KR101324338B1 - A compact quality sorting method and apparatus for cherry tomatoes based on real-time color image processing - Google Patents

Abstract

PURPOSE: A method for sorting the quality of cherry tomatoes in real time by using color images and a device thereof are provided to measure a coloration degree in real time by developing an algorithm for rapidly measuring and determining the coloration degree in an installed transfer device. CONSTITUTION: A method for sorting the quality of cherry tomatoes in real time by using color images comprises the following steps: arraying the cherry tomatoes and transferring them (S10); determining a coloration degree in a multi-step manner without interrupting the cherry tomatoes being transferred (S20); and classifying the cherry tomatoes according to determined coloration degrees (S30). The S20 step comprises the following steps: obtaining image data for analyzing images (S21); measuring a data value (S22); changing a color-difference meter after removing the noise (S23); applying a mask image by removing backgrounds (S24); obtaining an a* value, a b* value, and pixel data (S25); and determining the classification and class according to a boundary equation (S26). [Reference numerals] (AA) Remove noise; (BB) Remove a background; (S10) Step of arraying and transferring; (S21) Acquire image data; (S22) Measure a data value; (S23) Change a color-difference meter (RAB→LAB); (S24) Apply a mask image; (S25) Obtain an a* value, a b* value, and pixel data; (S26) Determine the classification and class according to a boundary equation; (S30) Step of classifying cherry tomatoes according to a determined coloring grade

Description

A compact quality sorting method and apparatus for cherry tomatoes based on real-time color image processing}

The present invention relates to a cherry tomato real-time quality screening method using a color image and the device, and more particularly, to develop an algorithm that can quickly determine the color measurement of cherry tomatoes through color image acquisition and real-time image processing, The present invention relates to a cherry tomato real-time quality sorting method and apparatus using color images that can be applied to image analysis equipment commonly adopted in general agricultural non-destructive sorting technology, thereby reducing production costs and ensuring the efficiency of maintenance.

Most domestically produced cherry tomatoes are exported to major Asian countries such as Japan, Hong Kong, etc., and it is important to produce products that meet the quality and safety requirements of the target countries. However, in the case of cherry tomatoes, research is relatively insufficient due to the difficulty in developing high-quality screening technology because of the higher quantity and faster screening compared to large tomatoes.

The quality factors of cherry tomatoes are various parts such as coloration, shape, size, heat, presence, sugar, and acidity, but as known in Korean Utility Model Publication No. 20-0242431 and Korean Utility Model Publication No. 20-0218218 The selection of cherry tomatoes is dependent on simple size selection by drum or line sorter, and the technical basis for producing high quality agricultural products is not sufficiently secured.

Recently, the applicant has analyzed the difference in the fluorescence characteristics of the skin and pulp of the tomato by using an electron multiplying chargecoupled device (EMCDC) camera to analyze the heat and parts that could not be detected well by a general CCD camera using a fluorescence image processing technique. By developing a method for detecting tomato fever (Korean Patent Publication No. 10-2012-0036467, a method for detecting cherry tomatoes using fluorescence image), the technical basis for producing high-quality agricultural products has been established, but Still, the technical foundation for producing high quality agricultural products is not sufficiently secured.

In particular, the degree of coloring of cherry tomatoes is a factor influencing the purchasing decision of the consumer. Since the coloration of cherry tomatoes shows maturity that changes with time, the coloration should be selected in consideration of the expiration date depending on the distance of the target country. However, at present, exports are made only in the red phase, and considering the quarantine waiting time in the exporting country and the movement time of goods in the exporting country, the product value period of cherry tomatoes is shortened and quality deterioration is incurred.

As shown in FIG. 1, the coloration of cherry tomatoes is currently only selected by the naked eye, and visual selection is not quantitative because the selector's fatigue level or individual criteria are different. Since cherry tomatoes are small in size and have a large quantity to be processed, they may provide a cause for a large labor force and a corresponding increase in production cost for visual selection.

