JP2012117962A - Crop assessment system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a crop assessment system capable of easily and surely evaluating crops without gathering crops into a place for assessment.SOLUTION: A crop assessment system has imaging means 11 for capturing visual images of crops at a farmhouse 1 which is in a contractual relationship and evaluation means 20 which generates an evaluation value of the crops from the image data of the crops transmitted from the farmhouse 1. The evaluation means 20 has: first processing means 21 for normalizing the color tone of the image data; second processing means 22 for extracting a target region from the normalized image data; third processing means 23 for calculating the average values of each of RGB components in the extracted region; and fourth processing means 24 for assigning evaluation values to the calculated average values of each of the RGB components based on previously set ideal values of each of the RGB components.

Description

  The present invention relates to a crop determination system for remotely evaluating a crop harvested by a farmer in a contract relationship.

  Conventionally, as a determination system for judging the quality of fruits and vegetables, a light source having a wavelength in a predetermined range is irradiated onto a target object, transmitted light from the target object is detected, and the transmitted light is compared with a reference waveform to form a pattern. A system for determining the quality of an object based on whether or not they match is known (see Patent Document 1 below).

JP 07-318490 A

  However, the conventional determination system requires a light source, an irradiation device for the light, a condenser and an optical table for measuring transmitted light, and has a problem that the introduction of the system is large and the introduction cost increases.

  Furthermore, in the conventional judgment system, it is necessary to collect fruits and vegetables at an evaluation place (for example, a selection place) and perform judgment by the judgment system there. Applicable to supply chain management (stable supply system of high freshness and high value-added crops by supplying farm products directly from farmers who have a contract with the contractor to the seller's store, hereinafter referred to as SCM system) There was a problem that I could not.

  Then, an object of this invention is to provide the crop determination system which can evaluate a crop simply and reliably, without collecting a crop in an evaluation place.

The crop judgment system of the first invention is a crop judgment system for remotely evaluating a crop harvested by a farmer in a contract relationship,
Imaging means for capturing a visible image of the crop in the farmhouse;
Evaluation means for creating an evaluation value for the crop from the image data of the crop imaged by the imaging means and transmitted from the farmhouse,
The evaluation means includes
First processing means for standardizing the color tone of the image data;
Second processing means for cutting out a target area from the image data standardized by the first processing means;
Third processing means for calculating an average value of each component of RGB of the target region cut out by the second processing means;
And fourth processing means for assigning an evaluation value from an ideal value of each RGB component set in advance to the average value of each RGB component calculated by the third processing means.

  According to the crop judgment system of the first invention, the captured image of the crop harvested by the contracted farmhouse is transmitted to the evaluation means, and the evaluation value of the crop is given by the evaluation means. Therefore, since it is not necessary to collect the crops in one place, the present invention can also be applied to a stable supply system for high freshness and high added value crops such as direct delivery from production areas and SCM systems for crops.

  Further, the light source, the light irradiation device, the condenser and the optical table for measuring the transmitted light are not required, and the system itself is simple and the introduction cost is not increased.

  On the other hand, since the visible images captured by each contracted farmhouse are uneven in brightness and hue, it is difficult to determine them according to the same criterion. Normalize the color tone of the data, and the second processing means cuts out the target area, so that the reflection portion and the like can be removed, and the third and fourth processing means make a determination based on the same evaluation criteria based on the RGB components. It can be performed.

  Thereby, it is possible to easily and reliably evaluate the crop without collecting the crop at the evaluation place.

The crop determination system of the second invention is the first invention,
Based on part or all of the amino acid content of the destructive test that destroyed the crop, the brix value, and the taste value such as the umami richness value measured by the taste recognition device, the ideal value of each component of RGB is determined. An ideal value setting means for setting is provided.

  According to the crop judgment system of the second invention, based on part or all of the amino acid content of the crop by a destructive test, the brix value, and the taste value such as the umami richness value measured by the taste recognition device, By setting the ideal value of each component of RGB, it is possible to set the ideal value of each component of RGB that is optimal from the type of crop, the harvesting time, the weather conditions at the time of growth, etc. Based on the ideal value, it is possible to simply and reliably evaluate the crop without collecting the crop at the evaluation place.

The crop determination system of the third invention is the first or second invention,
Evaluation transmission means for transmitting the evaluation value created by the evaluation means together with the image data of the farm product to the farmhouse is provided.

  According to the crop determination system of the third invention, the evaluation value created by the evaluation means is transmitted to the farmhouse as one piece with the image data of the crop, so that the crop is in the contractual relationship without being collected at the evaluation place. It is possible to make the farmer recognize what kind of production the farm has produced has a high evaluation value.

The perspective view which shows the whole structure of a crop determination system. The flowchart which shows the processing content in a crop determination system. Explanatory drawing which shows the processing content in a crop determination system. Explanatory drawing which shows the determination content in a crop determination system.

  1 is a system for remotely evaluating a crop harvested by a farmer 1 who has a contractual relationship, and an imaging means 11 provided on the farmer 1 side and a network. And a controller 2 provided at a position spatially separated from the farm 1.

