JP2642107B2 - Automatic grading method and device for fruits and vegetables - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial application field] The present invention takes fruits and vegetables which are conveyed in a fixed direction on a conveyor or the like with a color sensor camera, analyzes the images, and automatically processes the images. The present invention also relates to a method and apparatus for automatically determining the grade of fruits and vegetables.

[Prior Art] Generally, grades based on weight or size and grades based on appearance and the like are used as sorting criteria for fruits and vegetables.

As a device for classifying by class according to the weight or the size, for example, a weight sorter using an electronic scale or the like, a shape sorter using a camera device or the like is known.

In addition, in order to determine the grade of fruits and vegetables as infinitely varied natural products based on the grade selection criteria, the present situation relies on a visual inspection of a sorter. Although these grade selection criteria are expressed in sentences, they are inspection items that are not quantified, and the judgment results based on the grade selection criteria are not always constant when different classifiers are used.

Furthermore, even if the same discriminator is judged, it may not be possible to make a certain judgment due to environmental differences and time differences. As a result, erroneous judgments may occur in the quality (grade) judgment, and the credibility of market consumers will be reduced, and However, there is a disadvantage that the price is lowered.

On the other hand, a method for automatically inspecting the appearance of oranges such as shape and size, color, and scratches is disclosed in Japanese Patent Application Laid-Open No. 59-87081 filed by the group including the present applicant. According to this, the inspection of the coloring state is to inspect the average value of the color distribution of the imaged surface as the degree of coloring, and is suitable for fruits and vegetables of a single color such as tangerines and persimmons, but is infinite. For some varieties, such as apples having a lot of variation and color unevenness, the grading (selection) criteria based on the coloring state are different for each part such as the apex and the stalk. For example, varieties such as Tsugaru have the apex. Is green and has excellent grade and is colored (red)
The varieties such as Chiaki and Fuji have good grades, while the green ones at the top are good grades and the colored (red) varieties have good grades. There is a case where the quality (grade) is completely opposite depending on the part, and the conventional one in the above-mentioned gazette has a problem that the grading criteria based on the coloring state cannot be applied to fruits and vegetables different in each part such as a top part and a stalk part. was there.

[Problems to be Solved by the Invention] The present invention has been made in view of the above-mentioned problems, and an object of the present invention is to determine a grading standard based on a coloring state of a fruit and vegetable surface for each part such as a top part or a corneal part. Another object of the present invention is to provide a method and apparatus for automatically determining the grade of a fruit or vegetable, which can automatically and accurately determine the grade of the fruit or vegetable according to the coloring state of the fruit or vegetable.

[Means for Solving the Problems] The present invention employs the following technical means to achieve the above object.

That is, in the first invention of the present invention, on the conveyor, the fruit side illuminates the fruits and vegetables conveyed in a fixed posture facing the left and right sides of the conveyor, and emits light from the upper surface of the illuminated fruits and vegetables. At the same time as guiding to the sensor of a single color sensor camera placed above the conveyor, light from both sides of the fruits and vegetables reflected from a pair of reflecting means arranged on the left and right sides of the conveyor is sent to the sensor. By guiding the sensor to form three images of the upper surface and both side surfaces of the fruits and vegetables individually side by side and forming the images, the color sensor camera can be used to produce three surfaces of the upper surface of the waist of the fruits and vegetables, the side surface of the fruit stalk, and the side surface of the fruits top. Is imaged, the degree of coloring is calculated for each of the three surfaces from the imaging results of the three surfaces, and the calculated degree of coloring of each surface is compared with a color reference value preset for each surface, respectively. The grade is determined for each surface, and the It is obtained by the configuration of the automatic grade determination method of fruit or vegetable, characterized in that to determine the grade according to the colored state of the fruit or vegetable by integrating the grade determination result.

Further, in the second invention of the present invention, a fruit stalk side on the transport conveyor illuminates fruits and vegetables that are transported in a fixed posture facing the left and right side of the transport conveyor, and an illumination device that is disposed above the transport conveyor, A single color sensor camera that captures an image of the surface of the fruit and vegetable by imaging the surface of the fruit and vegetable on the sensor by light from the illuminated fruit and vegetable, and a pair of reflecting means disposed on the left and right sides of the conveyor. Light from the upper surface of the fruits and vegetables and light from both sides of the fruits and vegetables reflected from the reflection means are guided to the sensor to form three images of the upper surface and the both sides of the fruits and vegetables separately on the sensor. An optical system, and an arithmetic processing unit for inputting an imaging result of the color sensor camera, wherein the color sensor camera has three surfaces: a waist upper surface, a fruit stalk side surface, and a fruit apex side surface of the fruits and vegetables. The arithmetic processing device calculates the degree of coloring for each of the three surfaces from the imaging results of the three surfaces, and calculates the degree of coloring of each of the surfaces and a color reference value preset for each surface. A grade is determined for each surface by comparison, and the grade determination based on the coloring state of the fruit or vegetable is determined by integrating the grade determination results of the respective surfaces to form an automatic grade determination device for fruits and vegetables. .

[Example] Hereinafter, a description will be given based on the drawings showing a preferred example of the present invention.

1 to 6, a description will be given of an apparatus for sorting and sorting fruits and vegetables implemented using the present invention.

1 is a conveyor, 2 is a color sensor camera, 3 is a mirror, 4 is a small lamp, 5 is an arithmetic processing unit, and 6 is fruits and vegetables whose appearance is inspected.

The conveyer 1 supplies the fruits and vegetables 6 from the previous process in a posture in which the central axis of the fruit axis is turned sideways at right angles to the traveling direction, and the fruits and vegetables 6 on the conveyor 1 are oriented so that the fruit stalk side faces one side of the conveyor 1 left and right. Are separated one by one in a row and back and forth, and are randomly transported.
It is preferable to use a conveyor device as proposed in the publication. The conveyor 1 may be a chain conveyor, a belt conveyor, or the like, but the conveyor surface 11 is a black matte that hardly reflects light.

The color sensor camera 2 uses an image sensor of a MOS type or CCD type linear image sensor 201 (FIG. 7), and more specifically, a light receiving portion of a light receiving element used in a color facsimile, a color copier, and the like. A camera using a color linear image sensor in which R, G, and B filters are formed on-chip is used. The color sensor camera 2 is provided so that the scanning line 21 of the color sensor camera 2 crosses in a direction perpendicular to the transport path downward at a predetermined position above the transport path of the transport conveyor 1 so as to equally divide the visual fields on both the left and right sides. I do.

The mirrors 3, 3 are arranged on both outer sides of the conveyor 1, and are provided symmetrically at a predetermined angle so as to reflect light from both sides of the fruits and vegetables 6 to the color sensor camera 2 above. The mirrors 3 may be narrow as long as they are located at the position where the scanning line 21 of the color sensor camera 2 scans, but the length of the inclined surface covers the field of view determined from the size of the fruits and vegetables 6. Length.

By combining the color sensor camera 2 with the mirrors 3 and 3 as described above, the three images of the upper surface of the waist, the side surface of the fruit stalk, and the side surface of the fruit apex of the fruits and vegetables 6 are output to the linear image sensor 201 of the camera 2. It is configured so that images can be taken side by side individually (FIG. 7).

When the fruits and vegetables 6 conveyed by the conveyor 1 are on the scanning line 21 of the color sensor camera 2, the small lamps 4 are arranged on the front side and the rear side of the scanning line 21, centered on the center line of the conveying path. When the fruits and vegetables 6 pass through the scanning line 21, they are adjusted so that the light projection distances from the scanning surface of the fruits and vegetables are substantially equal.

That is, as shown particularly in FIG. 5, the small lamps 4 are arranged so that when the center of the fruit or vegetable 6 is on the scanning line 21, the respective distances are radially outward from the center of the fruit or vegetable 6 and the distances are substantially equal. It is provided so that the scanning surface of the fruits and vegetables 6 is uniformly irradiated.

Reference numeral 41 denotes a lamp box, in which a frosted glass material that diffuses and transmits light is used on the side facing the transport path and the side facing the scanning line 21, and the other surface that is not irradiated with light is made of a material that does not transmit light. It is preferable to apply a white paint such as magnesium to the other inner peripheral surface except the ground glass material.

The arithmetic processing unit 5 includes a camera signal control unit, a CP
It is composed of a main part (not shown) of U, a grade and class standard setting part, and a sorting control part.

In FIG. 1, reference numeral 7 denotes a drawer conveyor, which is a sorting / discharging section 100 provided for each grade and class of the conveyor 1.
Is provided in a direction crossing the conveyor 1. The drawer conveyor receives the fruits and vegetables 6 discharged from the conveyor 1 by the discharge device 101 of the sorting / discharging section 100 and carries it out of the machine. The transfer surface has a cushioning material (for example, artificial turf). It is preferable to use a belt conveyor provided with a soft nap material.

Next, an imaging procedure of the fruits and vegetables 6 conveyed on the conveyor 1 will be described with reference to FIGS.

Linear image sensor 201 is imaging the area of the right side surface A of the fruits or vegetables 6 element surface from A ₁ to A _2, the imaging area of the upper surface B of the fruits or vegetables 6 element surface from B ₁ to B _2, from C ₁ C ₂ The element surface up to this is set as an imaging area on the left side surface C of the fruit and vegetable 6.

That is, when fresh produce 6 conveyed by the conveyor 1 reaches the position across the scan line 21 of the color sensor camera 2, the color sensor camera 12 in each scanning of the A ₁ of the linear image sensor 201 to the C ₂ Right side A of fruits and vegetables 6
(Front side) and top B (top of waist) and right C
The image of the three surfaces (the side surface on the top side of the fruit) is separately arranged and photographed, and scanning is repeated by a scan start pulse (not shown) which is started every time the conveyor 1 advances by a predetermined unit size. The images of each surface of the fruits and vegetables 6 thus captured are formed as images as shown in FIG.

As described above, the signal (image pickup result) picked up by the color sensor camera 2 is analyzed by the arithmetic processing unit 5 as follows, and the grade and rank of the fruit and vegetable are determined.

Here, a description will be given of an example of an apple or fruit 6 having various colors and shapes depending on varieties as the fruit and vegetable 6 as an object for which the grade and the rank are determined.

I Class determination The class determination based on the size of the fruits and vegetables 6 is performed by using the image of the upper surface B at the center, calculating the area value, the body diameter dimension value, or the fruit height dimension value of the fruits and vegetables 6 from the upper surface image, and calculating the calculated value. Is compared with a size reference value preset in the class standard setting unit to determine the corresponding class.

II Grade judgment The inspection items for judging the grade are as shown below.

(A) Shape: Degree of deformation and circularity (b) Arozawa: Degree of coloring specific to each kind (c) Defects on the surface: Degree of defects such as pests, injuries, etc. The calculation is performed by the arithmetic processing unit 5 analyzing the three images (the waist part upper surface image, the cephalic part side surface image, and the apex part side surface image) captured by the sensor camera 2 as follows.

(A) Shape From the top surface image of the waist part, the degree of symmetry of the right and left shapes with the center line of the fruit and vegetable as the center line is calculated from the area (number of pixels) of the asymmetric part, and the degree of deformation is calculated. The grade is determined by comparing with a deformation degree reference value preset in the section. (See Fig. 12) From the side view image of the corneal part, the circularity (roundness) centered on the fruit core is calculated, and the circularity is set as a circularity criterion preset in the corneal part grade standard setting unit. Determine the grade by comparing with the value.
(See FIG. 13) From the image of the top side, the circularity of the top centered on the fruit core is calculated, and the circularity is compared with a circularity reference value preset in the top grade setting unit. To determine the grade.

(B) Arozawa The color value of each pixel is read from the image of each surface (the upper surface of the waist, the side of the stalk, and the side of the scalp) imaged by the color sensor camera 2, and for each image, Calculate the degree of coloring specific to the variety as Akizawa.

That is, the color information from green to red is decomposed into
A color value of ~ 126 is provided, the imaging resolution of the color sensor camera is 1 mm x 1 mm per pixel, and the color value of each pixel is read to create a pixel count histogram for each color value for each imaging surface. The center value of the histogram is calculated as the degree of coloring of each surface of the fruits and vegetables (the upper surface of the waist, the side surface of the fruit stalk, and the side surface of the fruit apex).

Then, the calculated degree of coloring of each surface, for each surface (that is, waist circumference grade standard setting section, fruit stem grade standard setting section,
The grade according to the coloring state of the fruits and vegetables is determined for each surface by comparing with a color reference value set in advance (by a top grade standard setting unit).

(C) Defects on the surface These defects on the surface are defects such as harmful fruits and bruises, bruises, branches, leaves, rust, etc. caused by pests and insects as shown in FIGS. 14 and 15. Since these are different colors from the surroundings,
The degree of defect is calculated for each image based on the discolored pixel area, and the degree of defect is determined for each surface by comparing each degree of defect with a reference value preset in each of the class setting units.

The grade determination result of each surface determined as described above is
In accordance with a chart diagram as shown in FIG. 10, a grade to sort the fruits and vegetables to be integrated is determined and graded, and the fruits and vegetables are classified based on the determination result based on the size of the fruits and vegetables.

The discharge device 101 of the corresponding sorting / discharging section 100 is operated by the signals of these grades and grades, so that the fruits and vegetables 6 can be automatically discharged from the conveyor 1 to the conveyor 7.

In the embodiment described above, (a) shape, (b) color, and (c) are test items for determining the grade of fruit and vegetables.
Although the description has been made by using the defect on the surface and performing the class determination based on the size of the fruits and vegetables, in the present invention, (b)
Is only an indispensable item, the shape of (a) and the defect on the surface of (c) are not indispensable items, and it is not essential to make a class judgment based on the size of fruits and vegetables, These may be adopted as necessary when carrying out the present invention. Further, although the embodiment has been described with reference to apples, the present invention is not limited to this, and may be fruits and vegetables of other different items.

[Effects of the Invention] As described above, the present invention captures three surfaces: the upper surface around the trunk of a fruit and vegetable, the side surface at the fruit stalk, and the side surface at the top of the fruit with a single color sensor camera, and obtains the three surfaces from the imaged results of the three surfaces. The degree of coloring is calculated for each surface of each surface, and the calculated degree of coloring of each surface is compared with a color reference value preset for each surface to determine a grade for each surface. Since the grades based on the coloring state of the fruits and vegetables are determined by integrating the grade determination results, even if the criteria for determining the grades based on the coloring states of the fruits and vegetables are different for each part such as the apex and the stalk, even if the fruits and vegetables are different, This has the effect of automatically and accurately determining the grade based on the coloring state.

[Brief description of the drawings]

Each of the drawings shows an embodiment of the present invention.
The figure is a plan view of a sorting and sorting apparatus implemented using the present invention,
FIG. 2 is a perspective view of a main part for explaining an imaging state, FIG. 3 is a plan view of the same, FIG. 4 is a cross-sectional view along AA of FIG. 3, and FIG. Fig. 6, Fig. 6 is a side view of the same, Fig. 7 is an explanatory view of an imaging procedure, Fig. 8 is an explanatory view of an imaging area of an image sensor, Fig. 9 is a view showing an imaging state of three surfaces, and Fig. 10 is a class. Class judgment chart, Fig. 11 shows excellent apple fruit,
FIGS. 12 to 15 are explanatory diagrams of defective fruits and vegetables. DESCRIPTION OF SYMBOLS 1 ... Conveyor 2 ... Color sensor camera 3 ... Mirror 4 ... Small lamp 5 ... Arithmetic processing unit 6 ... Fruits and vegetables 7 ... Drawer conveyor 11 ... Conveying surface 21 ... Scanning line 41 …… Lamp box 201 …… Linear image sensor, 100 …… Sorting and discharging unit 101 …… Discharge device

Continuation of the front page (56) References JP-A-49-55392 (JP, A) JP-A-56-108579 (JP, A)

Claims (2)

(57) [Claims] 1. A fruit or vegetable which is conveyed in a fixed posture in which the fruit stem side is directed to one side of the conveyor on the conveyor, and light from the upper surface of the illuminated vegetable is arranged above the conveyor. At the same time, the light from both sides of the fruits and vegetables reflected from the pair of reflecting means disposed on the left and right sides of the conveyor is guided to the sensors, and the fruits and vegetables are guided to the sensors. The three images of the upper surface and both side surfaces are individually arranged and imaged, so that the color sensor camera images three surfaces of the upper surface of the waist of the fruit and vegetables, the side surface of the fruit stalk, and the side surface of the fruit apex, and the three surfaces are imaged. Calculate the degree of coloring for each of the three surfaces from the imaging result of
The calculated degree of coloring of each surface is compared with a color reference value preset for each surface to determine the grade for each surface, and the grade determination result for each surface is integrated to determine the coloring state of the fruits and vegetables. A method for automatically judging the grade of fruits and vegetables, characterized by judging the grade of the fruits and vegetables. 2. A lighting device for illuminating fruits and vegetables conveyed in a fixed posture in which the fruit bud side is directed to one side of the conveyor on the left and right sides of the conveyor, and a lighting device arranged above the conveyor and illuminating the fruits and vegetables. A single color sensor camera for imaging the surface of the fruit and vegetable by imaging the surface of the fruit and vegetable on the sensor by the light of the light, and a pair of reflection means disposed on the left and right sides of the conveyor, An optical system for guiding light from both sides of the fruits and vegetables reflected from the upper surface and the reflection means to the sensor to form the three images of the upper surface and both sides of the fruits and vegetables on the sensor separately and form an image; An arithmetic processing unit for inputting an imaging result of the color sensor camera, wherein the color sensor camera images three surfaces of the upper surface of the waist, the side of the fruit stalk, and the side of the fruit apex of the fruits and vegetables; The processing device calculates the degree of coloring for each of the three surfaces from the imaging results of the three surfaces, compares the calculated degree of coloring of each surface with a color reference value preset for each surface, and calculates the degree of coloring for each surface. An automatic grade judging apparatus for fruits and vegetables, wherein grades are judged for each of the fruits and vegetables, and the grades according to the coloring state of the fruits and vegetables are determined by integrating the grade judgment results of the respective surfaces.