A tobacco combustion detection method and system based on image recognition

Abstract

The application discloses a tobacco combustion detection method and system based on image recognition. Image information of a tobacco sample at different time intervals during combustion is acquired by a camera device. The change information of the tobacco sample during combustion is analyzed according to the image information at different time intervals. The change information of the tobacco sample is compared with existing reference image information. Correction information of a tobacco production process is obtained. The puffing process and the impurity removal process in the tobacco production are adjusted accordingly. The number of process steps is reduced, the input of manpower and material resources is reduced, and the quality of the tobacco product is improved and the production cost is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of image recognition, and more particularly, to a tobacco combustion detection method and system based on image recognition. BACKGROUND

[0002] Due to the traditional technology, the production and processing yield of some tobacco manufacturers is often not ideal, and the quality of tobacco products processed from different batches of tobacco raw materials is uneven. Without targeted processing of different batches of tobacco raw materials, the quality of the finished products cannot be guaranteed. Therefore, there is an urgent need for a method that can effectively improve the quality of tobacco processing products. SUMMARY

[0003] To solve the above at least one technical problem, the present application provides a tobacco combustion detection method and system based on image recognition.

[0004] The first aspect of the present application provides a tobacco combustion detection method based on image recognition, comprising:

[0005] Obtain a tobacco raw material sample to be detected, perform a combustion test on the tobacco raw material sample, and obtain combustion process image information;

[0006] Obtain reference image information of a good product raw material combustion process, and perform image analysis, comparison, and obtain combustion process deviation information on the combustion process image information and the reference image information;

[0007] Perform data comparison and analysis on the combustion process deviation information, and generate manufacturing process correction information;

[0008] Send the manufacturing process correction information to a preset terminal device for display.

[0009] In the present application, the tobacco raw material sample to be detected is obtained, the tobacco raw material sample is subjected to a combustion test, and the combustion process image information is obtained, which includes:

[0010] Obtain different batches of tobacco raw materials, and number the different batches of tobacco raw materials;

[0011] According to the numbering order, obtain each batch of tobacco detection raw material;

[0012] Divide the tobacco detection raw material into three groups on average, and make each group of tobacco detection raw material into cigarettes to obtain the tobacco raw material sample to be detected.

[0013] In the present application, the tobacco raw material sample to be detected is obtained, the tobacco raw material sample is subjected to a combustion test, and the combustion process image information is obtained, which includes:

[0014] Place the tobacco raw material sample in a predetermined experimental environment for combustion testing;

[0015] According to a preset time interval, an original image of tobacco raw material sample combustion in each time interval is acquired by the camera device;

[0016] The original image of combustion is subjected to image enhancement processing, and the processed original image is subjected to data arrangement to obtain combustion process image information.

[0017] In the scheme, the reference image information of the good product raw material combustion process is acquired, the combustion process image information is compared with the reference image information through image analysis, and combustion process deviation information is obtained, specifically as follows:

[0018] The combustion process image information is subjected to image recognition analysis, and image feature extraction is performed on the smoke drifting in the combustion process to obtain smoke drifting feature data;

[0019] The smoke drifting feature data is subjected to data analysis to obtain smoke drifting shape information, and according to the smoke drifting shape information, combined with the change of the combustion process, smoke drifting shape change information is obtained;

[0020] The smoke drifting shape change information is compared with the reference image information through data analysis to obtain smoke drifting shape deviation information.

[0021] In the scheme, the reference image information of the good product raw material combustion process is acquired, the combustion process image information is compared with the reference image information through image analysis, and combustion process deviation information is obtained, and the scheme further includes:

[0022] The combustion process image information is acquired, and smoke drifting color feature values in the combustion process image information are extracted;

[0023] The smoke drifting color feature values are subjected to data summarization, arrangement and analysis to obtain smoke drifting color change information;

[0024] The smoke drifting color change information is compared with the reference image information through data analysis to obtain smoke drifting color deviation information.

[0025] In the scheme, the combustion process deviation information is subjected to data comparison and analysis, and production process correction information is generated, specifically as follows:

[0026] According to the smoke drifting shape deviation information, data splitting and data extraction are performed to obtain smoke drifting size deviation values and smoke drifting length deviation values;

[0027] The smoke drifting size deviation values and the smoke drifting length deviation values are subjected to numerical average processing, and the processed deviation average values are imported into a process correction model;

[0028] The process correction model performs comprehensive analysis according to the deviation average values to obtain puffing process correction information.

