Banana ripening prediction device and banana ripening prediction method

Abstract

To predict a force-ripening state of a banana in a force-ripening chamber.SOLUTION: A banana force-ripening prediction device comprises: a hue determination unit 12a which determines a hue from a color image of a banana 3 in a force-ripening chamber 2; a temperature cumulative value calculation unit 11a which calculates a temperature cumulative value on the basis of a force-ripening schedule and a temperature in the force-ripening chamber 2; an average value calculation unit 12b which calculates an average value of an individual score of the banana 3; an average value determination unit 12c which determines whether the average value of the individual score exceeds a threshold; a data output unit 13 which outputs the temperature cumulative value in the force-ripening chamber 2 as explanatory variable data and the average value of the individual score as target variable data when the average value of the individual score exceeds the threshold; a force-ripening model generation unit 14 which calculates a and b by means of a maximum likelihood estimation method on the basis of the explanatory variable data and the target variable data in a logistic regression model:1 / (1+exp(-a(x-b))); and a force-ripening state prediction unit 15 which predicts a force-ripening state of the banana 3 by applying the calculated a and b to the logistic regression model.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a banana ripening prediction device and a banana ripening prediction method for predicting the ripening state of bananas stored in a ripening chamber. [Background technology]

[0002] Once harvested, bananas are stored in a ripening room and allowed to ripen for a week to 10 days to increase their sugar content before being shipped.

[0003] To explain further, as described in the conventional technology section of Patent Document 1 below, bananas harvested in the production area are not shipped to the market in their current state, but rather unripe green bananas at the time of harvest are stored in a ripening room before being put on the market in stores, and the temperature inside the ripening room is controlled to adjust the ripening state, and the bananas are shipped when they have reached the appropriate level of ripening.

[0004] In Patent Document 1 below, in order to know the ripening state of bananas in a ripening chamber, a measuring end that comes into contact with the banana skins is placed in a ripening processing chamber that has ventilation means and stores a large number of bananas, and a photometer equipped with monochromatic light generating means that generates monochromatic light having a wavelength that is absorbed by chlorophyll or carotenoids and photoelectric conversion means is placed in a position outside the processing chamber close to the processing chamber, and the measuring end is connected to the monochromatic light generating means and photoelectric conversion means by an optical transmission means, and the color state of the banana skins based on the measurement information obtained by the photometer is displayed on a display means placed in a control room located away from the processing chamber. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Publication No. 7-128226 Summary of the Invention [Problem to be solved by the invention]

[0006] However, in the conventional method disclosed in the above-mentioned Patent Document 1, the color of the banana skin is directly measured and displayed, so that the ripening state of the banana can be known from the outside based on the color of the banana skin, but the ripening state of the banana cannot be predicted.

[0007] Therefore, the present invention has been made in consideration of the above problems, and aims to provide a banana ripening prediction device and a banana ripening prediction method that can predict the ripening state of bananas in a ripening chamber. [Means for solving the problem]

[0008] In order to achieve the above object, the banana ripening prediction device described in claim 1 of the present invention is a banana ripening prediction device that predicts the ripening state of bananas ripening in a ripening chamber, A setting unit that performs initial settings including setting a ripening schedule, initializing the cumulative temperature value in the ripening chamber, and normalizing the color of the banana; a banana photographing camera for photographing the bananas in the ripening room by illuminating them; A thermometer for measuring the temperature inside the ripening chamber; a color determination unit that determines the color of the bananas in the ripening room from the color image captured by the banana photographing camera; a temperature accumulation value calculation unit that calculates a temperature accumulation value within the ripening chamber based on the ripening schedule and the temperature within the ripening chamber measured by the thermometer; an average value calculation unit that calculates an average value of the individual scores of the bananas in the ripening chamber based on the judgment result of the color judgment unit; an average value determination unit that determines whether the average value of the individual scores calculated by the average value calculation unit exceeds a threshold; a data output unit that outputs the cumulative temperature value in the ripening chamber as explanatory variable data and the average value of the individual scores as objective variable data when the average value determination unit determines that the average value of the individual scores exceeds a threshold value; A ripening model generation unit that calculates a and b of a logistic regression model: p=1 / (1+exp(-a(xb))) (where p is the average value of the individual score, x is the cumulative temperature value in the ripening chamber, a is the growth rate, and b is the inflection point) by maximum likelihood estimation based on the explanatory variable data and the objective variable data output by the data output unit; and a ripening state prediction unit that sets the average value of the individual scores or the cumulative temperature value inside the ripening chamber, and applies a and b calculated by the ripening model generation unit to the logistic regression model to predict the ripening state of the bananas.