The present invention has been made to solve the above problems, to build an on-line transfer device capable of high-speed transfer of cherry tomatoes using the image analysis equipment generally adopted in the image analysis device, and the degree of coloring in the constructed transfer device It is an object of the present invention to provide a method and apparatus for real-time cherry tomato quality screening using color images that can be used to colorize screening in real time.

Cherry tomato real-time quality screening method using the color image of the present invention for achieving the above object comprises the steps of aligning and transporting cherry tomato; Determining the degree of coloring in multiple stages without interrupting the transferred cherry tomatoes; And classifying cherry tomatoes according to the determined degree of coloring, and determining the coloring degree in multiple steps without stopping the cherry tomatoes may include obtaining image data for image analysis, measuring the data values, and removing noise. Changing the color gamut afterwards, applying a mask image by removing the background, acquiring a *, b * values and pixel data, and determining classification and rating according to a boundary equation. It is done.

In addition, the step of determining the classification and class according to the boundary equation is characterized in that can be determined up to six steps using the following boundary equation.

[equation]

In addition, the step of aligning and transporting the cherry tomatoes is characterized in that the left and right belts inclined downward in the V-shape toward the center portion and by varying the speed of the left and right belts tangled cherry tomatoes in a row in a row.

In the cherry tomato real-time quality sorting device using the color image of the present invention for achieving the above object in the cherry tomato real-time quality sorting device using a color image to sort by the degree of coloring by receiving the cherry tomatoes,

The sorting apparatus includes a supply unit, a sorting unit and a sorting unit, wherein the sorting unit determines the coloring degree in multiple stages without stopping the drop tomato being transferred; And classifying cherry tomatoes according to the determined degree of coloring, and determining the coloring degree in multiple steps without stopping the cherry tomatoes may include obtaining image data for image analysis, measuring the data values, and removing noise. Changing the color gamut afterwards, applying a mask image by removing the background, acquiring a *, b * values and pixel data, and determining classification and rating according to a boundary equation. It is done.

In addition, the step of determining the classification and class according to the boundary equation is characterized in that can be determined up to six steps using the following boundary equation.

[equation]

In addition, the supply unit is a hopper for supplying the cherry tomatoes, the supply conveyor is connected to the hopper to put the drop tomatoes contained in the hopper on the same plane on the feed conveyor, and the left and right belts to transfer the cherry tomatoes in a row from the feed conveyor To the center of the downward inclined in the form of V and the left and right belts to feed the different feed, the alignment feeder is hinged in a plate shape at the bottom of the alignment feeder, and the tomatoes according to the maturity group determined according to the color grade determined in the sorting unit It characterized in that it comprises a drop guide for guiding the classification.

Real time quality screening method of cherry tomatoes using the color image according to the present invention and the image is used in color imaging technology that is a conventional agricultural non-destructive screening technology for image acquisition and real-time image processing for the rapid measurement step by step coloration of cherry tomatoes By operating under the general CCD camera and the Windows operating system, which is an analytical device, it is possible to reduce production costs and maintain efficiency. In addition, since the degree of coloring can be set in multiple stages, there is an advantage in that it is possible to increase farm imports by increasing user convenience and diversifying export products. In addition, since the coloration of cherry tomatoes, which are currently performed by hand, can be automated at a high speed, there is an advantage of reducing agricultural labor and production costs.

1 is an exemplary view according to the conventional visual selection.
2 is a flow chart schematically showing a method for real-time cherry tomatoes quality screening using color image according to the present invention.
3 is a schematic diagram of an alignment feeder according to the present invention;
4 is a block diagram of an apparatus for sorting according to the present invention;
5 is a program screen in a personal computer according to the present invention
6 is an image processing procedure for classifying cherry tomatoes according to the present invention.
7 is a result generation image for the real-time detection of the cherry tomato surface according to the present invention
8 is a six-stage classification diagram applying the boundary equation according to the present invention
9 is a six-stage classification color of cherry tomatoes according to the present invention
10 is a device configuration diagram according to an embodiment of the present invention
11 is a block diagram of an apparatus according to an embodiment of the present invention.
12 is an enlarged front view of the alignment feeder according to the present invention.
13 is a schematic view of the operation of the drop guide according to the present invention
14 is a schematic view of the sorting device according to the present invention

First, with reference to the accompanying Figures 2 to 9 will be described in detail with respect to the cherry tomato real-time quality screening method using a color image according to an embodiment of the present invention.

