A cigarette quality scoring method based on multiple sensory parameters

Abstract

The present invention discloses a cigarette quality scoring method based on multiple sensory parameters, comprising the following steps: (1) machine scoring; (2) manual scoring; (3) comprehensive scoring. The cigarette quality scoring method based on multiple sensory parameters of the technical solution combines machine scoring with manual scoring, and uses a scoring team to score and record samples in turn with a full score of 100. If there are samples with consistent scores, a re-evaluation is performed, and the re-evaluation score is used as the result. The machine score and the manual score are then added to obtain an average value, which is the final score. If there are samples with consistent scores, a re-evaluation is performed. If the re-evaluation scores are still the same, the score of the manual score is used as the result, thereby improving the rigor of the scoring.

Description

Technical Field

[0001] The invention relates to the technical field of cigarette quality scoring, in particular to a cigarette quality scoring method based on multiple sensory parameters. Background Art

[0002] Recently, the tobacco industry mainly describes the quality of cigarette cut tobacco through the whole-cut ratio, long-cut ratio, dust content, moisture content, and sensory quality of cigarette cut tobacco. By defining the factors that affect the appearance quality of cigarette cut tobacco and testing the appearance quality of tobacco cut tobacco, it is beneficial to the study of the appearance quality of cigarette cut tobacco and can also solve the problems that affect the quality of cigarette cut tobacco in production. At present, the domestic sensory evaluation method is generally adopted for the evaluation of the sensory quality of cigarettes, and it is judged according to the national standard. The current method is well known in the industry and can be used well. It is a better method for checking the compliance of cigarettes and comparing their characteristics.

[0003] However, it also has certain limitations. The smoking evaluation conclusions obtained using this method are not easy to intuitively determine the differences in the sensory characteristics of different cigarette products. Moreover, it relies entirely on manual scoring, which is prone to mistakes and affects the rigor of quality scoring. Summary of the invention

[0004] The object of the present invention is to provide a cigarette quality scoring method based on multiple sensory parameters to solve the problems raised in the above background technology.

[0005] To achieve the above objectives, this application is implemented through the following technical solutions:

[0006] A cigarette quality scoring method based on multiple sensory parameters comprises the following steps:

[0007] (1) Machine scoring;

[0008] (2) Human scoring

[0009] a. Establish a scoring team;

[0010] b. Sampling: first classify according to brand, specification, batch number, packaging, etc., and then randomly select a certain number of samples as identification and inspection samples according to the requirements of the article;

[0011] c. Sample sealing record: After sampling, the sampling team seals the sample, fills in the sampling form, and makes necessary records on the sampling form;

[0012] d. Scoring record: the scoring team will score the samples in turn with a full score of 100 and record the scores. If there are samples with the same scores, they will be re-evaluated and the re-evaluation score will be used as the result;

[0013] (3) Comprehensive score

[0014] The machine score and the manual score are added together to get the average value, which is the final score. If there are samples with the same score, a re-evaluation will be carried out; if the re-evaluation scores are still the same, the manual score will be used as the result.

[0015] Furthermore, the specific steps of step (1) are as follows:

[0016] 11) Establish a quality scoring system, identify the real-time data of qualified cigarette appearance in a steady state, establish a data sample for research, analyze the correlation between parameters and indicators of sample data, use statistical modeling methods to obtain factors affecting key quality indicators, screen out key scoring indicators, and based on the screened key evaluation indicators, combine the degree of influence of quality indicators, objectively weight them, and establish a cigarette appearance quality scoring system;

[0017] 12) Integrate quantitative indicators, classify images of qualified and unqualified cigarette appearance, normalize these images into the same size and name them, classify the separately collected cigarette quality indicators and parameters, integrate the names of the collected cigarette quality indicator parameters, and quantify the quality indicators;

[0018] 13) Extract feature information, input the characteristics of the cigarette into the generator to extract the feature information, and output the corresponding feature probability map;

[0019] 14) Comparative scoring: Based on the cigarette quality index parameter set, the color of tobacco leaves and cigarettes, as well as the shape, moisture, length and diameter of cigarettes are scored in turn, and the scoring results are output.