[0009] The banana ripening prediction method according to claim 2 of the present invention is a banana ripening prediction method for predicting the ripening state of bananas ripening in a ripening chamber, A step of performing initial settings including setting a ripening schedule, initializing the cumulative temperature value in the ripening chamber, and normalizing the color of the bananas; Reading the temperature inside the ripening chamber; A step of illuminating and photographing the bananas in the ripening chamber; determining the color of the bananas in the ripening chamber from the captured color image; calculating a cumulative temperature value within the ripening chamber based on the ripening schedule and the temperature within the ripening chamber; Calculating an average value of the individual scores of the bananas based on the color determination results of the bananas in the ripening chamber; determining whether the average value of the individual scores exceeds a threshold; When it is determined that the average value of the individual scores exceeds the threshold, outputting the cumulative temperature value in the ripening chamber as explanatory variable data and the average value of the individual scores as objective variable data; Based on the explanatory variable data and the objective variable data, a step of calculating a and b of the logistic regression model: p = 1 / (1 + exp (-a (xb))) (where p: average value of individual score, x: cumulative temperature value in the ripening chamber, a: growth rate, b: inflection point) by maximum likelihood estimation method; and setting the average value of the individual scores or the cumulative temperature value in the ripening chamber, and applying the calculated a and b to the logistic regression model to predict the ripening state of the bananas. [Effects of the Invention]

[0010] According to the present invention, the ripening state of bananas in the first half of the ripening period can be used to predict the ripening state of bananas in the second half of the ripening period. This makes it possible to ship bananas at the ripening state desired by the shipping destination based on the predicted ripening state of bananas. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a block diagram showing a schematic configuration of a banana ripening prediction device according to the present invention. [Figure 2] 1 is a flowchart showing the steps of a banana ripening prediction method according to the present invention. [Figure 3] FIG. 10 is a diagram showing a specific example of the relationship between the cumulative temperature value in the ripening chamber and the average value of the individual scores of bananas in the ripening chamber when data from the entire ripening period is used. [Figure 4] FIG. 4 is a diagram showing a specific example of the relationship between the cumulative temperature value in the ripening chamber and the average individual score of bananas in the ripening chamber when the banana ripening prediction method according to the present invention predicts the latter half of the ripening period using data from the first half of the ripening period under the same conditions as in FIG. 3. DETAILED DESCRIPTION OF THE INVENTION

[0012] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[0013] As shown in Figure 1, the banana ripening prediction device 1 of this embodiment predicts the ripening state of bananas 3 stored in a ripening chamber 2 after harvest, and is roughly composed of various measuring instruments 4 that measure the environment inside the ripening chamber 2, an illuminator 5 that illuminates the bananas 3 stored in the ripening chamber 2, a banana photography camera 6 that photographs the bananas 3 in the ripening chamber 2 illuminated by the illuminator 5, a setting unit 7, a central processing unit 8, and a display unit 9.

[0014] The measuring meter 4 measures the physical quantities of the environment in the ripening chamber 2 at predetermined time intervals (for example, every hour), and is composed of a thermometer 4a, a hygrometer 4b, a CO2 meter 4c, and an ethylene concentration meter 4d.

