CN115736298A - Image reasoning model-based tobacco primary baking control method - Google Patents
Image reasoning model-based tobacco primary baking control method Download PDFInfo
- Publication number
- CN115736298A CN115736298A CN202211608863.8A CN202211608863A CN115736298A CN 115736298 A CN115736298 A CN 115736298A CN 202211608863 A CN202211608863 A CN 202211608863A CN 115736298 A CN115736298 A CN 115736298A
- Authority
- CN
- China
- Prior art keywords
- dry
- proportion
- baking
- yellow
- bulb temperature
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Manufacture Of Tobacco Products (AREA)
Abstract
The invention provides a tobacco primary curing and baking control method based on an image inference model, which comprises the steps of establishing a yellow proportion and dry proportion prediction model, writing the prediction model into software, installing the software in an intelligent control system of a bulk curing barn, obtaining an actual yellow proportion and dry proportion in the current curing process through the intelligent control system with an on-site image acquisition function, comparing the actual yellow proportion and dry proportion with the yellow proportion and dry proportion required by a curing curve, if the yellow proportion and the dry proportion are the same, achieving a target in advance, if the yellow proportion and the dry proportion are different, estimating a delay time according to Huang Hongkao characteristics and dry curing characteristics of tobacco, automatically delaying fire, adjusting a dry bulb temperature and a wet bulb temperature, continuing curing, enabling the delay time to reach the target, entering the next curing stage, and automatically skipping to the next curing stage through real-time image acquisition and comparison without depending on the manual monitoring of experts.
Description
Technical Field
The invention belongs to the technical field of tobacco leaf baking, and particularly relates to a tobacco leaf primary baking control method based on an image inference model.
Background
The specific processing process of flue-cured tobacco is called flue-cured tobacco baking, baking for short, and is also called kang tobacco. It is not a simple dehydration drying process, but a series of unique and complex tobacco conditioning processes. The baking includes primary baking and secondary baking. The first curing of tobacco leaves harvested in the field (called fresh tobacco) is called primary curing and is usually carried out in a scattered manner in the tobacco field. The roasted tobacco leaves are called as primary flue-cured tobacco or raw tobacco (commonly called as dry tobacco), which are raw materials for people to use. The re-roasting of the raw tobacco is called re-roasting, which is mostly carried out in a re-roasting plant, and the roasted raw tobacco is called re-roasting tobacco.
The primary process of tobacco primary curing is to change planted tobacco into cigarette raw materials, and is a process that farmers place fresh tobacco harvested from the field into a curing barn to cure and modulate the fresh tobacco to become cigarette raw materials. Different baking temperatures are respectively set in the fresh tobacco leaf yellowing stage, the color fixing stage and the stem drying stage, so that the fresh tobacco leaves are sequentially baked to reach the tobacco leaf states of yellowing area, soft collapse and main vein softening in a certain proportion, and the yellow leaf yellow stem tip-hooked and curled tobacco leaf form is formed, and further the final main vein baked raw tobacco is obtained. The whole process not only needs to observe the change state of the tobacco leaves in real time to control the temperature and humidity of the curing barn through a flexible handle, but also needs to adjust the environment of the curing barn according to the change of the environment or the climate temperature and humidity.
In the traditional technology, the tobacco leaf primary baking is carried out in a manual judgment mode, time and labor are consumed, the traditional tobacco leaf primary baking control method is that an expert observes the color and the shrinkage state of the tobacco leaf through an observation window and serves as a judgment basis for adjusting the fire of a baking room, the method depends on the labor, the time is consumed, and the eye light scale is difficult to grasp. The current intensive curing barn technology adopts an intelligent controller to adjust the temperature and the humidity of the curing barn, and the specific method is that a curing curve is preset and is led into a control system to be cured. However, in this method, the actual change of the tobacco leaves during the baking process is not considered, specifically, the tobacco leaves are baked according to the baking curve even if the tobacco leaves reach the target, so that the over-baking condition is generated, the baking is stopped even if the optimal baking state is not reached, the baking stage is not accurately grasped, and the baking effect is difficult to ensure.