Here, cherry tomatoes are not limited to bovine tomatoes such as jujube tomatoes and cherry tomatoes.

In addition, in adding the reference numerals to the components of the referenced drawings, the same components are to have the same reference numerals as possible, even if displayed on different drawings, it is determined that may unnecessarily obscure the subject matter of the present invention Detailed descriptions of well-known functions and configurations will be omitted. Also, directional terms such as "top", "bottom", "front", "back", "front", "forward", "rear", etc. are used in connection with the orientation of the disclosed drawing (s). Since the elements of the embodiments of the present invention can be positioned in various orientations, the directional terminology is used for illustrative purposes, not limitation.

Referring to Figure 2, the cherry tomato real-time quality screening method using the color image related to the present invention in the step of sorting and transporting cherry tomatoes (S10), the step of determining the coloring degree in multiple stages without interrupting the transferred cherry tomatoes (S20) And classifying cherry tomatoes according to the determined degree of coloring (S30).

In the step (S10) of delivering the cherry tomatoes, the cherry tomatoes have an irregular shape having different shapes and sizes, so that the bridge phenomenon, which is a phenomenon in which cherry tomatoes are connected or overlapped, should not occur. For this purpose, a vibration feeder system that is capable of providing alignment while vibrating is generally adopted. However, this is not appropriate because there is a risk of necrosis of the skin tissue of cherry tomatoes with physical properties due to continuous vibration. As such, it is preferable that the alignment method for cherry tomatoes be designed in such a way that the image processing can be performed individually by being supplied in a line method.

Referring to FIG. 3, in the present invention, the conveying belt constituting the alignment feeder 120 is formed of left and right conveying belts 121 and 122 and is formed to be inclined downward in a V shape toward the center, and each belt speed is Each is controlled by two motors (not shown). In order to prevent the bridge phenomenon to separately shoot the cherry tomatoes transported from the selection unit to control the speed of the left and right conveying belts 121, 122 configured in a V through the motor differently. As a result, a large amount of cherry tomatoes dropped from the feed conveyor 110 to the alignment feeder 120 are transported while being aligned in a row while rotating during the transfer from the alignment feeder 120.

 The step S20 of determining the degree of coloring without interrupting the transferred cherry tomatoes may include obtaining image data for image analysis (S21), measuring the data value (S22), and changing the colorimeter after removing noise. (S23), applying a mask image by removing a background (S24), acquiring a * and b * values and pixel data (S25), and determining classification and grade according to a boundary equation (S26). ).

Referring to Figure 4 and Table 1, the image data acquisition step (S21) is a color CCD camera, which is an image analysis equipment adopted in a general image analysis system to reduce the production cost and maintenance costs according to the device configuration, Apply frame grabber and fluorescent lamp.

<Table 1> Composition of real-time video recording device

Referring to FIG. 5, in operation S22, the image data transmitted to a personal computer (a personal computer using a Windows operating system) through a frame grabber is opened in real time using OpenCV. By applying the GUI program using visual studio to process the image, the size of the cherry tomatoes and the R, G, B color difference data values are stored in real time based on the acquired image. do.

6 and 7, in the step of changing the color difference meter after removing the noise (S23), a general R, G, and B color difference obtained from the CCD camera in the image data acquisition step and data data in the step of measuring the data value After the noise is removed, the system data can be classified into L, a *, and b * color difference data so as to be classified into multi-levels according to the degree of coloring.

In the step S24 of applying the mask image by removing the background, the mask image is generated by removing the background of the cherry tomato and the top of the cherry tomato so as to detect only the epidermis of the cherry tomato.

In operation S25, the a * and b * values and pixel data are acquired based on a mask image, L, a * and b * color difference pixel data of the epidermal portion of cherry tomatoes are obtained.

In the step of classifying and classifying according to the boundary equation (S26), the screen is actually selected by applying to the boundary equation using an average value of pixels of a * and b * channels among the obtained L, a *, b * color difference pixel data. do. In general, the characteristic of La * b * channel is L, the brightness of the image, a * is red to green, b * is yellow to blue, so each drop using a * and b * channels except L channel is used. Apply a linear boundary equation to distinguish groups of tomato ripeness.

Here, the maturation group is a selection criteria according to the degree of coloration of cherry tomatoes, and it is possible to select and group the coloring degree for determining the release of cherry tomatoes in multiple stages (largely three and six stages) for export and domestic sales. Of course, the group can be further subdivided into groups.

For reference, in relation to the coloration of cherry tomatoes, the coloration of cherry tomatoes is not yet established in Korea, and according to the US Department of Agriculture (USDA, 1991), the coloration of cherry tomatoes is described with reference to <Table 2>. Green (GN), Color (Breaker, BK), Coloring (Turning, TN), Paint (PK), Light red (LR), Deep red Divide by (Red, hereinafter RD).

Table 2: US Department of Agriculture Standard Cherry Tomato Coloring Table

8 and 9 and <Table 3>, the linear boundary equation may be applied in up to six steps as follows. As a result, the degree of coloration is selected in six stages of rust, GN, color, BR, color, PK, light red, LR, and deep red. can do. Note that the user can further simplify the maturity group as needed.

<Table 3> Accuracy of calibration and verification of classification

In the step S30 of classifying cherry tomatoes according to the determined degree of coloring, the cherry tomatoes belonging to the maturity groups classified in stages in the step S26 of determining classification and grade according to the boundary equation are seated in the storage group of the same group. It includes a control unit (not shown) for controlling the ejection angle of the drop guide 130 to be made to be.

Next, the cherry tomato real-time quality screening apparatus using the color image according to an embodiment of the present invention will be described in detail with reference to FIGS. 10 to 14.

Referring to FIG. 10, the cherry tomato real-time quality sorting apparatus using the image related to the present invention includes a supply unit 100, a sorting unit 200, and a sorting unit 300.

The supply part 100 includes a supply conveyor 110, an alignment feeder 120, and a guide part 130.

3 and 10, the feed conveyor 110 is a conventional flatbed conveyor, and before the transfer of droplets of the tomato to be transported from the hopper (not shown) to the sorting feeder 120 is intertwined with each other The bridge phenomenon, which is a continuous or overlapping phenomenon, is primarily eliminated and placed on the same plane on the supply conveyor. Referring to FIG. 11, the supply conveyor 110 may be formed as a roller conveyor. In this case, each roller conveyor may be planted with a roller conveyor to remove foreign substances adhering to the surface of cherry tomatoes. Driven by a motor (not shown) and rotated, the cherry tomatoes loaded thereon are conveyed toward the downstream alignment feeder 120. In addition, the plate-shaped guide portion 112 is disposed at the lower end of the supply conveyor 110, and the drop tomatoes transported along the supply conveyor 110 may fall down to the alignment feeder 120 along the guide portion 112, and fall down. Skin tissue due to necrosis can be prevented. In addition, the side plate 111 may be added to prevent the cherry tomatoes fall out of the conveyor (110).

3 and 12, the alignment feeder 120 is composed of left and right conveying belts 121, 122, which is formed to be inclined downward in the V shape toward the center portion, each belt speed is two motors (not shown) Are controlled respectively. In order to prevent the bridging phenomenon, each of the cherry tomatoes transported by the sorting unit 200 can be individually photographed by varying the speeds of the left and right conveying belts 121 and 122 having a V shape, while a large amount of cherry tomatoes are rotated during transportation. It can be transported while being aligned in a row. The alignment feeder 120 may be inclined downward from the supply part 100 toward the sorting part 300. In addition, the alignment feeder 120 is configured in two rows as shown in the figure, but may be configured in a single row or a double row in consideration of the supply amount.

