JP2006021170A - Method and apparatus for sorting quality of vegetable and fruit - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method and an apparatus for accurately sorting the quality of vegetables and fruits according to the types of inner anomaly of the vegetables and fruits and the degree of the anomaly. <P>SOLUTION: Insides of the vegetables and fruits 1 are irradiated with measuring beam, on the basis of signals obtained by the transmitting beam of the measuring beam passing through the vegetables and fruits 1, and an inner anomaly value preset according to the signals of transmitting beam is determined. Appearance of the vegetables and fruits 1 is imaged with an imaging apparatus 6, and the image of the vegetables and fruits 1 obtained by the imaging is processed to discriminate color tone or shape of the vegetables and fruits 1. Types of inner anomaly of the vegetables and fruits 1 and the degree of inner anomaly are sorted based on the sectioned range preset to the inner anomaly value according to the degree of inner anomaly of the vegetables and fruits 1, and the discrimination result of the color tone or shape of the vegetables and fruits 1 by image processing. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a quality sorting method and a quality sorting apparatus for performing quality sorting according to internal abnormalities of fruits and vegetables, and more particularly, to an improvement in technology for improving the accuracy of sorting types and degrees of internal abnormalities.

  Fruits and vegetables, such as fruits and vegetables, are sorted and shipped according to the internal quality after harvesting. Conventionally, various techniques for measuring the internal quality of fruits and vegetables from the outside without destroying the fruits and vegetables have been proposed.

  Among these, as a method for measuring the sugar content, acidity, and the like of fruits and vegetables, a method by introspection measurement that measures transmitted light from fruits and vegetables irradiated with measurement light is known (see, for example, Patent Document 1). In this method, for each fruit and vegetables that are continuously transported through the transport line, irradiate the measurement light from the light source installed in the middle of the transport line, and detect the transmitted light that has passed through the inside of the fruit and vegetables, The sugar content, acidity, etc. are measured by analyzing the transmitted light signal from the light receiving section.

For example, when measuring internal abnormalities of fruits and vegetables such as hollow fruits, for example, by performing transmitted light measurement and visual observation on a large number of samples, parameters indicating the degree of internal abnormalities (internal abnormal values) ) In advance corresponding to the magnitude of the transmitted light signal, the transmitted light from the fruit or vegetable to be measured is detected by the light receiving unit, and an internal abnormal value is determined for this fruit or vegetable based on the transmitted light signal from the light receiving unit. The quality is selected according to the internal abnormal value.
JP 2003-021598 A

  However, taking tomato as an example, there are so-called bucchi balls and hollow fruits as types of internal abnormality, but in the above-described introspection measurement using transmitted light, the internal abnormality is determined based on the transmitted light. It is difficult to determine the type, that is, whether the internal abnormal value obtained by the measurement is due to a bunch ball or a hollow fruit.

  On the other hand, it is also conceivable to select internal abnormalities by measuring the appearance of fruits and vegetables. That is, each fruit and vegetable that is continuously conveyed on the conveyance line is imaged by an imaging device such as a color CCD camera installed in the middle of the conveyance line, and the obtained image data is subjected to image processing, and shape classification or tone classification is performed. By doing this, it is a method of identifying internal abnormalities according to the classification. For example, with tomatoes, it is possible to select hollow fruits according to the degree of deformation from a round shape, or select buku balls with an approximate degree of color tone to red and black peculiar to buku balls.

  However, even when trying to sort out the tomato hollow fruit according to the degree of deformation of the shape, it is difficult to accurately sort the presence / absence of the hollow fruit and its degree if the degree of deformation is relatively small.

Further, even when trying to sort buku balls by color tone, it is difficult to accurately sort the presence / absence of buku balls and the degree thereof only by the redness degree of the color tone.
The present invention has been made to solve the above-described problems in the prior art, and the quality of fruits and vegetables that can be accurately selected according to the types of internal abnormalities of fruits and vegetables and the degree of the internal abnormalities. An object is to provide a sorting method and a quality sorting apparatus.