Disclosure of Invention
The invention aims to solve the problems that the actual change condition of tobacco leaves is not considered and the baking is inaccurate in the baking process of the current intelligent controller of a bulk curing barn, and aims to provide a tobacco leaf primary baking control method based on an image inference model, which can accurately control the baking stage through the intelligent controller, can obtain the actual change condition of the tobacco leaves and ensure the baking effect.
The invention adopts the following technical scheme:
a tobacco primary baking control method based on an image inference model comprises the following steps:
s1, building a yellow proportion and dry proportion prediction model in a baking process, writing the prediction model into software, and installing the software in an intelligent control system of a bulk curing barn;
s2, acquiring an actual yellow proportion and an actual dry proportion in the current baking process through an intelligent control system with a field image acquisition function;
s3, comparing the actual yellow proportion and dry proportion in the baking process with the yellow proportion and dry proportion required by a baking curve at any time, if the yellow proportion and the dry proportion are the same, reaching the target in advance, entering the next baking stage, if the yellow proportion and the dry proportion do not reach the target after the time required by the baking curve is reached, estimating the fire delay time according to the Huang Hongkao characteristics and the dry baking characteristics of the tobacco leaves, automatically delaying fire, adjusting the dry-bulb temperature and the wet-bulb temperature, continuing baking, and entering the next baking stage after the fire delay time reaches the target, and if the fire delay time is not reached, automatically skipping to the next baking stage.
The tobacco primary baking control method of the image inference model comprises the following steps of:
s11, the expert evaluates the yellow proportion and the dry proportion of the tobacco leaves in the baking process in a grading way;
s12, carrying out image acquisition on a flue-cured tobacco sample to be detected, and reading in an image as a sample image;
and S13, establishing a polynomial regression model by taking the yellow ratio and the dry ratio which are evaluated by experts as dependent variables and taking the flue-cured tobacco sample image, the dry-bulb temperature, the wet-bulb temperature and the baking time as independent variables.
In the method for controlling the primary flue-curing of the tobacco leaves by the image inference model, the image acquisition of the flue-cured tobacco sample to be detected in S12 is acquired by a plurality of cameras in a curing barn.
In the method for controlling the primary tobacco curing and baking of the image inference model, a plurality of cameras in the curing barn in the S12 are moved back and forth in real time to take images.
In the tobacco leaf primary baking control method of the image inference model, when the sample image is read in S12, the sample image is respectively converted into HSV and Lab color spaces, and the pixel values of HSV and Lab channels of the image are obtained.
The tobacco primary baking control method of the image inference model is characterized in that a yellow proportion and dry proportion prediction polynomial regression model established in S13 is as follows:
in the formula, x 1 Represents the dry bulb temperature; x is the number of 2 Represents the wet bulb temperature; x is a radical of a fluorine atom 3 Represents an H-channel pixel value; x is the number of 4 Representing an S-channel pixel value; x is the number of 5 Represents a V-channel pixel value; x is the number of 6 Represents an L-channel pixel value; x is the number of 7 Representing a-channel pixel values; x is the number of 8 Representing a b-channel pixel value; x is the number of 9 Represents the baking time; w is a 0 Representing a bias term; w is a 1 A regression coefficient representing a dry-bulb temperature; w is a 2 A regression coefficient representing a wet bulb temperature; w is a 3 A regression coefficient representing the product of the dry bulb temperature and the wet bulb temperature; w is a 4 A regression coefficient representing the square of the H-channel pixel value; w is a 5 A regression coefficient representing the square of the S-channel pixel value; w is a 6 A regression coefficient representing a V-channel pixel value; w is a 7 A regression coefficient representing an L-channel pixel value; w is a 8 A regression coefficient representing a product of the V-channel pixel value and the L-channel pixel value; w is a 9 A regression coefficient representing the square of the a-channel pixel value; w is a 10 A regression coefficient representing the square of the b-channel pixel value; w is a 11 A regression coefficient representing a baking time; y represents the predicted values of the dependent variable yellow ratio and the dry ratio.
The tobacco primary baking and baking control method of the image inference model comprises the following steps of S3:
s31, obtaining the current yellow proportion and the dry proportion according to the yellow proportion and dry proportion prediction model, and subtracting the initial yellow proportion and dry proportion obtained by the yellow proportion and dry proportion measurement model to obtain a changed yellow value and a changed dry value;
s32, subtracting the initial baking time from the current baking time to obtain the accumulated baking time of the baking stage;
and S33, dividing the changed yellow and the changed dry and accumulated baking time, the dry bulb temperature and the wet bulb temperature to obtain Huang Hongkao characteristics and dry baking characteristics of the tobacco leaves.