[0020] Furthermore, the characteristics of the cigarette in step 13) include physical property inspection, mainstream smoke analysis, and conventional chemical analysis.

[0021] Furthermore, the content of step a of step (2) is as follows: the sensory scoring is completed in a scoring group, and a scoring group is composed of 8-10 people, with a group leader. The group members have undergone rigorous and detailed training in advance, are familiar with the cigarette sensory scoring process, and are good at perceiving the general aroma, taste and overall feeling. They must also be familiar with the aroma of standard spices. After the calibrated calibrant, the members of the same group tend to have the same perception of the sensory quality characteristics and style characteristics.

[0022] Furthermore, step a of step (2) includes the following five situations:

[0023] 21) For packs with less than 50 cigarettes, randomly select 1-2 cigarettes from each pack to form a sample, and then randomly select 2 cigarettes from all the samples as test pieces;

[0024] 22) If the quantity is less than 2 pieces, 1-2 pieces shall be randomly selected from all samples as test pieces;

[0025] 23) If the number is less than 2, select 1 or all of them as the sample; if the number is between 5 and 10, randomly select 1 from each piece to form a sample, and then randomly select 2 from the sample as the sample;

[0026] 24) If the quantity is between 10 and 50 pieces, randomly select 10 pieces, randomly select 1 piece from each piece to form a sample, and then randomly select 2 to 5 pieces from the sample as test pieces;

[0027] 25) If the quantity is more than 50 pieces, randomly select 20 pieces, randomly select 1 piece from each piece to form a sample, and then randomly select 5-10 pieces from the sample as test pieces.

[0028] The beneficial effects of the present invention are:

[0029] 1. The present invention combines machine scoring with manual scoring. The scoring team scores the samples in turn with a full score of 100 and records them. If there are samples with the same score, a re-evaluation is performed, and the re-evaluation score is used as the result. The machine score and the manual score are added to obtain the average value as the final score. If there are samples with the same score, a re-evaluation is performed. If the re-evaluation scores are still the same, the score of the manual score is used as the result, thereby improving the rigor of the scoring;

[0030] 2. After sampling, the sampling team seals the sample, fills in a sampling sheet, makes necessary records on the sampling sheet, and records on the corresponding sampling sheet during scoring. The data can be checked to intuitively determine the differences in sensory characteristics of different cigarette products. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a system block diagram of the present invention.

[0032] Figure 2 It is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] like Figure 1As shown, a cigarette quality scoring method based on multiple sensory parameters of the present application is implemented based on a scoring system. The scoring system of the present application includes a monitoring module, a data acquisition module, a data upload module, a data conversion module, a data comparison module, a data analysis module, a data sharing module, a database, an early warning module, a notification module and a central controller.

[0035] The monitoring module is electrically connected to the data acquisition module, the data acquisition module, the data upload module, the data conversion module, the data comparison module, and the data analysis module are electrically connected in sequence, the data analysis module is directly electrically connected to the data sharing module, the database, the early warning module, and the notification module, and the data analysis module and the data sharing module are directly electrically connected to the database.

[0036] The monitoring module includes a real-time monitoring module and a random monitoring module. The real-time monitoring module is directly electrically connected to the early warning module and the notification module, and the random monitoring module is directly electrically connected to the data comparison module, the data analysis module, and the data sharing module.

[0037] The data acquisition module includes the acquisition of image, graphic, color, quality, size and humidity data, and is equipped with a camera, infrared sensor, gravity sensor, distance sensor and humidity sensor.

[0038] The data upload module includes the scored cigarette data collected by the data collection module and the quality index data set and updated in real time.

[0039] The data comparison module compares the information after the data conversion module with the quality indicator parameter information actually set. After the information is analyzed by the data analysis module, two situations occur. One is that it meets the quality indicators set actually, which can be shared or uploaded to the database through the data sharing module. The other is that it does not meet the quality indicators set actually. A warning is issued through the early warning module and notification module and checked by the staff.