[0029] In the scheme, the combustion process deviation information is compared and analyzed, and the production process correction information is generated, specifically:

[0030] According to the smoke color deviation information, the smoke color deviation value is obtained, the smoke color deviation value is subjected to numerical average processing, and the color average deviation value is obtained.

[0031] Determine whether the color average deviation value is less than a preset threshold value.

[0032] If it is less than, the color average deviation value is introduced into the process correction model for data analysis, and the impurity removal process correction information is obtained.

[0033] The second aspect of the application also provides a tobacco combustion detection system based on image recognition, which comprises a memory and a processor, the memory comprises a tobacco combustion detection method based on image recognition, and the processor executes the following steps when the tobacco combustion detection method based on image recognition is executed:

[0034] Obtain a tobacco raw material sample to be detected, perform combustion test on the tobacco raw material sample, and obtain combustion process image information;

[0035] Obtain reference image information of a good product raw material combustion process, and perform image analysis, comparison, and obtain combustion process deviation information by comparing the combustion process image information with the reference image information;

[0036] The combustion process deviation information is compared and analyzed, and the production process correction information is generated;

[0037] The production process correction information is sent to a preset terminal device for display.

[0038] In the scheme, the tobacco raw material sample to be detected is obtained, the combustion test is performed on the tobacco raw material sample, and the combustion process image information is obtained, specifically:

[0039] Place the tobacco raw material sample in a predetermined experimental environment for combustion test;

[0040] According to a predetermined time interval, the combustion original image of the tobacco raw material sample in each time interval is obtained by a camera device;

[0041] The combustion original image is subjected to image enhancement processing, and the processed original image is subjected to data arrangement to obtain the combustion process image information.

[0042] In the scheme, the reference image information of the good product raw material combustion process is obtained, and the combustion process image information is compared with the reference image information for image analysis, comparison, and obtaining of the combustion process deviation information, specifically:

[0043] The combustion process image information is subjected to image recognition analysis, and image feature extraction is performed on the smoke in the combustion process to obtain smoke feature data;

[0044] The smoke feature data is subjected to data analysis to obtain smoke shape information, and according to the smoke shape information, combined with the change of the combustion process, smoke shape change information is obtained;

[0045] The smoke shape change information is compared with the reference image information through data analysis to obtain smoke shape deviation information.

[0046] The present application discloses a kind of tobacco combustion detection method and system based on image recognition, different time interval image information in tobacco sample combustion process is obtained by camera device, according to the image information of different time interval, smoke change information of tobacco sample in combustion process is analyzed, smoke change information is compared with the analysis of existing reference image information, and correction information of tobacco production process is obtained, so as to adjust the bulking process and impurity removal process in tobacco production pertinently, reduce the process step of redundant, so as to reduce the input of manpower and material resources, finally reach the beneficial effect of improving tobacco finished product quality and reducing production cost. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 The flow chart of the tobacco combustion detection method based on image recognition is shown;

[0048] Figure 2 The flow chart of the combustion process image information acquisition is shown;

[0049] Figure 3 The flow chart of the smoke shape deviation information acquisition is shown;

[0050] Figure 4 The block diagram of the tobacco combustion detection system based on image recognition is shown. DETAILED DESCRIPTION

[0051] In order to more clearly understand the above-mentioned purposes, features and advantages of the present application, the present application will be further described in detail below in combination with the drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0052] In the following description, many specific details are set forth in order to provide a thorough understanding of the present application, however, the present application can also be implemented in other ways different from those described herein, therefore, the protection scope of the present application is not limited by the specific embodiments disclosed below.

[0053] Figure 1 The flow chart of the tobacco combustion detection method based on image recognition is shown.

[0054] As Figure 1 shown, the first aspect of the present application provides a tobacco combustion detection method based on image recognition, comprising:

[0055] S102, obtaining a tobacco raw material sample to be detected, performing combustion test on the tobacco raw material sample, and obtaining combustion process image information;

[0056] S104, obtaining reference image information of a good product raw material combustion process, and performing image analysis, comparison, and obtaining combustion process deviation information on the combustion process image information and the reference image information;

[0057] S106, performing data comparison and analysis on the combustion process deviation information, and generating manufacturing process correction information;

[0058] S108, sending the manufacturing process correction information to a preset terminal device for display.