  The imaging means 11 is a visible camera that captures a visible image, and may be a general digital camera or a visible camera mounted on a mobile phone 11 ′.

  The controller 2 includes an evaluation unit 20, an ideal value setting unit 25, an evaluation database 26, and a history database 27.

  The evaluation unit 20 is a unit that creates an evaluation value for the crop from the image data of the crop captured by the imaging unit 11 and transmitted from the farm 1, and includes a first processing unit 21, a second processing unit 22, Third processing means 23 and fourth processing means 24 are provided.

  The first processing means 21 executes a first process for normalizing the color tone of the image data.

  The second processing means 22 executes a second process for cutting out the target area from the image data standardized by the first processing means 21.

  The third processing unit 23 calculates an average value of the RGB components of the target area cut out by the second processing unit 22.

  The fourth processing unit 24 assigns an evaluation value from the preset ideal values of the RGB components to the average values of the RGB components calculated by the third processing unit 23.

  Details of the processing contents of the first processing means 21 to the fourth processing means 24 will be described later.

  The ideal value setting means 25 includes an amino acid content in a destructive test in which various crops (for example, cucumbers, tomatoes, spinach, apples, etc.) are destroyed in advance, a brix value, an umami richness value measured by a taste recognition device, and the like. The ideal value of each RGB component of the crop is set based on the correlation between part or all of the taste value and the average value of each RGB component.

  Specifically, for each crop, for example, the relationship between the amino acid content in the destructive test, the brix value, the taste value such as the umami richness value measured by the taste recognition device, and the average value of each RGB component is obtained. Alternatively, a reference value is determined from all, and the ideal value of each component of RGB is determined by multivariate analysis based on the reference value.

  Note that, instead of or in addition to setting the ideal values of the RGB components based on the measurement values obtained by the taste recognition device, the ideal values may be determined or corrected based on the inspection results obtained by sensory inspection or the like.

  The evaluation database 26 includes an amino acid content of a destructive test in which various crops (for example, cucumbers, tomatoes, etc.) are destroyed, a brix value, a taste value such as a delicious taste value measured by a taste recognition device, and RGB of the crop. Storage means for storing past data (including weather conditions, harvest time, etc.) of ideal values of the respective components.

  The history database 27 is a storage unit that stores a history of evaluation values (including additional information such as weather conditions and harvest time) of various crops evaluated by the evaluation unit 20.

  In the present embodiment, the processing units 21 to 27 constituting the controller 2 are each configured by hardware such as a CPU, a ROM, and a RAM, and these processing units 21 to 27 are each configured by common hardware. These may be partly or entirely configured by different hardware.

  The imaging unit 11 and the controller 2 are connected by a network such as the Internet via a personal computer 12 that reads image data from the imaging unit 11. Further, in the case of the mobile phone 11 ′ provided with the image pickup means, it is connected to the controller 2 through a network via the base station 13 by wireless communication.

  Next, with reference to FIG. 2, a method for evaluating a crop harvested by a farmer having a contractual relationship with the crop determination system configured as described above will be described.

  First, when the image data picked up by the image pickup means 11, 11 ′ is sent, the controller 2 recognizes the type of crop of the image data from the additional information attached to the image data (FIG. 2 / STEP 11).

  Here, the additional information attached to the image data is information that is described when the farmer 1 sends the image data to the controller 2, and includes, for example, the type of crop of the image data, the farming method (organic farming method), and the harvest date and time. Information such as harvest conditions (tree ripening, early ripening, etc.).

  In addition, without attaching additional information to image data, based on only the image data, the type of the crop may be determined and recognized by pattern matching or the like from the shape of the crop of the image data.

  Next, the controller 2 reads out the ideal value of each component of RGB corresponding to the type of crop recognized in STEP 11 from the evaluation DB 26 (FIG. 2 / STEP 12).

  Here, the read ideal values of the RGB components may be used as they are for the evaluation of the image data by the evaluation means 20, but may be corrected by the user. For example, by correcting the ideal value in consideration of the production location, the weather conditions of the year, etc., it is possible to set a more optimal ideal value and evaluate it according to reality.

  Next, the first processing means 21 of the evaluation means 20 executes a first process for normalizing the color tone of the image data (FIG. 2 / STEP 21).

  Specifically, as shown in FIG. 3 (a), the entire image is captured so that the background color (black) of the captured image of the crop originally captured under the black unified background is equal to the reference value. Standardize by correcting the color tone.

  In order to improve evaluation accuracy, the luminance of the entire image may be corrected and normalized so that the luminance of the background color (black) of the captured image of the crop is equal to the reference value.

  Next, the second processing means 22 of the evaluation means 20 executes a second process for cutting out the target area from the image data (FIG. 2 / STEP 22).

  Specifically, as indicated by hatching in FIG. 3B, an area having the same color tone other than the background color (black) is extracted from the image data as an area of the crop (cucumber) that is the object. At this time, in the extracted region of the same color tone, a predetermined range of color tone may be extracted as an area of the same color tone from the average value of the color tone of the region other than the background color. You may make it designate the color tone of a predetermined predetermined range.