[0015] The thermometer 4a is installed near the bananas 3 in the ripening chamber 2 and measures the temperature inside the ripening chamber 2, which is the most important physical quantity for ripening the bananas 3. The hygrometer 4b measures the humidity inside the ripening chamber 2. The CO2 meter 4c measures the CO2 concentration inside the ripening chamber 2. The ethylene concentration meter 4d measures the ethylene concentration inside the ripening chamber 2. The information measured at predetermined intervals (for example, 1 hour) (temperature, humidity, CO2 concentration, and ethylene concentration inside the ripening chamber 2) is sent sequentially to the central processing unit 8.

[0016] The setting unit 7 performs various settings, including initial settings, necessary for ripening the bananas 3 stored in the ripening chamber 2. The initial settings include setting a ripening schedule, initializing the cumulative temperature value in the ripening chamber 2, and normalizing the color of the bananas 3.

[0017] The ripening schedule consists of the ripening period (e.g., 7 to 10 days), the ethylene gas injection schedule (e.g., the first 24 hours at a concentration of 500 ppm or more), and the temperature schedule (e.g., 20°C (24 hours) → 18°C ​​(24 hours) → 16°C thereafter, 22°C (24 hours) → 20°C (24 hours) → 18°C ​​thereafter, etc.).

[0018] The cumulative temperature value in the ripening chamber 2 corresponds to the amount of heat given to the bananas 3, that is, temperature (temperature in the ripening chamber 2) x time (time of the temperature schedule).

[0019] The individual score of the banana 3 is determined by normalizing the color of the banana 3. Specifically, the color of the banana 3, which changes as it ripens, is divided into multiple color codes, and the color codes are quantified from 0.0 to 1.0. An example of a color code is the Dole (registered trademark) banana color chart. In this embodiment, the color of the banana 3 is divided into six levels: all green (color code 1), light green (color code 2), half green (color code 3), half yellow (color code 4), green tip (color code 5), and full yellow (color code 6), and the individual score is quantified as follows: color code 1 → 0.0, color code 2 → 0.2, color code 3 → 0.4, color code 4 → 0.6, color code 5 → 0.8, and color code 6 → 1.0. In other words, an individual score of 0 indicates that the banana is immature, and the closer the individual score is to 1, the more ripened the banana is.

[0020] The central processing unit 8 controls all parts when predicting the ripening state of the bananas 3 in the ripening chamber 2, and is composed of a ripening chamber environment measurement unit 11, a ripening state measurement and control unit 12, a data output unit 13, a ripening model generation unit 14, and a ripening state prediction unit 15.

[0021] The ripening chamber environment measurement unit 11 acquires information measured by the measuring meter 4 (temperature, humidity, CO2 concentration, ethylene concentration inside the ripening chamber 2) and measures the physical quantities of the environment inside the ripening chamber 2 as specific numerical values, and is equipped with a temperature cumulative value calculation unit 11a.

[0022] The temperature accumulation value calculation unit 11a calculates the temperature accumulation value in the ripening chamber 2 (temperature in the ripening chamber 2 x time of the temperature schedule) based on the temperature schedule of the initially set ripening schedule and the temperature in the ripening chamber 2 measured by the thermometer 4a of the measuring instrument 4.

[0023] The ripening state measurement and control unit 12 controls the on / off of the illuminator 5 and acquires color images of the bananas 3 taken by the banana photographing camera 6, and is equipped with a color judgment unit 12a, an average value calculation unit 12b, and an average value judgment unit 12c. In this embodiment, the illuminator 5 uses a light that emits white light.

[0024] The color determination unit 12a determines the color of each banana 3 in the ripening chamber 2 from a color image of the banana 3 illuminated by the illuminator 5, taken by the banana photographing camera 6. Specifically, the color determination unit 12a simultaneously photographs the banana 3 and the color code chart placed in the ripening chamber 2, and compares the color image of the banana 3 with the color image of the color code chart to determine for each banana 3 whether the color of the banana 3 in the ripening chamber 2 corresponds to all green, light green, half green, half yellow, green tip, or full yellow on the color code chart. In this embodiment, by simultaneously photographing the banana 3 and the color code chart in the ripening chamber 2, the influence of the color development by the illuminator 5 is eliminated.

[0025] The average value calculation unit 12b calculates the average value (0.0 to 1.0) of the individual scores of the bananas 3 in the ripening chamber 2 based on the judgment result of the color judgment unit 12a.

[0026] The average value determination unit 12c determines whether or not the average value of the individual scores of the bananas 3 calculated by the average value calculation unit 12b exceeds a preset threshold value (for example, 0.5).

[0027] When the average value determination unit 12c determines that the average value of the individual scores of bananas 3 exceeds the threshold value, the data output unit 13 outputs data using the cumulative temperature value in the ripening chamber 2 as explanatory variable data and the average value of the individual scores of bananas 3 as target variable data.

[0028] The ripening model generation unit 14 calculates a (growth rate) and b (inflection point) using the maximum likelihood estimation method based on the explanatory variable data and the objective variable data output by the data output unit 13 in the logistic regression model: p=1 / (1+exp(-a(xb))) (where p: average value of individual scores of bananas 3, x: cumulative temperature value in the ripening chamber 2, a: growth rate, b: inflection point).

[0029] The ripening state prediction unit 15 predicts the ripening state of the bananas 3 by applying a (increase rate) and b (inflection point) calculated by the ripening model generation unit 14 to a logistic regression model. Specifically, the ripening model generation unit 14 applies a (increase rate) and b (inflection point) calculated by the ripening model generation unit 14 to a logistic regression model to set an average value p of the individual scores of the bananas 3, and predicts the cumulative temperature value x in the ripening chamber 2 required to obtain the set average value p of the individual scores of the bananas 3. Alternatively, the ripening model generation unit 14 applies a (increase rate) and b (inflection point) calculated by the ripening model generation unit 14 to a logistic regression model to set an accumulated temperature value x in the ripening chamber 2, and predicts the average value p of the individual scores of the bananas 3 obtained at the set accumulated temperature value x in the ripening chamber 2.

[0030] The display unit 9 is composed of various displays such as an LCD display, and displays the setting screen, the ripening schedule, the information measured by the measuring meter 4 (temperature, humidity, CO2 concentration, ethylene concentration in the ripening chamber 2), and the prediction results.

[0031] Next, a method for predicting the ripening state of bananas 3 using the banana ripening prediction device 1 configured as described above will be described with reference to the flowchart of FIG.

[0032] A predetermined number of harvested bananas 3 are stored in the ripening chamber 2, and initial settings are performed (ST1). In the initial settings, a ripening schedule is set, the cumulative temperature value in the ripening chamber 2 is initialized, and the color of the bananas 3 is normalized.

[0033] After the initial setting is completed and the ripening of the bananas 3 starts according to the ripening schedule, the temperature inside the ripening chamber 2 is read by the thermometer 4a of the measuring meter 4 every predetermined time (for example, every hour) (ST2).

[0034] Next, the ripening state measurement and control unit 12 turns on the illuminator 5 to illuminate the bananas 3, and the banana photographing camera 6 photographs the bananas 3 at predetermined intervals (for example, every six hours). The photographed color images are sequentially sent to the color judgment unit 12a of the ripening state measurement and control unit 12. The color judgment unit 12a then judges the color of each banana 3 in the ripening chamber 2 from the color images photographed by the banana photographing camera 6 (ST3).

[0035] Further, the temperature accumulation value calculation unit 11a of the ripening chamber environment measurement unit 11 calculates the temperature accumulation value in the ripening chamber 2 (temperature in the ripening chamber 2×time of the temperature schedule) (ST4).

[0036] Then, the average value calculation unit 12b of the ripening state measurement / control unit 12 calculates the average value of the individual scores of the bananas 3 in the ripening chamber 2 based on the judgment result of the color judgment unit 12a (ST5).

[0037] Next, the average value determining unit 12c of the ripening state measurement / control unit 12 determines whether or not the average value of the individual scores of the bananas 3 calculated by the average value calculating unit 12b is equal to or greater than the threshold value (0.5) (ST6).

[0038] Then, if it is determined that the average value of the individual scores of banana 3 is greater than or equal to the threshold value (0.5) (ST6-Yes), the data output unit 13 outputs data with the cumulative temperature value in the ripening chamber 2 as the explanatory variable data and the average value of the individual scores of banana 3 as the objective variable data (ST7).

[0039] If it is determined that the average value of the individual scores of banana 3 is not equal to or greater than the threshold value (0.5) (ST6-No), the process returns to ST2, and the processes of ST2 to ST5 are repeatedly executed.

[0040] Next, the ripening model generation unit 14 calculates a (growth rate) and b (inflection point) using the maximum likelihood estimation method based on the explanatory variable data and target variable data output by the data output unit 13 in the logistic regression model: p=1 / (1+exp(-a(xb))) (ST8).

[0041] The maximum likelihood estimation method is a well-known method for point-estimating the parameters (a (growth rate), b (inflection point)) of the probability distribution that given data (explanatory variable data based on the cumulative temperature value in the ripening chamber 2, and dependent variable data based on the average value of the individual scores of bananas 3) follow.

[0042] The ripening state prediction unit 15 then applies a (rate of increase) and b (inflection point) to a logistic regression model to predict the ripening state (ST9). In predicting the ripening state, by applying a and b to the logistic regression model and determining the average value p of the individual scores of the bananas 3 or the cumulative temperature value x in the ripening chamber 2, it is possible to predict the cumulative temperature value x in the ripening chamber 2 required to obtain the average value of the individual scores p of the bananas 3, or to predict the average value p of the individual scores of the bananas 3 obtained at the cumulative temperature value x in the ripening chamber 2. In other words, by processing data up to the middle of the ripening period, it is possible to predict the time required to ripen the bananas 3 to a desired color code, or the color code of the bananas 3 obtained from the cumulative temperature value in the ripening chamber 2.

[0043] [Example of predicted ripening state] The ripening schedule was set as follows: ripening period: 7 days, ethylene gas injection schedule: at a specified concentration for the first 24 hours only, temperature schedule: 22°C (24 hours) → 20°C (24 hours) → 18°C ​​(5 days), and the bananas 3 in ripening chamber 2 were photographed every 6 hours to investigate the ripening state.

[0044] We then investigated the relationship between the cumulative temperature value in ripening chamber 2 (temperature in ripening chamber 2 x time of temperature schedule) and the average value of the individual scores of bananas 3 in ripening chamber 2. When data from the entire ripening period (7 days) was used, the results shown in Figure 3 were obtained. When the banana ripening prediction method based on the flowchart in Figure 2 was used to predict the second half (3 days) of the ripening period using data from the first half (4 days), the results shown in Figure 4 were obtained.

[0045] Comparing the two, the logistic regression model a (increase rate) is as follows: a[10 -3]=1.825, when predicted using the data from the first half of the ripening period in Figure 4: a[10 -3 ]=1.502, and although there was some discrepancy between the two, the predicted value captured the characteristics. Furthermore, regarding b (inflection point) of the logistic regression model, when all data from the ripening period in Figure 3 were used: b[10 3 ]=0.909, whereas when predicted using the data from the first half of the ripening period in Figure 4: b[10 3 ]=0.961, which means that the two results are almost identical.

[0046] In this way, according to the embodiment described above, it is possible to predict the ripening state of bananas in the latter half of the ripening period using data on the ripening state of bananas in the first half of the ripening period, which makes it possible to ship bananas at the ripening state desired by the shipping destination based on the predicted ripening state of bananas.