The tobacco primary baking and baking control method of the image inference model is characterized in that a calculation formula of the yellow baking characteristic and the dry baking characteristic in S33 is as follows:
huang Hongkao characteristics = change yellow/(cumulative bake time dry bulb temperature wet bulb temperature)
Dry bake characteristics = change dry/(cumulative bake time dry bulb temperature wet bulb temperature)
According to the control method for primary tobacco leaf baking based on the image inference model, the delay time is 2 hours.
The field image acquisition refers to the acquisition of a plurality of cameras in a curing barn, and the cameras in the curing barn move back and forth in real time to capture images.
The invention has the beneficial effects that: the invention solves the problems that the actual change condition of tobacco leaves is not considered and the baking is inaccurate in the baking process of the current intelligent controller of the bulk curing barn, a yellow proportion and dry proportion prediction model is established, the prediction model is written into software and is installed in an intelligent control system of the bulk curing barn with an on-site image acquisition function, the actual yellow proportion and dry proportion in the baking process are predicted and are compared with the yellow proportion and dry proportion required by a baking curve, if the yellow proportion and the dry proportion are the same, the target is reached in advance, if the yellow proportion and the dry proportion are not the same, the delay time is estimated according to Huang Hongkao characteristics and dry baking characteristics, the automatic delay is carried out, the dry bulb temperature and the wet bulb temperature are adjusted, the baking is continued, the delay time reaches the target, the next baking stage can be entered, and if the delay time is not up to the standard after the baking, the next baking stage is automatically jumped. The baking stage can be accurately evaluated through real-time image acquisition and comparison without depending on manual monitoring of experts, and the phenomena of excessive baking and substandard baking are prevented.
And a plurality of cameras moving back and forth in real time are used for image capture, so that the image of the tobacco leaves can be accurately acquired at any time, the yellow proportion and the dry proportion of the tobacco leaves are calculated in real time, and the yellow proportion and the dry proportion of the tobacco leaves are compared with those of the baking curve, so that the baking control stage is accurately carried out.
By means of the Huang Hongkao characteristic and the dry-baking characteristic, when the yellow proportion and the dry proportion do not reach the standard, the changed yellow and the changed dry needed for reaching the target yellow proportion and dry proportion can be calculated according to the Huang Hongkao characteristic and the dry-baking characteristic, and therefore the dry-bulb temperature and the wet-bulb temperature are adjusted during automatic fire delay, the target yellow proportion and dry proportion are achieved, and the baking effect is guaranteed.
Drawings
Fig. 1 is a flow chart of a tobacco primary flue-curing control method based on an image inference model.
Detailed Description
The technical content of the present invention is further described below with reference to the specific embodiments and the accompanying drawings, but the scope of the present invention is not limited thereto.
Referring to fig. 1, a tobacco primary flue-curing control method based on an image inference model comprises the following steps:
s1, building a yellow proportion and dry proportion prediction model in the baking process, writing the prediction model into software, and installing the software in an intelligent control system of a bulk curing barn.
The intelligent control system controls the working temperature and humidity in the curing barn and the curing time of tobacco leaves through a temperature and humidity sensor, a plurality of cameras which are moved back and forth and have changeable shooting angles, a single chip microcomputer technology and fuzzy logic software, so that the intelligent and digital control of the existing curing barn is achieved, the whole tobacco curing process is programmed and digitized, the phenomenon of unstable tobacco curing quality is changed, and the tobacco curing efficiency is improved.
The yellow proportion and dry proportion prediction model is established by the following steps:
s11, a specialist scores and evaluates the yellow proportion and the dry proportion of the tobacco leaves in the baking process;
s12, carrying out image acquisition on a flue-cured tobacco sample to be detected, and reading in an image as a sample image;
and S13, establishing a polynomial regression model by taking the yellow ratio and the dry ratio which are evaluated by experts as dependent variables and taking the flue-cured tobacco sample image, the dry-bulb temperature, the wet-bulb temperature and the baking time as independent variables.