  As a result, the target area can be cut out from the target area except for the high-luminance area due to reflection or the like and the low-luminance area that is shaded.

  In addition to or instead of the processing by the second processing means 22, the user may manually specify the target area and cut out the target area.

  Next, the third processing unit 23 of the evaluation unit 20 calculates an RGB average value that is an average value of the RGB components of the cut out target region (FIG. 2 / STEP 23).

  Here, the third processing means 23 uses the RGB value of each dot in this area as an RGB average value of the target area cut out in STEP 23 (R, G, and B each have an intensity of 0 or more and 255 or less) The average value of is calculated.

  Next, the fourth processing unit 24 gives an evaluation value based on the ideal value of each component of RGB read in STEP 12 to the RGB average value calculated in STEP 23 (FIG. 2 / STEP 24).

  Specifically, as shown in FIG. 4, the RGB average value calculated in STEP 23 shows a distribution different from the ideal value depending on the production area, production method, harvest time, etc., even for the same crop.

  In FIG. 4, the ideal value is 170 for the R component (about 640 nm) and 170 for the G component (about 540 nm), whereas the average RGB value is 140 for the G component (about 540 nm). In actual cucumbers having different values, the RGB average value does not become lower than the ideal value, and deviates from the ideal value in the order of A to D.

  In the ideal component B (about 440 nm), the RGB average value is 220, whereas A is close to the ideal value, but C and D do not have the RGB average value as high as the ideal value. It is off.

  Therefore, for example, the fourth processing unit 24 sets 100 points when the average value is equal to the ideal value for each component of RGB, 80 points within a predetermined first range from the ideal value, and further, the first value outside the first range. The overall evaluation of the three components (maximum of 300 points) is performed with 60 points in the 2 range, 40 points in the third range outside the second range, and 0 points in the other range.

  Note that this evaluation value is an example, and depending on the type of agricultural product, one of the three RGB components may be weighted for evaluation, or the evaluation may be performed based on one of the components.

  Next, the controller 2 stores the evaluation value given by the fourth processing unit 24 in the history database 27 (FIG. 2 / STEP 31). At this time, the controller 2 integrally stores additional information (such as the type of crop, farming method (organic farming method), harvesting date and time, harvesting condition) attached to the image data transmitted from the farmer 1 together with the evaluation value. .

  As a result, the evaluation information stored in the history database 27 can be statistically processed by dividing the evaluation information stored in the history database 27 into the type of crop, the farming method, the harvesting time, and the harvesting conditions, thereby producing a crop with higher added value. You can get useful materials for.

  Next, the controller 2 transmits the evaluation value given in STEP 24 together with the corresponding image data to the farm 1 and ends the series of processes (FIG. 2 / STEP 32). Thereby, it is possible to make the farmer 1 who has the contractual relationship recognize the high evaluation value of the produced crops without collecting the crops at the evaluation place.

  The above is the details of the method for evaluating the crops harvested by the farmer 1 who has a contractual relationship with the crop determination system of the present embodiment, and according to the crop determination system, it is not necessary to collect the crops in one place. The present invention can also be applied to a stable supply system for high freshness and high value-added crops such as direct delivery from production areas and SCM systems for crops, and can easily and reliably evaluate crops.

  In addition, in this embodiment, although the case where cucumber was evaluated as an agricultural product was demonstrated, it is not limited to this, In addition to various fruits and vegetables, such as a spinach, a tomato, and an apple, plants, such as a flower and a foliage plant You may apply to.

DESCRIPTION OF SYMBOLS 1 ... Farmer, 2 ... Controller, 11 ... Imaging means, 11 '... Mobile phone, 20 ... Evaluation means, 21 ... 1st evaluation means, 22 ... 2nd evaluation means, 23 ... 3rd evaluation means, 24 ... 4th evaluation Means 25 ... Ideal value setting means 26 ... Evaluation database 27 ... History database

Claims (3)

A crop judgment system for remotely evaluating crops harvested by contracted farmers,
Imaging means for capturing a visible image of the crop in the farmhouse;
Evaluation means for creating an evaluation value for the crop from the image data of the crop imaged by the imaging means and transmitted from the farmhouse,
The evaluation means includes
First processing means for standardizing the color tone of the image data;
Second processing means for cutting out a target area from the image data standardized by the first processing means;
Third processing means for calculating an average value of each component of RGB of the target region cut out by the second processing means;
And a fourth processing unit for assigning an evaluation value from an ideal value of each RGB component set in advance to an average value of each RGB component calculated by the third processing unit. Judgment system. The crop determination system according to claim 1,
Based on part or all of the amino acid content of the destructive test that destroyed the crop, the brix value, and the taste value such as the umami richness value measured by the taste recognition device, the ideal value of each component of RGB is determined. An agricultural product determination system comprising ideal value setting means for setting. In the crop judgment system according to claim 1 or 2,
An agricultural product determination system comprising: an evaluation transmission unit that transmits the evaluation value created by the evaluation unit to the farmhouse as an integral part of image data of the agricultural product.

Images