[0059] According to the embodiment of the present application, the obtaining of the tobacco raw material sample to be detected, the combustion test on the tobacco raw material sample, and the obtaining of the combustion process image information, before including:

[0060] Obtaining different batches of tobacco raw materials, and numbering the different batches of tobacco raw materials;

[0061] According to the numbering order, obtaining each batch of tobacco detection raw material;

[0062] Dividing the tobacco detection raw material into three groups on average, and making each group of tobacco detection raw material into cigarettes to obtain the tobacco raw material sample to be detected.

[0063] It should be noted that the quality of tobacco raw materials between the same batches is also relatively consistent, while due to the difference in tobacco raw materials and the inconsistency in processing time between different batches, there are often large differences in quality, so in subsequent processing, different batches of tobacco raw materials need to be processed in a positive manner to obtain better cut tobacco and cigarette products. In the obtained tobacco raw material sample to be detected, the weight of each batch of tobacco raw material sample to be detected is consistent, and in the dividing the tobacco detection raw material into three groups on average, the grouping is based on weight average division.

[0064] Figure 2 The flow chart of obtaining the combustion process image information of the present application is shown.

[0065] According to the embodiment of the present application, the obtaining of the tobacco raw material sample to be detected, the combustion test on the tobacco raw material sample, and the obtaining of the combustion process image information, specifically:

[0066] S202, placing the tobacco raw material sample in a preset experimental environment for combustion test;

[0067] S204, acquiring, by the camera, a tobacco raw material sample combustion original image in each time interval according to a preset time interval;

[0068] S206, performing image enhancement processing on the combustion original image, and performing data arrangement on the processed original image to obtain combustion process image information.

[0069] It should be noted that in the step of acquiring a tobacco raw material sample combustion original image in each time interval, the acquired combustion original image contains a plurality of image data, and the data amount of the original image is affected by the preset time interval, and the preset time interval is generally 30-50 seconds. The image enhancement processing on the combustion original image is to highlight the color features of the smoke in the combustion process by enhancing the color details of the enhanced image, so that subsequent image recognition and processing can be better performed.

[0070] Figure 3 A flowchart for acquiring smoke shape deviation information is shown.

[0071] According to the embodiment of the present application, the reference image information of the good product raw material combustion process is acquired, and the combustion process image information is compared with the reference image information for image analysis, and the combustion process deviation information is obtained, specifically:

[0072] S302, performing image recognition analysis on the combustion process image information, and performing image feature extraction on the smoke in the combustion process to obtain smoke feature data;

[0073] S304, performing data analysis on the smoke feature data to obtain smoke shape information, and obtaining smoke shape change information according to the smoke shape information and the change of the combustion process;

[0074] S306, performing data analysis and comparison on the smoke shape change information and the reference image information to obtain smoke shape deviation information.

[0075] It should be noted that the smoke shape deviation information is deviation information obtained by comparing the reference image information. The reference image information of the good product raw material combustion process is image information acquired in the combustion process of high-quality finished cut tobacco artificially selected, and the image information contains reference smoke shape change information and has good reference significance.

[0076] According to the embodiment of the present application, the reference image information of the good product raw material combustion process is acquired, and the combustion process image information is compared with the reference image information for image analysis, and the combustion process deviation information is obtained, and further comprising:

[0077] Acquiring the combustion process image information and extracting smoke color feature values in the combustion process image information;

[0078] The smoke color feature values are summarized, sorted and analyzed to obtain smoke color change information;

[0079] The smoke color change information is compared with reference image information to obtain smoke color deviation information.

[0080] It should be noted that the combustion process deviation information includes smoke shape deviation information and smoke color deviation information. The smoke color feature values are multiple sets of feature values, and different sets of feature values correspond to image information at different time periods in the combustion process. The smoke color change information is color change information of the smoke in the combustion process, which can effectively reflect the color concentration change of the smoke. By studying the color change of the smoke, the quality of the tobacco sample can be further analyzed. In addition, the reference image information is image information obtained during the combustion process of high-quality finished cut tobacco selected by artificial selection, and the image information contains reference smoke color change information, which has good reference significance.

[0081] According to the embodiment of the present application, the combustion process deviation information is compared and analyzed, and manufacturing process correction information is generated, which further comprises:

[0082] According to a preset combustion time interval, image information within a preset start time and image information within a preset end time in the combustion process are obtained;

[0083] The image information is summarized to obtain start image data and end image data;

[0084] The start image data and the end image data are analyzed to obtain the initial change time of the smoke and the end change time of the smoke;

[0085] The initial change time of the smoke and the end change time of the smoke are analyzed to obtain tobacco humidity correction information.

[0086] It should be noted that the tobacco raw material has a certain hygroscopicity, and contains water inside. The amount of water has a certain influence on the transportation, storage and processing of the tobacco raw material, and also has a certain influence on the quality of the tobacco raw material. By studying the change of the smoke within 30 to 40 seconds at the beginning and end of the combustion process of the tobacco raw material sample, the humidity content of the current tobacco raw material can be accurately analyzed to obtain the tobacco humidity correction information, thereby achieving the purpose of improving the quality of the tobacco raw material.

[0087] In addition, the preset combustion time interval is 2 to 3 seconds, the preset start time and the preset end time are 30 to 40 seconds, and the start image data and the end image data each contain multiple image data. The size of the image data is affected by the size of the preset combustion time interval, and the tobacco humidity correction information includes the content and concentration of the added humectant.

[0088] According to the embodiment of the present application, the reference image information of the good-quality raw material combustion process is acquired, the combustion process image information is analyzed, compared with the reference image information, and the combustion process deviation information is obtained, and the embodiment further comprises:

[0089] The combustion process deviation information in different batches of tobacco raw material samples is acquired;

[0090] The combustion process deviation information is re-numbered according to the numbers of different batches of tobacco raw material, and a plurality of groups of combustion process deviation information is obtained;

[0091] Each two groups of combustion process deviation information are analyzed and compared, and a data difference rate in the two groups of combustion process deviation information is obtained;

[0092] If the data difference rate is lower than a preset difference rate, the two batches of tobacco raw material corresponding to the two groups of combustion process deviation information are combined.

[0093] It should be noted that in the plurality of groups of combustion process deviation information, each group of combustion process deviation information corresponds to the combustion process deviation information in a batch of tobacco raw material samples, and each group of combustion process deviation information corresponds to a number, which is consistent with the number of the corresponding batch of tobacco raw material. In the data difference rate is lower than a preset difference rate, if the data difference rate is lower than the preset difference rate, it means that the quality of the corresponding two batches of tobacco raw material is relatively consistent, and the two batches of tobacco raw material are combined. Through the batch combination operation, the consumption of manpower and material resources in subsequent processing can be reduced, and the redundant processing steps caused by batch processing of raw materials can be reduced.

[0094] According to the embodiment of the present application, the combustion process deviation information is compared and analyzed, and the manufacturing process correction information is generated, specifically:

[0095] According to the data splitting and data extraction of the smoke shape deviation information, the smoke size deviation value and the smoke length deviation value are obtained;

[0096] The smoke size deviation value and the smoke length deviation value are subjected to numerical averaging processing, and the processed deviation average value is introduced into the process correction model;

[0097] The process correction model comprehensively analyzes the deviation average value, and obtains the puffing process correction information.

[0098] It should be noted that the smoke size deviation value and the smoke length deviation value each include multiple sets of values. In addition, data analysis based on the obtained smoke size deviation value and the smoke length deviation value can obtain puffing process correction information of the tobacco raw material. The puffing process can increase the bulkiness of the tobacco raw material, increase the volume, reduce the weight, and make the combustion more complete. The size and length of the smoke generated by the combustion are closer to the size and length of the smoke in the reference image information. The puffing process correction information includes puffing treatment intensity information and puffing treatment time information.

[0099] According to the embodiment of the present application, the combustion process deviation information is compared and analyzed, and production process correction information is generated. Specifically,

[0100] The smoke color deviation value is obtained based on the smoke color deviation information. The smoke color deviation value is subjected to numerical averaging processing to obtain a color average deviation value.

[0101] It is determined whether the color average deviation value is less than a preset threshold value.

[0102] If it is less than, the color average deviation value is introduced into a process correction model for data analysis to obtain impurity removal process correction information.

[0103] It should be noted that the smoke color deviation value includes multiple sets of values. The production process correction information includes puffing process correction information and impurity removal process correction information. In the smoke color deviation value obtained based on the smoke color deviation information, the impurity content in the tobacco sample can be effectively determined by analyzing the smoke color deviation information, so as to determine whether the tobacco raw material needs to be subjected to impurity removal operation. If the impurity content of the tobacco sample is high, the color of the smoke generated during combustion will be black or white, and the color average deviation value obtained by analysis will also be larger. In addition, in the determination of whether the color average deviation value is less than a preset threshold value, if the color average deviation value is greater than or equal to the preset threshold value, it means that the impurity content of the tobacco sample is high, and the tobacco raw material of the corresponding batch needs to be subjected to complete screening and processing to meet the qualified tobacco standard. The impurity removal process correction information includes impurity removal time information and impurity removal method information.

[0104] According to the embodiment of the present application, the production process correction information is sent to a preset terminal device for display. The preset terminal device includes a preset mobile terminal device and a preset computer terminal device.

[0105] Figure 4 A block diagram of a tobacco combustion detection system based on image recognition is shown.

[0106] The second aspect of the present application also provides an image recognition-based tobacco combustion detection system 4, comprising a memory 41 and a processor 42, wherein the memory comprises an image recognition-based tobacco combustion detection method program, and the image recognition-based tobacco combustion detection method program is executed by the processor to implement the following steps:

[0107] obtaining a tobacco raw material sample to be detected, performing a combustion test on the tobacco raw material sample, and obtaining combustion process image information;

[0108] obtaining reference image information of a good-quality raw material combustion process, performing image analysis on the combustion process image information and the reference image information, comparing, and obtaining combustion process deviation information;

[0109] performing data comparison and analysis on the combustion process deviation information, and generating manufacturing process correction information;

[0110] sending the manufacturing process correction information to a preset terminal device for display.

[0111] According to the embodiment of the present application, the obtaining of the tobacco raw material sample to be detected, the combustion test on the tobacco raw material sample, and the obtaining of the combustion process image information comprises the following steps:

[0112] obtaining different batches of tobacco raw materials, and numbering the different batches of tobacco raw materials;

[0113] obtaining tobacco detection raw materials of each batch according to the numbering sequence;

[0114] dividing the tobacco detection raw materials into three groups on average, and making cigarettes from the tobacco detection raw materials of each group to obtain the tobacco raw material sample to be detected.

[0115] It should be noted that the tobacco raw materials of the same batch are generally relatively consistent in quality, while the tobacco raw materials of different batches are different in quality due to different tobacco raw materials and inconsistent processing times, and therefore need to be processed in a targeted manner in subsequent processing to obtain better cut tobacco and cigarette products. In the obtaining of the tobacco raw material sample to be detected, the weight of the tobacco raw material sample to be detected of each batch is consistent, and in the dividing of the tobacco detection raw materials into three groups on average, the grouping is based on weight average division.

[0116] According to the embodiment of the present application, the obtaining of the tobacco raw material sample to be detected, the combustion test on the tobacco raw material sample, and the obtaining of the combustion process image information specifically comprises the following steps:

[0117] placing the tobacco raw material sample in a preset experimental environment for combustion test;

[0118] According to a preset time interval, the tobacco raw material sample combustion original image in each time interval is acquired by the camera device;

[0119] The combustion original image is subjected to image enhancement processing, and the processed original image is subjected to data arrangement to obtain the combustion process image information.

[0120] It should be noted that in the acquisition of the tobacco raw material sample combustion original image in each time interval, the acquired combustion original image contains a plurality of image data, and the data amount of the original image is affected by the preset time interval, and the time interval is generally 30-50 seconds. The combustion original image is subjected to image enhancement processing to highlight the color features of the smoke smoke in the combustion process, so that subsequent image recognition and processing can be better performed.

[0121] According to the embodiment of the present application, the reference image information of the good product raw material combustion process is acquired, the combustion process image information is compared with the reference image information for image analysis, and the combustion process deviation information is obtained, specifically:

[0122] The combustion process image information is subjected to image recognition analysis, and the smoke in the combustion process is subjected to image feature extraction to obtain smoke feature data;

[0123] The smoke feature data is subjected to data analysis to obtain smoke shape information, and according to the smoke shape information, the smoke shape change information is obtained in combination with the change of the combustion process;

[0124] The smoke shape change information is compared with the reference image information for data analysis to obtain the smoke shape deviation information.

[0125] It should be noted that the smoke shape deviation information is the deviation information obtained by comparing the reference image information. The reference image information of the good product raw material combustion process is the image information acquired in the combustion process of the high-quality finished cut tobacco selected by artificial, and the image information contains reference smoke shape change information and has good reference significance.

[0126] According to the embodiment of the present application, the reference image information of the good product raw material combustion process is acquired, the combustion process image information is compared with the reference image information for image analysis, and the combustion process deviation information is obtained, and further comprising:

[0127] The combustion process image information is acquired and the smoke color feature value in the combustion process image information is extracted;

[0128] The smoke color feature value is subjected to data summarization, arrangement and analysis to obtain the smoke color change information;

[0129] The smoke color change information is compared with the reference image information through data analysis to obtain smoke color deviation information.

[0130] It should be noted that the combustion process deviation information includes smoke shape deviation information and smoke color deviation information. The smoke color feature value is a plurality of feature values, and different groups of feature values correspond to image information of different time periods in the combustion process. The smoke color change information is color change information of the smoke in the combustion process, which can effectively reflect the color concentration change of the smoke. By studying the color change of the smoke, the quality of the tobacco sample can be further analyzed. In addition, the reference image information is image information obtained in the combustion process of high-quality finished cut tobacco artificially selected, and the image information contains reference smoke color change information, which has good reference significance.

[0131] According to the embodiment of the present application, the combustion process deviation information is compared and analyzed through data comparison, and manufacturing process correction information is generated, which further comprises:

[0132] According to the preset combustion time interval, image information within a preset start time and image information within a preset end time in the combustion process are obtained;

[0133] The image information is summarized to obtain start image data and end image data;

[0134] The start image data and the end image data are analyzed to obtain the initial change time of the smoke and the end change time of the smoke;

[0135] The initial change time of the smoke and the end change time of the smoke are analyzed to obtain tobacco humidity correction information.

[0136] It should be noted that the tobacco raw material has a certain hygroscopicity, and contains moisture inside. The amount of moisture has a certain influence on the transportation, storage, processing of the tobacco raw material and the quality of the tobacco raw material. By studying the change of the smoke within 30 to 40 seconds at the beginning and end of the combustion process of the tobacco raw material sample, the humidity content of the current tobacco raw material can be accurately analyzed to obtain the tobacco humidity correction information, thereby achieving the purpose of improving the quality of the tobacco raw material.

[0137] In addition, the preset combustion time interval is 2 to 3 seconds, the preset start time and the preset end time are 30 to 40 seconds, and the start image data and the end image data each contain a plurality of image data. The size of the image data is affected by the size of the preset combustion time interval, and the tobacco humidity correction information includes the content and concentration of the added humectant.

[0138] According to the embodiment of the present application, the reference image information of the good-quality raw material combustion process is acquired, the combustion process image information is analyzed, compared with the reference image information, and the combustion process deviation information is obtained, and the method further comprises:

[0139] The combustion process deviation information in different batches of tobacco raw material samples is acquired;

[0140] The combustion process deviation information is re-numbered according to the numbers of different batches of tobacco raw material, and a plurality of groups of combustion process deviation information is obtained;

[0141] Each two groups of combustion process deviation information are analyzed and compared, and a data difference rate in the two groups of combustion process deviation information is obtained;

[0142] If the data difference rate is lower than a preset difference rate, the two groups of combustion process deviation information corresponding to the batches of tobacco raw material are merged.