[0040] like Figure 1 As shown, the present application provides a cigarette quality scoring method based on multiple sensory parameters, comprising the following steps:

[0041] (1) Machine scoring: The steps of machine scoring are as follows:

[0042] 11) Establish a quality scoring system, conduct steady-state identification of real-time data on qualified cigarette appearance, establish data samples for research, analyze the correlation between parameters and indicators of sample data, use statistical modeling methods to obtain factors affecting key quality indicators, screen out key scoring indicators, and objectively weight them based on the screened key evaluation indicators combined with the degree of influence of quality indicators to establish a cigarette appearance quality scoring system.

[0043] 12) Integrate quantitative indicators, classify images of qualified and unqualified cigarette appearance, normalize these images to the same size and name them, classify the separately collected cigarette quality indicators and parameters, integrate the names of the collected cigarette quality indicator parameters, and quantify the quality indicators.

[0044] 13) Extract characteristic information, input the characteristics of cigarettes into the generator to extract characteristic information, and output the corresponding characteristic probability map; the characteristics of cigarettes include physical property inspection, mainstream smoke analysis, and conventional chemical analysis.

[0045] 14) Comparative scoring: Based on the cigarette quality index parameter set, the color of tobacco leaves and cigarettes, as well as the shape, moisture, length and diameter of cigarettes are scored in turn, and the scoring results are output.

[0046] (2) Human scoring

[0047] a. Establish a scoring team: Sensory scoring is completed in scoring teams. A scoring team consists of 8-10 people and a team leader. Team members undergo rigorous and detailed training in advance, are familiar with the sensory scoring process of cigarettes, and are good at perceiving the general aroma, taste and overall feeling. They must also be familiar with the aroma of standard spices. Members of the same team tend to have consistent feelings about sensory quality characteristics and style characteristics after calibrating the caliber.

[0048] b. Sampling: first classify according to brand, specification, batch number, packaging, etc., and then randomly select a certain number of samples as identification and inspection samples according to the requirements of the article; the specific steps of sampling include the following five situations:

[0049] 21) For packs with less than 50 cigarettes, randomly select 1-2 cigarettes from each pack to form a sample, and then randomly select 2 cigarettes from all the samples as test pieces;

[0050] 22) If the quantity is less than 2 pieces, 1-2 pieces shall be randomly selected from all samples as test pieces;

[0051] 23) If the number is less than 2, select 1 or all of them as the sample; if the number is between 5 and 10, randomly select 1 from each piece to form a sample, and then randomly select 2 from the sample as the sample;

[0052] 24) If the quantity is between 10 and 50 pieces, randomly select 10 pieces, randomly select 1 piece from each piece to form a sample, and then randomly select 2 to 5 pieces from the sample as test pieces;

[0053] 25) If the quantity is more than 50 pieces, randomly select 20 pieces, randomly select 1 piece from each piece to form a sample, and then randomly select 5-10 pieces from the sample as test pieces.

[0054] c. Sealing record: After sampling, the sampling team will seal the sample, fill in the sampling sheet, and make necessary records on the sampling sheet.

[0055] d. Scoring record: the scoring team will score the samples in turn with a full score of 100 and record the scores. If there are samples with the same scores, they will be re-evaluated and the re-evaluation score will be used as the result;

[0056] (3) Comprehensive score

[0057] The machine score and the manual score are added together to get the average value, which is the final score. If there are samples with the same score, a re-evaluation will be carried out; if the re-evaluation scores are still the same, the manual score will be used as the result.

