Tobacco leaf production line sampling monitoring system and monitoring method

Abstract

The invention provides a tobacco leaf production line sampling monitoring system and monitoring method, and belongs to the field of tobacco leaf production monitoring, and the tobacco leaf production line sampling monitoring system comprises a quantity sampling module and an uploading monitoring module which are in telecommunication connection with an image recognition module; the weighing and sampling module comprises a baking machine with a tobacco leaf sampling box, a fixed card punching device and a weighing device; the uploading monitoring module comprises a fixing device and a monitoring camera; the image recognition module comprises a transmission device, an alarm device and a processor and is used for monitoring operation steps and operation specifications of workers and forming statistical reports of samples and the workers. Through the system and the method, material loss can be reduced, tobacco production quality can be improved, operation of personnel is standardized, and the system and the method are convenient to popularize and apply in production lines of tobacco, agricultural products and the like.

Description

Technical Field

[0001] The invention belongs to the field of tobacco production line monitoring, and in particular relates to a tobacco production line sampling monitoring system and a monitoring method. Background Art

[0002] In the process of tobacco production and processing in tobacco factories, accurate measurement of the leaf structure of threshed tobacco is an important step to ensure the quality of tobacco leaves. Through accurate measurement of the leaf structure of threshed tobacco leaves, problems in the production process can be discovered and solved in a timely manner, such as leaves that are too large, too small, or too high in stem content, thereby improving the uniformity and consistency of re-roasted tobacco leaves and ultimately improving the overall quality of tobacco products.

[0003] When inspecting the leaf structure of threshed tobacco leaves, the existing production line requires workers to manually operate the weighing device to intercept and weigh the tobacco leaves in production at one time. During this process, it is impossible to monitor the workers' operating steps. The workers are prone to negligence and carelessness, which may lead to irregular operations, resulting in uneven structure of the samples taken, affecting the quality of the produced tobacco. Therefore, there is an urgent need for a monitoring system that can standardize the weighing operations of the workers. Summary of the invention

[0004] The purpose of the present invention is to provide a production line sampling monitoring system and method thereof, which can capture the operation process of the staff in real time, make the staff more cautious, and improve the staff's operation standardization, in order to achieve the above purpose, the present invention adopts the following technical solutions.

[0005] A tobacco production line sampling monitoring system, the tobacco production line sampling monitoring system comprises a weighing sampling module and an uploading monitoring module which are telecommunication-connected to an image recognition module; the weighing sampling module comprises a roasting machine with a tobacco sampling box, a fixed punching device and a weighing device; the uploading monitoring module comprises a fixing device and a monitoring camera; the monitoring camera is installed on the weighing device through the fixing device, and is used to monitor the operating steps of the staff, and transmit the operating process of the staff to the image recognition module in the form of video, so as to monitor the operating steps of the staff; the image recognition module comprises a transmission device, an alarm device and a processor, and the image recognition module identifies the irregular operation of the staff according to the operation video fed back by the uploading monitoring module, takes a screenshot and issues an alarm, and generates statistics based on the data transmitted by the weighing sampling module to form a report.

[0006] Preferably, the weighing device samples and weighs the tobacco at a fixed point, at a fixed time, and in a fixed quantity.

[0007] Preferably, a weighing alarm light is installed on the side of the weighing device, and the upper and lower weighing limits can be set according to the sampling requirements. When the weighing weight exceeds the range, the weighing alarm light flashes and an alarm sound is heard.

[0008] Preferably, the weighing device samples the tobacco once when weighing and sampling.

[0009] Preferably, the transmission device transmits the data recorded by the weighing device, including weighing data, weighing time, staff, and category parameters, to the processor, and generates a report through statistics.

[0010] Preferably, the alarm device identifies the operating steps and attire of the staff based on the video uploaded by the monitoring module, and warns the staff and management personnel through warning lights or buzzer alarms when the staff operate and wear improperly.

[0011] Preferably, the processor of the image recognition module uses a deep learning model to perform graphic detection, and sets a detection threshold, a frame counter, and a continuous operation threshold for the model.

[0012] The present invention also provides a monitoring method based on the above-mentioned tobacco production line sampling monitoring system, and the monitoring method includes the following steps.

[0013] S1. Tobacco weighing: The tobacco at the exit of the roasting machine is weighed by a weighing device.

[0014] S2. Operation monitoring: The monitoring camera fixedly installed on the weighing device captures the worker's clothing and operating steps and transmits them to the image recognition module.

[0015] S3. Image recognition and report generation: The processor processes and compiles statistics on the recorded data including weighing data, weighing time, staff, and category parameters to form reports, and identifies the staff's operating steps and wear. When the staff's operation and wear are not standardized, the alarm device will be used to warn the staff and management personnel.

[0016] Compared with the existing technology, the Bunsen monitoring system and monitoring method include the following advantages.

[0017] 1. The tobacco at the exit of the roasting machine is intercepted and weighed at one time through a weighing device. The qualified samples are separated into leaves of different sizes through a blade vibration sorting screen. According to the proportion of leaves of different sizes, the operator is guided to accurately control the structure of the re-roasted leaves, reduce the production of unqualified products, and reduce material loss.

[0018] 2. The present invention can capture the operation process of the staff in real time by setting up an upload monitoring module, so that the staff can be more cautious and improve the standardization of the staff's operation, avoid the staff's irregular operation due to negligence and carelessness, make the tobacco structure detection more accurate, and improve the quality of the produced tobacco.

[0019] 3. By setting up an image recognition module, the present invention can count and form reports on the weighing data, weighing time, staff, categories, etc. transmitted by the weighing sampling module, which is convenient for managers to record the data. At the same time, according to the video feedback from the uploaded monitoring module, the operating steps and wear of the staff are identified and screenshoted, and the staff and managers are warned in time, so as to further improve the standardization of the operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic diagram of the structural composition of the monitoring system of the present invention.

[0021] In the figure: 1. Weighing and sampling module; 2. Upload monitoring module; 3. Image recognition module; 4. Baking machine; 5. Fixed punching device; 6. Weighing device; 7. Fixing device; 8. Monitoring camera; 9. Transmission device; 10. Alarm device. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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.

[0023] It should be understood that the "system", "device", "unit" and / or "module" used in this specification is a method for distinguishing different components, elements, parts, parts or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0024] Flowcharts are used in this specification to illustrate the operations performed by the system according to the embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed precisely in order. Instead, the steps may be processed in reverse order or simultaneously. At the same time, other operations may also be added to these processes, or one or more operations may be removed from these processes.

[0025] Monitoring system A tobacco production line sampling monitoring system, see Figure 1The tobacco production line sampling monitoring system includes a weighing sampling module 1 and an uploading monitoring module 2 which are connected to an image recognition module 3 by telecommunication. Specifically, the weighing sampling module 1 and the uploading monitoring module 2 are connected to the image recognition module 3 via a network.

[0026] The weighing and sampling module 1 includes a tobacco leaf sampling box 4, a fixed punching device 5 and a weighing device 6. The weighing and sampling module 1 is used to transmit data to the image recognition module 3.

[0027] Among them, the upload monitoring module 2 includes a fixing device 7 and a monitoring camera 8; the monitoring camera 8 is installed on the weighing device 6 through the fixing device 7, and is used to monitor the operating steps of the staff, and transmit the operating process of the staff to the image recognition module 3 in the form of video, so as to realize the monitoring of the operating steps of the staff; in view of the problem that the staff in the prior art is prone to irregular operation due to negligence and carelessness, which in turn causes inaccurate tobacco structure detection and affects the quality of the produced tobacco, the present invention, by setting the upload monitoring module 2, can use the monitoring camera 8 to capture the operating process of the staff in real time, so that the staff can be more cautious, improve the standardization of the staff's operation, avoid the irregular operation of the staff due to negligence and carelessness, and improve the quality of the produced tobacco.

[0028] Among them, the image recognition module 3 includes a transmission device 9, an alarm device 10 and a processor. The image recognition module 3 identifies the irregular operations of the staff and takes screenshots to issue an alarm based on the operation video fed back by the uploaded monitoring module 2, and generates statistics based on the data transmitted by the weighing and sampling module 1 to form a report.

[0029] Furthermore, the weighing device 6 samples and weighs the tobacco at a fixed point, at a fixed time and in a fixed quantity. This not only complies with national standards, but also facilitates the staff to make timely adjustments when the sampling operation does not comply with the regulations, thereby reducing losses. The specific regulations are as follows.

[0030] The fixed point regulations for sampling and weighing are as follows: sampling and weighing of the tobacco at the outlet of the roasting machine 4 is carried out at a fixed point.

[0031] The timing of sampling and weighing is as follows: the staff starts timing at the fixed punch card device 5, and the sampling interval is 1h or 2h. Of course, other time limits are also considered in this application, such as 30min, 40min, 50min, 1.5h, etc.

[0032] The quantitative regulation of sampling and weighing is: the standard sampling weight of tobacco leaves is 3000g, and the error does not exceed 300g. Other standard sampling weights, such as 2000g, are also considered in this application.

[0033] Furthermore, a weighing alarm light is installed on the side of the weighing device 6. The upper and lower weighing limits can be set according to the sampling requirements. When the weighing weight exceeds the range, the weighing alarm light flashes and an alarm sound is heard.

[0034] Furthermore, the weighing device 6 takes a sample of the tobacco at one time during weighing and sampling, so as to ensure that the weighing of the tobacco meets the requirements of national standards.

[0035] In a specific example, the weighing and quotation codes of the weighing device 6 are as follows.

[0036] Assuming we have a variable measured_weight to represent the weight of the tobacco measured (in grams), the code could look like this: Python measured_weight=3000#Assume the measured weight is 3000g allowed_error=300 #The maximum allowed error is 300g ifmeasured_weight>3000+allowed_error or measured_weight<3000-allowed_error: print("The alarm device is triggered! The tobacco weight is abnormal.") #Here you can add specific code to trigger the alarm or call the alarm function else: print("Tobacco weight is normal. No alarm was triggered.") Explanation: measured_weight is the weight of the tobacco measured by the weighing device, which is assumed to be 3000g here.

[0037] Allowed_error is the maximum allowable error, that is, the floating range can be plus or minus 300g based on 3000g.

[0038] The if statement checks whether the measured weight is outside the allowable error range. If it is, the print("Trigger alarm! Abnormal tobacco weight.") line of code will be executed, and the specific code or function that triggers the alarm can be added here.

[0039] If the measured weight is within the allowable error range, print("Tobacco weight is normal. No alarm is triggered.") will be output.

[0040] In a specific example, the fixed card punching device 5 adopts a fingerprint pressing type, a radio frequency card swiping type or a facial recognition card swiping type.

[0041] The working time of the staff, including the start and end time, etc., is recorded by the fixed clock-in device 5 .

[0042] The transmission device 9 transmits the data including weighing data, weighing time, staff, and category parameters recorded by the weighing device 6 to the processor, and forms a report through statistics.

[0043] Among them, the alarm device 10 identifies the operating steps and wear of the staff according to the video feedback from the uploaded monitoring module 2, and warns the staff and management personnel through warning lights or buzzer alarms when the staff operate and wear irregularly.

[0044] The staff's operations include separating the sampled tobacco leaves into leaves of different sizes through a vibrating sorting screen, thereby guiding the operators of the previous process to accurately control the structure of the re-roasted leaves, reduce the production of unqualified products, and reduce material loss.

[0045] Furthermore, the processor of the image recognition module 3 uses a deep learning model to perform graphic detection, and sets a detection threshold, a frame counter, and a continuous operation threshold for the model.

[0046] Among them, the detection threshold is the single-frame detection threshold of the graphic, and the continuous operation threshold is the time or frame number threshold of the staff's continuous irregular operations.

[0047] In the specific example, the code for standardizing and warning the staff's operating procedures is as follows.

[0048] Use Python and OpenCV to detect manipulation and wear status and trigger an alarm when needed: importcv2 importnumpyasnp #Initialize the camera or load the video file #Here it is assumed that the video source is a local file, which can be modified according to the actual situation cap=cv2.VideoCapture('your_video_file.mp4') # Load the operation steps and the detection model for improper wearing #Here we assume that a pre-trained model or a custom model is used for detection. The specific model can be selected according to the needs. #For example, using deep learning models such as YOLO, SSD, etc. defdetect_operations(frame): #Here is the logic for detecting operation steps and wearing, which can be filled in according to actual conditions #Return True if normal, return False if abnormal #In practical applications, this needs to be implemented according to specific models and algorithms #This is simplified to randomly return True or False returnnp.random.choice([True,False]) #Set the alarm threshold THRESHOLD_COUNT=30#The alarm is triggered only when irregular operations are detected for 30 consecutive frames # Initialize counter and alarm status frame_count=0 alarm_triggered=False whileTrue: ret,frame=cap.read() ifnotret: break #Check whether the operation steps and wearing are standard in each frame is_normal=detect_operations(frame) ifnotis_normal: frame_count+=1 else: frame_count=0 #If continuous irregular operations exceeding the threshold are detected, an alarm is triggered ifframe_count>=THRESHOLD_COUNTandnotalarm_triggered: #Here you can add the code that triggers the alarm, such as sending an alarm signal to the management staff print("ALERT:Detectedimproperoperationsorattire!") alarm_triggered=True cv2.imshow('MonitoringVideo',frame) ifcv2.waitKey(1)&0xFF==ord('q'): break #Release resources cap.release() cv2.destroyAllWindows() Explanation: Video reading and initialization: Use OpenCV to initialize the video stream or load a local video file.

[0049] detect_operations function: This is a simplified function that simulates the logic of detecting operation steps and wearing status. In practical applications, you can use pre-trained deep learning models (such as YOLO, SSD, etc.) or custom models for actual detection.

[0050] THRESHOLD_COUNT and frame_count: Set a threshold and frame counter to trigger an alarm when the number of consecutive irregular operations is detected.

[0051] Alarm trigger logic: When continuous frames detect irregular operations exceeding the threshold, an alarm is triggered. In actual applications, specific alarm mechanisms can be added according to needs, such as sending alarm information to management personnel or triggering alarms through physical devices.

[0052] Display Video and Exit: Use OpenCV to display the monitored video stream and exit the loop and release resources by pressing the 'q' key.

[0053] Monitoring methods A monitoring method according to the aforementioned tobacco production line sampling monitoring system, the monitoring method comprising the following steps (steps S1 and S2 in no particular order).

[0054] S1. Tobacco weighing: The tobacco at the exit of the roasting machine is weighed by a weighing device 6.

[0055] The weighed samples are separated into blades of different sizes by a blade vibration sorting screen, so as to guide the operator to accurately control the structure of the re-roasted blades, reduce the production of unqualified products, reduce material loss, and record the working time of the staff through a fixed punching device 5.

[0056] S2, operation monitoring: the monitoring camera 8 fixedly installed on the weighing device 6 captures the worker's clothing and operation steps and transmits them to the image recognition module 3.

[0057] S3, image recognition and report generation: The processor processes and compiles statistics on the recorded data including weighing data, weighing time, staff, and category parameters to form a report. According to the video feedback from the uploaded monitoring module 2, the operating steps and wear of the staff are identified, and when the staff's operation and wear are not standardized, the alarm device 10 is used to warn the staff and management personnel.

[0058] The functional principle of the present invention can be explained through the following operation mode.

[0059] The tobacco at the exit of the roasting machine 4 can be weighed. The weighing device 6 works at intervals of 1h or 2h. During operation, the tobacco is weighed regularly by the staff punching in. When the weighing device 6 weighs the tobacco quantitatively, the weight of the tobacco is 3000g, and the error does not exceed 300g. The fixed-point, regular, and quantitative weighing not only meets the national standards, but also improves the accuracy of tobacco structure detection. In a specific example, a weighing record is shown in the table below.

[0060] Table 1: Sampling and weighing record The monitoring camera 8 fixedly installed on the weighing device 6 captures the worker's clothing and operating steps and transmits them to the image recognition module 3, so that the worker can be more cautious, improve the worker's operating standardization, and avoid the worker's improper operation due to negligence and carelessness.

[0061] The weighing data, weighing time, staff, category, etc. transmitted by the weighing sampling module 1 are counted and reported through the transmission device 9. According to the video feedback from the uploaded monitoring module 2, the operation steps and wear of the staff are identified and screenshots are taken. When the operation and wear of the staff are not standardized, the staff and management personnel are warned through warning lights or buzzer alarms.

[0062] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A tobacco production line sampling monitoring system, characterized in that: The tobacco production line sampling monitoring system comprises a weighing sampling module (1) and an uploading monitoring module (2) which are connected to an image recognition module (3) by telecommunication; The weighing and sampling module (1) comprises a tobacco leaf sampling box (4), a fixed punching device (5) and a weighing device (6); The upload monitoring module (2) comprises a fixing device (7) and a monitoring camera (8); the monitoring camera (8) is installed on the weighing device (6) via the fixing device (7) and is used to monitor the operating steps of the staff and transmit the operating process of the staff to the image recognition module (3) in the form of a video, thereby realizing the monitoring of the operating steps of the staff; The image recognition module (3) comprises a transmission device (9), an alarm device (10) and a processor. The image recognition module (3) identifies irregular operations of staff members based on the operation video fed back by the upload monitoring module (2), takes screenshots and issues alarms, and generates statistics based on the data transmitted by the weighing and sampling module (1) to form a report.

2. The tobacco production line sampling monitoring system according to claim 1, characterized in that: The weighing device (6) samples and weighs tobacco at a fixed point, at a fixed time, and in a fixed quantity.

3. The tobacco production line sampling monitoring system according to claim 2, characterized in that: The fixed-point regulations for sampling and weighing are: sampling and weighing of tobacco at a fixed point at the exit of the roasting machine (4); the timing regulations for sampling and weighing are: the staff turns on the timing at the fixed punching device (5) and the sampling interval is 1 hour or 2 hours; the quantitative regulations for sampling and weighing are: the standard sampling weight of tobacco leaves is 3000g, and the error does not exceed 300g.

4. The tobacco production line sampling monitoring system according to claim 1, characterized in that: A weighing alarm light is installed on the side of the weighing device (6). An upper weighing limit and a lower weighing limit can be set according to sampling requirements. When the weighing weight exceeds the range, the weighing alarm light flashes and an alarm sound is heard.

5. The tobacco production line sampling monitoring system according to claim 1, characterized in that: The weighing device (6) takes a single sample of the tobacco during weighing and sampling.

6. The tobacco production line sampling monitoring system according to claim 1, characterized in that: The fixed card punching device (5) adopts a fingerprint pressing type, a radio frequency card swiping type or a facial recognition card swiping type.

7. The tobacco production line sampling monitoring system according to claim 1, characterized in that: The transmission device (9) transmits the data recorded by the weighing device (6), including weighing data, weighing time, staff, and category parameters, to the processor, and generates a report through statistics.

8. The tobacco production line sampling monitoring system according to claim 1, characterized in that: The alarm device (10) identifies the operating steps and attire of the staff member based on the video fed back by the uploaded monitoring module (2), and warns the staff member and management personnel through a warning light or a buzzer alarm when the staff member's operation and attire are not in compliance with regulations.

9. The tobacco production line sampling monitoring system according to claim 1, characterized in that: The processor of the image recognition module (3) uses a deep learning model to perform image detection and sets a detection threshold, a frame counter and a continuous operation threshold for the model.

10. A monitoring method for a tobacco production line sampling monitoring system according to any one of claims 1 to 9, characterized in that: The monitoring method includes the following steps: S1. Tobacco weighing: weighing the tobacco at the exit of the roasting machine by means of a weighing device (6); S2, operation monitoring: using a monitoring camera (8) fixedly installed on the weighing device (6) to capture the worker's clothing and operation steps and transmit the images to the image recognition module (3); S3, image recognition and report generation: the processor processes the recorded data including weighing data, weighing time, staff, and category parameters and generates statistics to generate reports, and identifies the staff's operation steps and clothing for identification. When the staff's operation and clothing are not in accordance with the regulations, the alarm device (10) is used to warn the staff and management personnel.

Images