[0143] It should be noted that in the plurality of groups of combustion process deviation information, each group of combustion process deviation information corresponds to the combustion process deviation information in a batch of tobacco raw material samples, and each group of combustion process deviation information corresponds to a number, which is consistent with the number of the corresponding batch of tobacco raw material. In the data difference rate is lower than a preset difference rate, if the data difference rate is lower than the preset difference rate, it means that the quality of the corresponding two batches of tobacco raw material is relatively consistent, and the two batches of tobacco raw material are merged. Through the batch merging operation, the consumption of manpower and material resources in subsequent processing can be reduced, and the redundant processing steps caused by batch processing of raw materials can be reduced.

[0144] According to the embodiment of the present application, the combustion process deviation information is compared and analyzed, and the manufacturing process correction information is generated, specifically:

[0145] According to the data splitting and data extraction of the smoke shape deviation information, the smoke size deviation value and the smoke length deviation value are obtained;

[0146] The smoke size deviation value and the smoke length deviation value are subjected to numerical averaging processing, and the processed deviation average value is introduced into the process correction model;

[0147] The process correction model comprehensively analyzes the deviation average value, and obtains the puffing process correction information.

[0148] It should be noted that the smoke size deviation value and the smoke length deviation value each include multiple sets of values. In addition, data analysis based on the obtained smoke size deviation value and the smoke length deviation value can obtain puffing process correction information of the tobacco raw material. Puffing treatment can increase the bulkiness of the tobacco raw material, increase the volume, reduce the weight, and make the combustion more complete. The size and length of the smoke generated by combustion are closer to the size and length of the smoke in the reference image information. The puffing process correction information includes puffing treatment intensity information and puffing treatment time information.

[0149] According to the embodiment of the present application, the combustion process deviation information is compared and analyzed, and production process correction information is generated. Specifically,

[0150] The smoke color deviation value is obtained based on the smoke color deviation information. The smoke color deviation value is subjected to numerical averaging processing to obtain a color average deviation value.

[0151] It is determined whether the color average deviation value is less than a preset threshold value.

[0152] If the color average deviation value is less than the preset threshold value, the color average deviation value is imported into a process correction model for data analysis to obtain impurity removal process correction information.

[0153] It should be noted that the smoke color deviation value includes multiple sets of values. The production process correction information includes puffing process correction information and impurity removal process correction information. In the smoke color deviation value obtained based on the smoke color deviation information, the impurity content in the tobacco sample can be effectively determined by analyzing the smoke color deviation information, so as to determine whether the tobacco raw material needs to be subjected to impurity removal operation. If the impurity content of the tobacco sample is high, the color of the smoke generated during combustion will be black or white, and the color average deviation value obtained by analysis will also be larger. In addition, in the determination of whether the color average deviation value is less than the preset threshold value, if the color average deviation value is greater than or equal to the preset threshold value, it indicates that the impurity content of the tobacco sample is high, and the tobacco raw material of the corresponding batch needs to be subjected to complete screening and processing to meet the qualified tobacco standard. The impurity removal process correction information includes impurity removal time information and impurity removal method information.

[0154] According to the embodiment of the present application, the production process correction information is sent to a preset terminal device for display. The preset terminal device includes a preset mobile terminal device and a preset computer terminal device.

[0155] The application discloses a tobacco combustion detection method and system based on image recognition, and obtains image information of different time intervals in a tobacco sample combustion process through a camera device, analyzes tobacco sample smoke change information in the combustion process according to the image information of different time intervals, compares the smoke change information with existing reference image information, obtains correction information of a tobacco production process, and thus the puffing process and the impurity removal process in tobacco production can be adjusted, the redundant process steps are reduced, the manpower and material resources are saved, and finally the beneficial effects of improving the quality of tobacco finished products and reducing production costs are achieved.

[0156] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other manners. The above described device embodiments are merely schematic. For example, the division of the units is only a logical function division. There can be another division manner for the actual implementation. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the various components shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, and can be electrical, mechanical or other forms.

[0157] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units; they can be located in one place, or distributed on multiple network units; and part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0158] In addition, each functional unit in each embodiment of the present application can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be realized in the form of hardware or in the form of hardware plus software functional unit.

[0159] Those skilled in the art can understand that all or part of the steps of the above method embodiments can be completed by program instruction related hardware, and the foregoing program can be stored in a computer readable storage medium, and the program is executed to perform the steps of the above method embodiments; and the foregoing storage medium includes mobile storage devices, read-only memories (ROMs), random access memories (RAMs), magnetic discs or optical discs and various storage media that can store program codes.