[0047] Although the best mode of the banana ripening prediction device and method according to the present invention has been described above, the present invention is not limited by the description and drawings of this mode. In other words, all other modes, examples, and operational techniques that can be made by those skilled in the art based on this mode are naturally included in the scope of the present invention. [Explanation of symbols]

[0048] 1. Banana ripening prediction device 2. Ripening Room 3 bananas 4 Measuring meter 4a thermometer 4b Hygrometer 4c CO2 meter 4d Ethylene concentration meter 5. Illuminators 6. Banana camera 7. Settings 8 Central Processing Unit 9 Display section 11. Ripening Room Environmental Measurement Section 11a Temperature accumulation value calculation section 12 Ripening state measurement and control unit 12a Color judgment section 12b Average value calculation section 12c Average value judgment section 13 Data output section 14. Ripening model generation section 15 Ripening state prediction section

Claims

1. A banana ripening prediction device that predicts the ripening state of bananas ripening in a ripening chamber, A setting unit that performs initial settings including setting a ripening schedule, initializing the cumulative temperature value in the ripening chamber, and normalizing the color of the banana; a banana photographing camera for photographing the bananas in the ripening room by illuminating them; A thermometer for measuring the temperature inside the ripening chamber; a color determination unit that determines the color of the bananas in the ripening room from the color image captured by the banana photographing camera; a temperature accumulation value calculation unit that calculates a temperature accumulation value within the ripening chamber based on the ripening schedule and the temperature within the ripening chamber measured by the thermometer; an average value calculation unit that calculates an average value of the individual scores of the bananas in the ripening chamber based on the judgment result of the color judgment unit; an average value determination unit that determines whether the average value of the individual scores calculated by the average value calculation unit exceeds a threshold; a data output unit that outputs the cumulative temperature value in the ripening chamber as explanatory variable data and the average value of the individual scores as objective variable data when the average value determination unit determines that the average value of the individual scores exceeds a threshold value; A ripening model generation unit that calculates a and b of a logistic regression model: p = 1 / (1 + exp (-a (x - b))) (where p is the average value of the individual scores, x is the cumulative temperature value in the ripening chamber, a is the growth rate, b is the inflection point) by maximum likelihood estimation based on the explanatory variable data and the objective variable data output by the data output unit; a ripening state prediction unit that sets the average value of the individual scores or the cumulative temperature value inside the ripening chamber, and applies a and b calculated by the ripening model generation unit to the logistic regression model to predict the ripening state of the bananas.

2. A banana ripening prediction method for predicting the ripening state of bananas ripened in a ripening chamber, comprising: A step of performing initial settings including setting a ripening schedule, initializing the cumulative temperature value in the ripening chamber, and normalizing the color of the bananas; Reading the temperature inside the ripening chamber; A step of illuminating and photographing the bananas in the ripening chamber; determining the color of the bananas in the ripening chamber from the captured color image; calculating a cumulative temperature value within the ripening chamber based on the ripening schedule and the temperature within the ripening chamber; Calculating an average value of the individual scores of the bananas based on the color determination results of the bananas in the ripening chamber; determining whether the average value of the individual scores exceeds a threshold; When it is determined that the average value of the individual scores exceeds the threshold, outputting the cumulative temperature value in the ripening chamber as explanatory variable data and the average value of the individual scores as objective variable data; A step of calculating a and b of a logistic regression model: p = 1 / (1 + exp (-a (x - b))) (where p is the average value of the individual score, x is the cumulative temperature value in the ripening chamber, a is the growth rate, b is the inflection point) by maximum likelihood estimation based on the explanatory variable data and the objective variable data; and setting an average value of the individual scores or a cumulative temperature value in the ripening chamber, and applying the calculated a and b to the logistic regression model to predict the ripening state of the bananas.

Images