The image acquisition of the flue-cured tobacco sample to be detected in S12 is acquired through a plurality of cameras in the curing barn, the cameras in the curing barn in S12 move back and forth in real time to take images, and when the images are read in, the images are respectively converted into HSV (hue saturation value) and Lab (Lab) color spaces, so that HSV (hue saturation value) and Lab channel pixel values of the images are acquired.
The established yellow ratio and dry ratio prediction polynomial regression model is as follows:
in the formula, x 1 Represents the dry bulb temperature; x is the number of 2 Represents the wet bulb temperature; x is the number of 3 Represents an H-channel pixel value; x is a radical of a fluorine atom 4 Representing an S-channel pixel value; x is the number of 5 Represents a V-channel pixel value; x is the number of 6 Represents an L-channel pixel value; x is the number of 7 Representing a-channel pixel values; x is the number of 8 Representing a b-channel pixel value; x is the number of 9 Represents the baking time; w is a 0 Representing a bias term; w is a 1 A regression coefficient representing a dry bulb temperature; w is a 2 A regression coefficient representing a wet bulb temperature; w is a 3 A regression coefficient representing the product of dry-bulb temperature and wet-bulb temperature; w is a 4 A regression coefficient representing the square of the H-channel pixel value; w is a 5 A regression coefficient representing the square of the S-channel pixel value; w is a 6 A regression coefficient representing a V-channel pixel value; w is a 7 A regression coefficient representing an L-channel pixel value; w is a 8 A regression coefficient representing a product of the V-channel pixel value and the L-channel pixel value; w is a 9 A regression coefficient representing the square of the a-channel pixel value; w is a 10 A regression coefficient representing the square of the b-channel pixel value; w is a 11 A regression coefficient representing a baking time; y represents the predicted values of the dependent variable yellow ratio and the dry ratio.
And S2, acquiring the actual yellow proportion and dry proportion in the current baking process through an intelligent control system with a field image acquisition function.
The field image acquisition is acquired by a plurality of cameras in the baking room, and the cameras in the baking room move back and forth in real time to acquire images.
S3, comparing the actual yellow proportion and dry proportion in the baking process with the yellow proportion and dry proportion required by a baking curve at any time, if the yellow proportion and the dry proportion are the same, reaching the target in advance, entering the next baking stage, if the yellow proportion and the dry proportion do not reach the target after the time required by the baking curve is reached, estimating the fire delay time according to the Huang Hongkao characteristics and the dry baking characteristics of the tobacco leaves, automatically delaying fire, adjusting the dry-bulb temperature and the wet-bulb temperature, continuing baking, and entering the next baking stage after the fire delay time reaches the target, and if the fire delay time is not reached, automatically skipping to the next baking stage.
The estimated delay time from the Huang Hongkao and dry-bake characteristics of tobacco leaves may be 2 hours.
The Huang Hongkao characteristic and the dry baking characteristic are established by the following steps:
s31, obtaining the current yellow proportion and the dry proportion according to the yellow proportion and dry proportion prediction model, and subtracting the initial yellow proportion and dry proportion obtained by the yellow proportion and dry proportion measurement model to obtain a changed yellow value and a changed dry value;
s32, subtracting the initial baking time from the current baking time to obtain the accumulated baking time of the baking stage;
and S33, dividing the changed yellow and the changed dry and accumulated baking time, the dry bulb temperature and the wet bulb temperature to obtain Huang Hongkao characteristics and dry baking characteristics of the tobacco leaves.
5363 the formula for the calculation of the characteristic Huang Hongkao and the dry bake characteristic is:
huang Hongkao characteristics = change yellow/(cumulative bake time dry bulb temperature wet bulb temperature)
Dry bake characteristics = change dry/(cumulative bake time dry bulb temperature wet bulb temperature)
The following is a comparison of the Huang Hongkao and dry-bake characteristics obtained by this method with the expert's evaluation of the cured tobacco Huang Hongkao and dry-bake characteristics.