[0160] Alternatively, the above-mentioned integrated unit of the present application, if realized in the form of a software function module and sold or used as an independent product, can also be stored in a computer-readable storage medium. Based on such an understanding, the technical solutions of the embodiments of the present application can be embodied in the form of a software product in essence or in the form of a part of the prior art that contributes to the present application. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the method described in the embodiments of the present application. The aforementioned storage medium includes mobile storage devices, ROM, RAM, magnetic discs or optical discs, and various media that can store program codes.

[0161] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. An image recognition-based tobacco combustion detection method, characterized by, The method comprises the following steps: obtaining a tobacco raw material sample to be detected, performing a combustion test on the tobacco raw material sample, and obtaining combustion process image information; obtaining reference image information of a good product raw material combustion process, performing image analysis on the combustion process image information and the reference image information, comparing them, and obtaining combustion process deviation information; performing data comparison and analysis on the combustion process deviation information, and generating production process correction information; sending the production process correction information to a preset terminal device for display; wherein the combustion process deviation information is obtained by: performing image recognition analysis on the combustion process image information, and extracting image features of the smoke in the combustion process to obtain smoke feature data; performing data analysis on the smoke feature data to obtain smoke shape information, and obtaining smoke shape change information based on the smoke shape information and the change in the combustion process; performing data analysis and comparison on the smoke shape change information and the reference image information to obtain smoke shape deviation information; obtaining the combustion process image information and extracting smoke color feature values from the combustion process image information; performing data summarization, arrangement, and analysis on the smoke color feature values to obtain smoke color change information; performing data analysis and comparison on the smoke color change information and the reference image information to obtain smoke color deviation information; wherein the production process correction information is generated by: performing data splitting and data extraction based on the smoke shape deviation information to obtain smoke size deviation values and smoke length deviation values; performing numerical averaging processing on the smoke size deviation values and the smoke length deviation values, and importing the processed deviation average values into a process correction model; performing comprehensive analysis on the deviation average values by the process correction model to obtain puffing process correction information; obtaining smoke color deviation numerical values based on the smoke color deviation information, performing numerical averaging processing on the smoke color deviation numerical values to obtain color average deviation values; determining whether the color average deviation values are less than a preset threshold value; if so, importing the color average deviation values into the process correction model for data analysis to obtain impurity removal process correction information.

2. The image recognition-based tobacco combustion detection method according to claim 1, characterized by, The method further comprises the following steps before obtaining the tobacco raw material sample to be detected, performing a combustion test on the tobacco raw material sample, and obtaining combustion process image information: obtaining different batches of tobacco raw materials, and numbering the different batches of tobacco raw materials; obtaining tobacco detection raw materials according to the numbering sequence; dividing the tobacco detection raw materials into three groups, and making each group of tobacco detection raw materials into cigarettes to obtain the tobacco raw material sample to be detected.

3. The image recognition-based tobacco combustion detection method of claim 1, wherein, The method further comprises the following steps before obtaining the tobacco raw material sample to be detected, performing a combustion test on the tobacco raw material sample, and obtaining combustion process image information: placing the tobacco raw material sample in a preset experimental environment for combustion testing; obtaining tobacco raw material sample combustion original images in each time interval through a camera device according to a preset time interval. The combustion original image is subjected to image enhancement processing, and the processed original image is subjected to data arrangement to obtain combustion process image information.

4. An image recognition based tobacco combustion detection system, characterized in that, The system comprises a memory and a processor, the memory comprises a tobacco combustion detection method based on image recognition program, and the processor executes the tobacco combustion detection method based on image recognition program to realize the steps of the tobacco combustion detection method based on image recognition in claim 1.

5. An image recognition based tobacco combustion detection system according to claim 4, wherein, The tobacco raw material sample to be detected is acquired, the tobacco raw material sample is subjected to combustion test, and combustion process image information is obtained, specifically as follows: The tobacco raw material sample is placed in a preset experimental environment for combustion test; According to a preset time interval, a camera device is used to acquire combustion original images of the tobacco raw material sample in each time interval; The combustion original image is subjected to image enhancement processing, and the processed original image is subjected to data arrangement to obtain combustion process image information.

Images