The quality sorting method for fruits and vegetables of the present invention is a quality sorting method for fruits and vegetables that performs sorting according to its internal abnormality,
Irradiating the inside of the fruits and vegetables with the measurement light, based on the signal by the transmitted light that has passed through the inside of the fruits and vegetables, the internal abnormal value set in advance corresponding to the signal by the transmitted light,
Imaging the appearance of the fruits and vegetables by the imaging device, image processing the captured images of the fruits and vegetables by the imaging device to identify the color tone or shape of the fruits and vegetables,
Depending on the degree of internal abnormality of the fruit and vegetable, the division range preset to the internal abnormality value, and the identification result of the color or shape of the fruit and vegetable by the image processing, the type of internal abnormality of the fruit and vegetable and the internal abnormality It is characterized by selecting the degree.

The fruit and vegetable quality sorting device of the present invention is a fruit and vegetable quality sorting device that performs sorting according to internal abnormalities in the fruits and vegetables that are continuously transported through the transport line,
A light source for irradiating measurement light to the fruits and vegetables conveyed through the conveyance line;
A light receiving means for receiving the transmitted light transmitted from the light source through the fruit and vegetables;
An internal abnormal value determining means for determining an internal abnormal value set in advance corresponding to the signal by the transmitted light based on the signal by the transmitted light received by the light receiving means;
An imaging device for imaging the appearance of the fruits and vegetables;
Image processing means for image-processing the captured image of the fruits and vegetables by the imaging device, and giving a measurement result for identifying the color tone or shape of the fruits and vegetables;
The type and internal abnormality of the internal abnormality of the fruit and vegetable according to the division range preset to the internal abnormality value according to the degree of internal abnormality of the fruit and vegetable and the identification result of the color tone or shape of the fruit or vegetable by the image processing means And a selecting means for specifying the degree.

  In the above invention, for a range where the internal abnormal value by the transmitted light measurement is relatively high, a division range is set in advance as the internal abnormal value, and the shape or color tone obtained by the external appearance measurement by the imaging device is identified. The result identifies the type of internal abnormality (for example, boiled balls, hollow fruits in tomatoes) and the size of the internal abnormality.

  In this way, by the introspection measurement with transmitted light, it is possible to discriminate fruits and vegetables that have an internal abnormality reliably or likely to exist, and the identification result of the shape or color tone obtained by the appearance measurement by the imaging device Since the selection is made in combination with the result of introspection, the types of internal abnormalities of fruits and vegetables can be accurately determined, and the degree of internal abnormalities can be accurately determined for each type of internal abnormalities. it can.

  ADVANTAGE OF THE INVENTION According to this invention, the selection according to the kind of internal abnormality of fruit and vegetables and the degree of internal abnormality can be performed with sufficient precision.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a schematic diagram of an apparatus used for a quality selection method according to an embodiment of the present invention. In the present embodiment, an example is shown in which the quality of the harvested tomato is selected for each grade (A to D and extraordinary) according to internal abnormalities (buck balls, hollow fruits). As shown in the drawing, an introspection measuring unit in which a light source 4 and a light receiving unit 5 are arranged is provided in the middle of a transport line 2 through which the tomato 1 is transported. On the downstream side in the transport direction, an appearance measuring unit in which the imaging device 6 and the light 7 are arranged in the light shielding chamber 8 is provided.

  The tomato 1 placed on the tray 3 is transported on the transport line 2, and in the above interior measurement unit, the measurement light is irradiated from the light source 4 to the tomato 1, and the transmitted light transmitted through the inside of the tomato 1 is received by the light receiving unit. 5 is detected.

  As the light source 4, for example, a halogen lamp, a xenon lamp, or the like can be used, and the light source 4 is disposed at an appropriate position such as above, laterally, or obliquely above the transport line 2. A plurality of light sources may be arranged.