The method comprises the following steps of taking primarily-cured tobacco leaves in the curing process as a test object, selecting a tobacco leaf sample from a Yunnan Chuxiong producing area of 2022 years, scoring the yellow proportion and the dry proportion of the tobacco leaves by an expert, wherein the initial yellow proportion and the dry proportion are 0, subtracting the initial yellow proportion and the dry proportion from the current yellow proportion and the dry proportion to obtain a changed yellow value and a changed dry value, and obtaining the current curing time, the dry bulb temperature and the wet bulb temperature from a control terminal sensor according to a formula: huang Hongkao characteristics = change yellow/(cumulative bake time dry bulb temperature wet bulb temperature); dry bake characteristics = change dry/(cumulative bake time dry bulb temperature wet bulb temperature), the table uses a 0.05% integer multiple upper limit counting method to obtain Huang Hongkao characteristics and dry bake characteristics as follows.
TABLE 1 expert evaluation of flue-cured tobacco samples
Selecting pre-cured tobacco leaves in Yunan Chuxiong producing area of 2022 years as samples, and obtaining Huang Hongkao characteristics and dry-curing characteristics of the samples by adopting a tobacco leaf pre-curing control method based on an image inference model as follows:
firstly, establishing a yellow proportion and dry proportion prediction model based on correlation between HSV and Lab channel pixel values obtained in real time by a camera moving back and forth in a baking room and a yellow proportion and a dry proportion evaluated by an expert, taking the yellow proportion and the dry proportion evaluated by the expert as dependent variables, reading the dry and wet temperature, the HSV and Lab channel pixel values and the baking time as independent variables, and establishing a polynomial regression model by applying the independent variables and the dependent variables as follows to realize the prediction of the yellow proportion and the dry proportion:
y represents a dependent variable yellow ratio predicted value; y2 represents a dependent variable dry ratio predicted value; x is the number of 1 Represents the dry bulb temperature; x is the number of 2 Represents the wet bulb temperature; x is the number of 3 Represents an H-channel pixel value; x is the number of 4 Representing an S-channel pixel value; x is the number of 5 Represents a V-channel pixel value; x is the number of 6 Represents an L-channel pixel value; x is the number of 7 Representing a-channel pixel values; x is the number of 8 Representing a b-channel pixel value; x is the number of 9 Indicating the baking time.
Next, huang Hongkao characteristics and dry bake characteristics were calculated.
(1) And (3) applying the model to predict and obtain the yellow ratio and dry ratio values of the current time, wherein the prediction results are shown in a table 2.
TABLE 2 yellow ratio and Dry ratio prediction values of flue-cured tobacco samples
(2) And subtracting the obtained yellow ratio and dry ratio from the yellow ratio and dry ratio of the starting time to obtain a changed yellow value and a changed dry value.
(3) And subtracting the initial baking time from the current baking time to obtain the accumulated baking time of the baking stage.
(4) Dividing change yellow and change dry by the product of the cumulative baking time, the mean value of dry bulb temperature and the mean value of wet bulb temperature in the baking time period: huang Hongkao characteristics = change yellow/(cumulative bake time dry bulb temperature wet bulb temperature); dry bake characteristics = change dry/(cumulative bake time dry bulb temperature wet bulb temperature) to obtain Huang Hongkao characteristics and dry bake characteristic values, and the average of Huang Hongkao characteristics and dry bake characteristics of bake stage 6, bake stage 7, and bake stage 8 is shown in table 3.
TABLE 3 Huang Hongkao mean values for Properties and Dry baking Properties
| Current stage of toasting | Huang Hongkao characteristic | Dry bake characteristics |
| 6 | 0.07％ | 0.02％ |
| 7 | 0.05％ | 0.02％ |
| 8 | 0.04％ | 0.01％ |
The Huang Hongkao characteristic and the dry-baking characteristic results of the method are compared with the evaluation of the experts on the baked tobacco leaves, and the comparison results are shown in table 4.
TABLE 4 Huang Hongkao Properties and Dry bake Property comparison results
As can be seen from Table 4, according to the evaluation of Huang Hongkao characteristic and dry-baking characteristic degree of tobacco leaf baking by experts in the baking process and the comparison of Huang Hongkao characteristic and dry-baking characteristic results obtained by applying the method, the consistent ratio of the two evaluation results is found to be 100%, and the accuracy is high.