Referring to FIG. 13, the drop guide 130 is hinged in the shape of a plate at the bottom of the alignment feeder 120 to classify the waterproof tomato according to the maturation groups determined according to the color grade determined by the sorting unit 200. It serves as a guide.

To this end, the drop guide 130 is hinged to the top of the alignment feeder 120, the bottom is a hydraulic cylinder (wire or chain 132 formed on the bottom of the alignment feeder 120 by the roller or sprocket 133) 131 is connected. The hydraulic cylinder 131 has a stretch length according to the color grade transmitted from the sorting unit 200 by the control unit (not shown). As a result, the length of the wire or chain 132 is variable so that the lower end of the drop guide 130 is raised and lowered. As the drop guide 130 descends, the droplet tomato is guided to the classification unit 300 for each ripening group.

14, in the configuration of the sorting unit 200, in order to image the cherry tomatoes individually, the background for efficient associative processing should be set and lighted. Since cherry tomatoes produce atypical irregular movement, four 25W fluorescent lamps 220 are placed at the top of the case 210 so that the light is uniformly illuminated on the subject so that the background and the object can be clearly distinguished. Place 230.

Referring to FIG. 13 again, the classification unit 300 may be formed in the form of a storage box. In FIG. 13, the classification unit 300 may be divided into two stages for each set maturity group. In addition, in partitioning the sorter 130 in the form of a storage box, the partition width is set in consideration of the feed rate of the alignment feeder 120 and the lifting width of the drop guide 130.

Although the present invention has been described in detail with reference to the drawings, the present invention is not limited thereto.

100:
110: supply conveyor 120: alignment feeder 130: drop guide

200:
210: case 220: fluorescent lamp 230: CCD camera
300: classification unit

Claims (6)

Sorting and transporting the cherry tomatoes (S10); Determining the degree of coloring in multiple stages without interrupting the transferred cherry tomatoes (S20); And classifying cherry tomatoes according to the determined degree of coloring (S30),
The step S20 of determining the coloring degree without stopping the cherry tomato may include obtaining image data for image analysis (S21); Measuring the data value (S22); Changing the colorimeter after noise removal (S23); Applying a mask image by removing the background (S24); acquiring a *, b * values and pixel data (S25); And determining the classification and the grade according to the boundary equation (S26),
The classifying and classifying according to the boundary equation may be determined up to six levels using the following boundary equation.
Real-time quality screening method of cherry tomato using color image.
[equation]

The method of claim 1,
Aligning and transporting the cherry tomatoes (S10) is characterized in that the left and right belts to be inclined downward in the V shape toward the center of the alignment feeder, and by moving the scrambled cherry tomatoes in a row by varying the speed of the left and right belts in a row
Real-time quality screening method of cherry tomato using color image.
In the cherry tomato real-time quality sorting device using a color image to receive the cherry tomatoes to sort by the degree of coloring,
The sorting apparatus includes a supply unit, a sorting unit and a sorting unit, wherein the sorting unit acquires image data for image analysis (S21); Measuring the data value (S22); Changing the colorimeter after noise removal (S23); Applying a mask image by removing the background (S24); acquiring a *, b * values and pixel data (S25); And determining the classification and the grade according to the boundary equation (S26),
The classifying and classifying according to the boundary equation may be determined up to six levels using the following boundary equation.
Cherry tomato real-time quality screening device using color image.
[equation]

The method of claim 3,
The supply unit is a hopper for supplying cherry tomatoes, a supply conveyor connected to the hopper so that the drop tomatoes contained in the hopper on the same plane on the feed conveyor, and the left and right belts to transfer the cherry tomatoes in a row from the feed conveyor Aligned feeder which is inclined downward toward the center of the alignment feeder in a V-shape and is transferred at different speeds of the left and right belts, and is connected to the end of the alignment feeder in a plate shape, and then dropped into tomatoes according to maturity groups determined according to the coloring grades determined by the sorting unit. Characterized in that it comprises a drop guide for guiding the sorting unit
Cherry tomato real-time quality screening device using color image.
delete delete

Images