[0058] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A cigarette quality scoring method based on multiple sensory parameters, characterized in that: The following steps are involved: (1) Machine scoring; (2) Human scoring a. Establish a scoring team; b. Sampling: first classify according to brand, specification, batch number and packaging, and then randomly select a certain number of samples as identification and inspection samples according to the requirements of the article; c. Sample sealing record: After sampling, the sampling team seals the sample, fills in the sampling form, and makes necessary records on the sampling form; d. Scoring record: the scoring team will score the samples in turn with a full score of 100 and record the scores. If there are samples with the same scores, they will be re-evaluated and the re-evaluation score will be used as the result; (3) Comprehensive score The machine score and the human score are added together to get the average value, which is the final score. If there are samples with the same score, they will be re-evaluated. If the re-evaluation scores are still the same, the human score will be used as the result. The specific steps of step (1) are as follows: 11) Establish a quality scoring system, identify the real-time data of qualified cigarette appearance in a steady state, establish a data sample for research, analyze the correlation between parameters and indicators of sample data, use statistical modeling methods to obtain factors affecting key quality indicators, screen out key scoring indicators, and based on the screened key evaluation indicators, combine the degree of influence of quality indicators, objectively weight them, and establish a cigarette appearance quality scoring system; 12) Integrate quantitative indicators, classify images of qualified and unqualified cigarette appearance, normalize these images into the same size and name them, classify the separately collected cigarette quality indicators and parameters, integrate the names of the collected cigarette quality indicator parameters, and quantify the quality indicators; 13) Extract feature information, input the characteristics of the cigarette into the generator to extract the feature information, and output the corresponding feature probability map; 14) Comparative scoring: according to the cigarette quality index parameter set, the color of tobacco leaves and cigarettes, as well as the shape, humidity, length and diameter of cigarettes are scored in turn, and the scoring results are output; The characteristics of the cigarette in step 13) include physical characteristics inspection, mainstream smoke analysis, and conventional chemical analysis; The cigarette quality scoring method based on multiple sensory parameters is implemented based on a scoring system, which includes a monitoring module, a data acquisition module, a data upload module, a data conversion module, a data comparison module, a data analysis module, a data sharing module, a database, an early warning module, a notification module and a central controller; The monitoring module is electrically connected to the data acquisition module, the data acquisition module, the data upload module, the data conversion module, the data comparison module, and the data analysis module are electrically connected in sequence, the data analysis module is electrically directly connected to the data sharing module, the database, the early warning module, and the notification module, and the data analysis module and the data sharing module are electrically directly connected to the database; The monitoring module includes a real-time monitoring module and a random monitoring module. The real-time monitoring module is directly connected to the early warning module and the notification module through electrical connection, and the random monitoring module is directly connected to the data comparison module, the data analysis module and the data sharing module through electrical connection; The data acquisition module includes the acquisition of image, graphic, color, quality, size and humidity data, and is equipped with a camera, infrared sensor, gravity sensor, distance sensor and humidity sensor; The data upload module includes the scored cigarette data collected by the data collection module and the quality index data set and updated in real time; The data comparison module compares the information after the data conversion module with the quality indicator parameter information actually set. After the information is analyzed by the data analysis module, two situations occur. One is that it meets the quality indicators set actually, which can be shared or uploaded to the database through the data sharing module. The other is that it does not meet the quality indicators set actually. A warning is issued through the early warning module and notification module and checked by the staff.

2. The cigarette quality scoring method based on multiple sensory parameters according to claim 1, characterized in that: The content of step a of step (2) is as follows: the sensory scoring is completed in a scoring group. A scoring group consists of 8-10 people and a group leader. The group members have undergone rigorous and detailed training in advance, are familiar with the cigarette sensory scoring process, and are good at perceiving the general aroma, taste and overall feeling. They must also be familiar with the aroma of standard spices. After calibration of the caliber, the members of the same group tend to have the same perception of the sensory quality characteristics and style characteristics.

3. The cigarette quality scoring method based on multiple sensory parameters according to claim 1, characterized in that: Step b of step (2) includes the following five situations: 21) For packs of 5, i.e. less than 50 cartons, randomly select 1-2 cartons from each pack to form a sample, and then randomly select 2 cartons from all the samples as test pieces; 22) If the quantity is less than 2 pieces, 1-2 pieces shall be randomly selected from all samples as test pieces; 23) When the quantity is less than 2, select 1 or all of them as the sample; when the quantity is between 5 and 10, randomly select 1 from each piece to form a sample, and then randomly select 2 from the sample as the sample; 24) If the quantity is between 10 and 50 pieces, randomly select 10 pieces, randomly select 1 piece from each piece to form a sample, and then randomly select 2 to 5 pieces from the sample as test pieces; 25) If the quantity is more than 50 pieces, randomly select 20 pieces, randomly select 1 piece from each piece to form a sample, and then randomly select 5-10 pieces from the sample as test pieces.

Images