Therefore, the baking control method based on the image reasoning model can be used for accurately evaluating the baking effect of the flue-cured tobacco leaves in the baking stage and judging whether the baking reaches the standard or not. According to the image reasoning model-based tobacco primary baking control method, the baking stage can be accurately controlled through the intelligent controller, the baking effect is ensured, and the phenomena of over-baking and substandard baking are prevented. The method provides a research basis for improving the efficiency of the digital tobacco leaf baking by incorporating the characteristics of Huang Hongkao baking and the characteristics of dry baking in the digital baking in the future.
Claims (10)
1. A tobacco primary baking control method based on an image inference model comprises the following steps:
s1, building a yellow proportion and dry proportion prediction model in a baking process, writing the prediction model into software, and installing the software in an intelligent control system of a bulk curing barn;
s2, acquiring an actual yellow proportion and an actual dry proportion in the current baking process through an intelligent control system with a field image acquisition function;
s3, comparing the actual yellow proportion and dry proportion in the baking process with the yellow proportion and dry proportion required by a baking curve at any time, if the yellow proportion and the dry proportion are the same, reaching the target in advance, entering the next baking stage, if the yellow proportion and the dry proportion do not reach the target after the time required by the baking curve is reached, estimating the delay time according to the Huang Hongkao characteristic and the dry baking characteristic of the tobacco leaves, automatically delaying, adjusting the dry-bulb temperature and the wet-bulb temperature, continuing baking, and entering the next baking stage after the delay time reaches the target, and if the delay time is not reached, automatically skipping to the next baking stage.
2. The tobacco flue-curing control method of the image inference model according to claim 1, wherein the yellow ratio and dry ratio prediction model in the S1 is established by the steps of:
s11, the expert evaluates the yellow proportion and the dry proportion of the tobacco leaves in the baking process in a grading way;
s12, carrying out image acquisition on a flue-cured tobacco sample to be detected, and reading in an image as a sample image;
and S13, establishing a polynomial regression model by taking the yellow ratio and the dry ratio which are evaluated by experts as dependent variables and taking the flue-cured tobacco sample image, the dry-bulb temperature, the wet-bulb temperature and the baking time as independent variables.
3. The tobacco flue-curing control method of the image inference model according to claim 2, wherein the image acquisition of the flue-cured tobacco sample to be tested in S12 is acquired by a plurality of cameras in the curing barn.
4. The tobacco curing control method according to claim 3, wherein the cameras in the curing barn in S12 are moved back and forth in real time for image capture.
5. The tobacco flue-curing control method according to claim 4, wherein when the sample image is read in S12, the sample image is converted into HSV and Lab color spaces respectively to obtain HSV and Lab channel pixel values.
6. The tobacco flue-curing control method of the image inference model according to claim 5, wherein the yellow ratio and dry ratio prediction polynomial regression model established in S13 is:
in the formula, x 1 Represents the dry bulb temperature; x is the number of 2 Represents the wet bulb temperature; x is the number of 3 Represents an H-channel pixel value; x is the number of 4 Representing an S-channel pixel value; x is the number of 5 Represents a V-channel pixel value; x is the number of 6 Represents an L-channel pixel value; x is the number of 7 Representing a-channel pixel values; x is the number of 8 Representing a b-channel pixel value; x is the number of 9 Represents the baking time; w is a 0 Representing a bias term; w is a 1 A regression coefficient representing a dry bulb temperature; w is a 2 A regression coefficient representing a wet bulb temperature; w is a 3 A regression coefficient representing the product of the dry bulb temperature and the wet bulb temperature; w is a 4 A regression coefficient representing the square of the H-channel pixel value; w is a 5 A regression coefficient representing the square of the S-channel pixel value; w is a 6 A regression coefficient representing a V-channel pixel value; w is a 7 A regression coefficient representing an L-channel pixel value; w is a 8 A regression coefficient representing a product of the V-channel pixel value and the L-channel pixel value; w is a 9 A regression coefficient representing the square of the a-channel pixel value; w is a 10 A regression coefficient representing the square of the b-channel pixel value; w is a 11 A regression coefficient representing a baking time; y represents the predicted values of the dependent variable yellow ratio and the dry ratio.