  The light receiving unit 5 is disposed below the transport line 2 and receives light transmitted downward from a through hole provided in the tray 3. The light receiving unit 5 is composed of, for example, a photomultiplier tube, a photodiode, or the like. When the incidence of disturbance light on the light receiving unit 5 is large, a light shielding member for blocking disturbance light is installed as necessary.

  Techniques relating to the internal measurement of fruits and vegetables using transmitted light are well known to those skilled in the art, and an appropriate measurement system may be configured according to circumstances. For example, in order to remove disturbance light components, the illumination light may be chopped, and the transmitted light signal may be synchronously detected by a lock-in amplifier using a reference signal corresponding to the chopping frequency, and the light source may be a carrier line. 2 may be disposed below and the light receiving unit 5 may be disposed above.

  The transmitted light incident on the light receiving unit 5 is photoelectrically converted, and the electric signal is sent to the computer 9 through an amplifier, an AD converter, and the like. In the computer 9, the correlation between the transmitted light signal of the tomato measured by this introspection measuring unit and the degree (size) of internal abnormality by visual observation of the tomato is based on the results of examining a large number of tomatoes. An internal abnormal value is set according to the transmitted light signal. In this embodiment, an internal abnormal value is set according to the intensity of the transmitted light signal, and this correlation data is stored in the storage unit of the computer 9.

  The actually measured value of the transmitted light signal of the tomato 1 as the measurement target is compared with the correlation data stored in the storage unit by the calculation unit of the computer 9, and an internal abnormal value is determined for the tomato 1.

  The computer 9 can set an arbitrary partition range for determining the normal product and the abnormal product or the degree of the internal abnormality in the storage unit for the internal abnormal value. FIG. 2 is a graph showing an example of the relationship between visual measurement of internal abnormalities (bump balls, hollow fruit) of tomatoes and internal abnormal values determined from the transmitted light signal by the above-mentioned interior measurement unit. In the example of the figure, for example, by setting the threshold value to 3.4 and setting the range in which the internal abnormal value is 3.4 or more as the partition range, normal products and abnormal products including broom balls or hollow fruits are obtained. Can be determined.

  In FIG. 1, the tomato 1 whose transmitted light has been measured by the interior measurement unit is conveyed to an appearance measurement unit provided on the downstream side of the interior measurement unit. In this appearance measuring unit, the appearance of the tomato 1 is imaged by the imaging device 6 such as a CCD camera or a television camera, and an image signal from the imaging device 6 is sent to the computer 9. In the computer 9, image data of the tomato 1 is subjected to image processing in order to identify the shape or color tone of the tomato 1.

  Techniques related to appearance inspection of the shape or color tone of fruits and vegetables using a CCD camera, a TV camera, etc. are well known to those skilled in the art, and the measurement system, image processing method, shape or color tone identification method can be changed according to circumstances. Can be configured appropriately.

When identifying by the shape of a tomato, as an example, the upper part of the tomato 1 is irradiated with the light 7, the tomato 1 is imaged from above by the imaging device 6 installed above the center of the tomato 1, and this captured image is A / D The image of tomato 1 and the background image are separated by binarization using a converter. Next, for example, the center of gravity of the image of the tomato 1 is obtained, and the center of the mask image provided with a straight line at every predetermined angle from the center is matched with the center of gravity of the binarized image of the tomato 1, and the logical product is calculated between each pixel. The distance from the center of gravity to the contour of the tomato can be calculated for each straight line, and the shape can be determined based on this distance. The shapes can be classified as, for example, “circular”, “elliptical”, “square”, “distorted”, and the like. The identification result of the tomato shape by the appearance measurement may be obtained as a shape classification such as “circular” or “ellipse”, or may be obtained as an appearance measurement value obtained by quantifying the shape.

  When discriminating according to the color tone of tomato, as an example, color image of tomato 1 is taken from above, and intensity data of each component of red, green, and blue of this captured image is stored for each pixel, and identification is performed based on this stored data. Calculate the result.

  For example, for each pixel of the captured image, the brightness obtained by the sum of each component of red, green, and blue and the chromaticity obtained by the ratio of each component to the brightness of that pixel are used for the entire image. The target tomato is discriminated based on the degree, and an average value is obtained for the pixels constituting the target tomato. The identification result of tomato color tone by appearance measurement may be obtained by classifying the average value of such color tone into “red black”, “red”, “pink”, “green”, etc. It may be a measured value.

In the case of tomatoes, the color tone tends to become red and black when buccal balls are present inside, and when hollow fruits are present inside, the shape tends to be deformed from a round shape.
In the computer 9, the internal abnormal value obtained by measuring the transmitted light of the tomato 1 in the introspection measuring unit and the color tone or shape obtained by image processing of the image data of the tomato 1 imaged in the appearance measuring unit. Based on the identification result, the degree and type of internal abnormality are identified and sorted into each grade.

  As shown in FIG. 3, quality selection can be performed based on the internal abnormal value obtained by measuring the transmitted light of the tomato and the appearance identification result regarding the shape of the tomato. In the example of the figure, the threshold value of the internal abnormal value is set to 3.4 and 3.9, and as a result of the transmitted light measurement, the internal abnormal value is included in this section range 3.4 to 3.9, Depending on the result of identification by appearance imaging, the product is selected as a D product (small to medium hollow fruit) or a non-standard product (small to medium buzz). That is, as a result of appearance imaging, for those whose shape is identified as a round shape or a shape close to a round shape, in this partition range 3.4 to 3.9, small to medium buku balls are inside the tomato. There is a high possibility that it exists, and it is selected as an unconventional product. On the other hand, as a result of appearance imaging, small to medium hollow fruits are present in the tomato inside the section range 3.4 to 3.9 in which the shape is identified as a shape deformed from a round shape. The possibility is high and it is sorted into D products.

Those whose internal abnormal value by transmitted light measurement exceeds 3.9 are highly likely to have large buccal balls or large hollow fruits in the tomato, and are selected as extraordinary products.
Those having an internal abnormal value of less than 3.4 as measured by transmitted light are selected from products A to C according to the internal abnormal value.

Moreover, quality selection can be performed as shown in FIG. 3 based on the internal abnormal value obtained by measuring the transmitted light of the tomato and the appearance identification result of the color tone of the tomato. In the example of the figure, the threshold value of the internal abnormal value is set to 3.4, 3.9, and a certain value less than 3.4 (X in the figure). As a result of the transmitted light measurement, for products whose internal abnormal values are included in the partition range 3.4 to 3.9, D products (small to medium hollow fruits) or non-standard products (small) ~ Medium Bukudama). That is, as a result of appearance imaging, with respect to the one whose color tone is identified as red-black or red-black to red, in this section range 3.4 to 3.9, small to medium buzz balls are inside the tomato. There is a high possibility that it exists, and it is selected as an unconventional product. On the other hand, as a result of appearance imaging, for those whose color tone is identified as pink to green, there is a possibility that small to medium hollow fruits may exist inside the tomato in this section range 3.4 to 3.9. High and sorted into D products.

  Furthermore, as a result of the transmitted light measurement, for those whose internal abnormal values are included in the section ranges X to 3.4, the D product is selected according to the identification result by the appearance imaging. That is, as a result of the appearance imaging, with respect to those whose color tone is identified as red or black or a color tone close to red to red, there is a possibility that a small boulder is present inside the tomato in this section range X to 3.4. Is high and sorted into D products. On the other hand, as a result of appearance imaging, those whose color tone is identified as pink to green are sorted into A product to C product in this section range X to 3.4 according to the internal abnormal value. In addition, as shown in the figure, the discrimination boundary of bukudama by appearance imaging in the section range X to 3.4 is more black than the discrimination boundary between bukudama and hollow fruit in the section range 3.4 to 3.9. You may make it set to the near side.

Those whose internal abnormal value by transmitted light measurement exceeds 3.9 are highly likely to have large buccal balls or large hollow fruits in the tomato, and are selected as extraordinary products.
Those having an internal abnormal value of less than X by transmitted light measurement are selected from A to C according to the internal abnormal value.

  Especially in tomatoes, as shown in FIG. 2, there are many small round balls in the internal abnormal value near the lower part of the threshold value 3.4, and the value of X should be determined appropriately. This makes it possible to select the degree of baku balls with high accuracy.

In FIG. 1, the tomatoes 1 selected for each grade as described above are further sorted for each grade selected on the downstream side of the transport line 2.
As described above, the measurement result of transmitted light measurement and appearance imaging of the fruits and vegetables to be measured is a preliminarily set division range in a range where the internal abnormal value by transmission light measurement is relatively high, and the shape or color tone by appearance imaging. By determining which of the areas divided by the partition range based on the identification result belongs to, it is possible to accurately select according to the type and size of the internal abnormality.

  As mentioned above, although embodiment of this invention was described, this invention is not limited to this embodiment, A various deformation | transformation and change are possible. For example, for fruits and vegetables such as fruits other than tomatoes, as long as the transmitted light can be measured for the fruits and vegetables, and there is a tendency for the shape or color tone according to the internal abnormality, internal abnormality It can be applied to sorting according to the type and the degree.

FIG. 1 is a schematic diagram of an apparatus used for a quality selection method according to an embodiment of the present invention. FIG. 2 is a graph showing an example of the relationship between the visual measurement of internal abnormalities (bump balls, hollow fruits) of tomatoes and the internal abnormal values determined from the transmitted light signal by the interior measurement unit. FIG. 3 is a diagram illustrating an example in which quality selection is performed based on an internal abnormal value obtained by measuring transmitted light of a tomato and an appearance identification result regarding the shape of the tomato. FIG. 4 is a diagram illustrating an example in which quality selection is performed based on an internal abnormal value obtained by measuring transmitted light of a tomato and an appearance identification result regarding the color tone of the tomato.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Tomato 2 Conveyance line 3 Tray 4 Light source 5 Light-receiving part 6 Imaging device 7 Light 8 Light shielding room 9 Computer

Claims (2)

A quality sorting method for fruits and vegetables that performs sorting according to internal abnormalities.
Irradiating the inside of the fruits and vegetables with the measurement light, based on the signal by the transmitted light that has passed through the inside of the fruits and vegetables, the internal abnormal value set in advance corresponding to the signal by the transmitted light,
Imaging the appearance of the fruits and vegetables by the imaging device, image processing the captured images of the fruits and vegetables by the imaging device to identify the color tone or shape of the fruits and vegetables,
Depending on the degree of internal abnormality of the fruit and vegetable, the division range preset to the internal abnormality value, and the identification result of the color or shape of the fruit and vegetable by the image processing, the type of internal abnormality of the fruit and vegetable and the internal abnormality A method for quality selection of fruits and vegetables characterized by selecting the degree. A fruit and vegetable quality sorting device that performs sorting according to internal abnormalities in the fruits and vegetables that are continuously conveyed through the transportation line,
A light source for irradiating measurement light to the fruits and vegetables conveyed through the conveyance line;
A light receiving means for receiving the transmitted light transmitted from the light source through the fruit and vegetables;
An internal abnormal value determining means for determining an internal abnormal value set in advance corresponding to the signal by the transmitted light based on the signal by the transmitted light received by the light receiving means;
An imaging device for imaging the appearance of the fruits and vegetables;
Image processing means for image-processing the captured image of the fruits and vegetables by the imaging device, and giving a measurement result for identifying the color tone or shape of the fruits and vegetables;
The type and internal abnormality of the internal abnormality of the fruit and vegetable according to the division range preset to the internal abnormality value according to the degree of internal abnormality of the fruit and vegetable and the identification result of the color tone or shape of the fruit or vegetable by the image processing means And a quality sorting device for fruits and vegetables, comprising: sorting means for specifying the degree of the above.

Images