing granular color information indicating colors specific to a predetermined granular object using hue and saturation as parameters;
Conversion means for converting the color indicated by the pixels constituting the granular object into discrimination information using hue and saturation as parameters using an image photographed by the photographing device;
For each of the discrimination information of each pixel constituting the granular material, the granular material color information stored in the reference table is used to determine whether or not the color of the predetermined granular material is indicated. When the number of discriminating information determined to indicate the color of the predetermined granular object out of the number of pixels constituting the pixel area of the object is equal to or greater than a predetermined ratio, the granular object is the predetermined granular object and discriminating means for discriminating that there is,
Screening means for excluding the particulate matter determined not to be the predetermined particulate matter by the discrimination means;
A granular material sorter characterized by comprising: The selecting means includes
The granular material sorter according to claim 2, wherein a granular material determined not to be the predetermined granular material is excluded from a falling trajectory of the granular material discharged from an end portion of the transfer device. The imaging device
Irradiating means for irradiating the granular material to be transported with light;
The granular material sorter according to any one of claims 1 to 3, further comprising a photographing element that captures an image of the granular material irradiated with light by the irradiation unit. The converting means includes
In the image photographed by the photographing element, the color indicated by each pixel constituting the pixel area is specified by specifying the pixel area of each granular object from the difference between the color of the background member of the granular object and the color of the granular object The granular material sorter according to claim 4, wherein the granular material sorter is converted into discrimination information.

Images