7. The tobacco primary baking control method of the image inference model according to any one of claims 1-6, wherein the yellow baking characteristics and the dry baking characteristics are established in step S3 by:
s31, obtaining the current yellow proportion and the dry proportion according to the yellow proportion and dry proportion prediction model, and subtracting the initial yellow proportion and dry proportion obtained by the yellow proportion and dry proportion measurement model to obtain a changed yellow value and a changed dry value;
s32, subtracting the initial baking time from the current baking time to obtain the accumulated baking time of the baking stage;
and S33, dividing the changed yellow and the changed dry and accumulated baking time, the dry bulb temperature and the wet bulb temperature to obtain Huang Hongkao characteristics and dry baking characteristics of the tobacco leaves.
8. The method for controlling the primary baking and the baking of the tobacco leaves through the image inference model according to claim 7, wherein the calculation formula of the yellow baking characteristic and the dry baking characteristic in S33 is as follows:
huang Hongkao characteristics = change yellow/(cumulative bake time dry bulb temperature wet bulb temperature)
Dry bake characteristics = change dry/(cumulative bake time dry bulb temperature wet bulb temperature).
9. The image inference model-based tobacco primary flue-curing control method according to any one of claims 1-8, wherein the delay time is 2 hours.
10. The tobacco curing control method based on image inference model according to any one of claims 1-8, wherein the field image collection is collection by multiple cameras in the curing barn, and the multiple cameras in the curing barn move back and forth in real time for image capture.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211608863.8A CN115736298A (en) | 2022-12-14 | 2022-12-14 | Image reasoning model-based tobacco primary baking control method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211608863.8A CN115736298A (en) | 2022-12-14 | 2022-12-14 | Image reasoning model-based tobacco primary baking control method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115736298A true CN115736298A (en) | 2023-03-07 |
Family
ID=85345984
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211608863.8A Pending CN115736298A (en) | 2022-12-14 | 2022-12-14 | Image reasoning model-based tobacco primary baking control method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115736298A (en) |
-
2022
- 2022-12-14 CN CN202211608863.8A patent/CN115736298A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104914720A (en) | Electronic nose intelligent baking control system having automatic learning function and control method | |
| CN110179151B (en) | Estimation method of initial moisture content of tobacco leaves during baking, baking method of fresh tobacco leaves, mathematical model and construction method of mathematical model | |
| CN109708459B (en) | Intelligent drying control method, system and device | |
| CN111109643B (en) | Method and system for controlling water content of cut tobacco of sheet cut tobacco drying | |
| CN104729174A (en) | Heat pump drying controller for crop drying and control method thereof | |
| Condorí et al. | Image processing for monitoring of the cured tobacco process in a bulk-curing stove | |
| CN110946313B (en) | Method and system for controlling water content of outlet of cut tobacco drying process | |
| CN102722201A (en) | Automatic baking monitoring system and method | |
| CN105334738B (en) | A kind of method of evaluating performance suitable for tobacco processing course pid control circuit | |
| CN114115393A (en) | Method for controlling moisture and temperature at outlet of cut tobacco dryer for sheet cut tobacco making line | |
| CN109724398B (en) | Wood drying control method and device based on artificial intelligence | |
| CN115736298A (en) | Image reasoning model-based tobacco primary baking control method | |
| CN113358647A (en) | Tobacco baking data prediction monitoring model establishing method | |
| CN114403487B (en) | Water adding control method for loosening and dampening | |
| CN114868948A (en) | Tobacco leaf intelligent baking dynamic management method | |
| CN103745087B (en) | A kind of dynamic changes of forest resources Forecasting Methodology based on remote sensing technology | |
| CN105029672A (en) | Precise characterization method for water at outlet of baked cut tobacco | |
| WO2022052547A1 (en) | Method for predicting energy efficiency of air-conditioning system, and air-conditioning system | |
| CN112335923B (en) | Tobacco dryer | |
| CN107808209B (en) | Wind power plant abnormal data identification method based on weighted kNN distance | |
| CN112987590A (en) | Intelligent household control method and system based on intelligent analysis of environmental laws | |
| CN115628526B (en) | Dry and wet load regulation and control method and system based on air conditioning unit data processing | |
| CN110826229A (en) | Cut tobacco drying head process modeling and optimal setting control method based on long-term prediction | |
| CN116182321A (en) | Automatic temperature regulating system of heating ventilation air conditioner based on machine learning | |
| CN115349654A (en) | Calibration method of tobacco leaf baking parameters |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |