A method for detecting the maximum length of a cigarette burning cone

Abstract

The application provides a cigarette combustion cone maximum length detection method, and belongs to the technical field of cigarette detection. The cigarette combustion cone maximum length detection method adjusts the cigarette measurement environment; fixes the cigarette on a smoking machine, installs a high-definition camera in the position directly in front of the smoking machine, records a video by using the high-definition camera during the process of the smoking machine sucking the cigarette; establishes a cigarette combustion cone length inversion model, clearly processes the video data, analyzes the video data, obtains the distance between the cigarette combustion line and the combustion cone end during the process of sucking the cigarette, compares the distance, and obtains the maximum distance between the cigarette combustion line and the combustion cone end; repeats the above operation under the same environment; and the cigarette combustion cone maximum length detection method can solve the problem that the traditional measurement of the cigarette combustion cone length is a dynamic process, which causes the measurement result accuracy to be insufficient.

Description

Technical Field

[0001] This invention belongs to the field of cigarette detection technology, and more specifically, relates to a method for detecting the maximum length of the cigarette combustion cone. Background Technology

[0002] The maximum length of the cigarette combustion cone refers to the maximum distance between the cigarette combustion line and the end of the combustion cone. It is one of the important indicators of smoke emissions. Since cigarette smoke contains a variety of harmful substances, such as carbon monoxide, benzene, and polycyclic aromatic hydrocarbons, these substances are harmful to the environment and human health. The maximum length of the cigarette combustion cone is one of the important reference parameters for assessing the impact of cigarette smoke emissions on environmental quality.

[0003] The combustion line refers to the portion of the cigarette paper that burns during smoking and static burning. The end of the combustion cone is the furthest point of the combustion cone corresponding to the combustion line. The length of the combustion cone is a major factor affecting the burning of cigarette butts, and the problem of cigarette butts falling off has been a long-standing challenge for cigarette manufacturers, seriously impacting the consumer experience. Furthermore, the establishment of a method for measuring the maximum length of the cigarette combustion cone is also intended to strengthen and strictly regulate environmental protection and hygiene standards in the tobacco industry.

[0004] Currently, most methods for measuring the length of the combustion cone use traditional methods, which can lead to insufficient accuracy in the measurement results because cigarette combustion is a dynamic process. Summary of the Invention

[0005] In view of this, the present invention provides a method for detecting the maximum length of a cigarette combustion cone, which can solve the problem that the traditional method of measuring the length of a cigarette combustion cone is a dynamic process, which leads to insufficient accuracy of the measurement results.

[0006] This invention is implemented as follows:

[0007] This invention provides a method for detecting the maximum length of a cigarette combustion cone, comprising the following steps:

[0008] S10: Adjust the cigarette measurement environment;

[0009] S20: Fix the cigarette to the smoking machine, install a high-definition camera at the front of the smoking machine, and record video during the process of smoking the cigarette using the high-definition camera to obtain video data;

[0010] S30: Establish a cigarette combustion cone length inversion model, analyze video data, and obtain the maximum distance between the cigarette combustion line and the end of the combustion cone;

[0011] S40: Repeat the operations of S10-S30 under the same conditions.

[0012] Light the cigarette, start the smoking machine to inhale the cigarette, and press the shutter of the high-definition camera before taking the second puff to record the process of inhaling the cigarette using the smoking machine.

[0013] The phrase "repeating the above operations under the same conditions" refers to the same batch of cigarette samples being tested in the same laboratory, using the same testing instrument, and by the same operator within a short period of time.

[0014] The high-definition camera used is one with a resolution of 4K or higher.

[0015] Based on the above technical solution, the method for detecting the maximum length of a cigarette combustion cone according to the present invention can be further improved as follows:

[0016] The specific steps for establishing the cigarette combustion cone length inversion model include:

[0017] The first step is to analyze the video recording data;

[0018] The second step is to establish a three-dimensional correspondence model, which maps the images captured by the high-definition camera to the actual objects.

[0019] The third step is to use an electronic measuring ruler to measure the length of the cigarette in the image when the cigarette is being smoked, and input the length of the cigarette in the image into the corresponding three-dimensional model to obtain the actual length of the cigarette.

[0020] The fourth step involves collecting cigarette combustion data for a specified time period and the corresponding cigarette combustion cone length. The deep network training samples are then divided into a training set, a validation set, and a test set in a 6:2:2 ratio. The cigarette combustion data includes the cigarette length, temperature, wind direction, and wind speed within the image.

[0021] The fifth step is to use a convolutional neural network to establish a preliminary model for retrieving the length of the cigarette combustion cone;

[0022] The sixth step involves using a training set to train the prototype of the cigarette combustion cone length inversion model through a convolutional neural network to obtain the cigarette combustion cone length inversion model.

[0023] The seventh step is to use validation and test sets to validate and test the cigarette combustion cone length inversion model and optimize the cigarette combustion cone length inversion model.

[0024] Furthermore, the single-puff suction capacity of the smoking machine is 51ml, the suction time is 1.6s, and the interval between two puffs is 17s.

[0025] Furthermore, the specific steps for analyzing the video data to obtain the maximum distance between the cigarette's combustion line and the end of the combustion cone include:

[0026] The first step is to perform frame processing on the video data to obtain the target frame image corresponding to the time when smoking a cigarette;

[0027] The second step is to perform a sharpening process on the target frame image;

[0028] The third step is to use the electronic ruler to obtain the length of the cigarette in the image, and then input it into the cigarette combustion cone length inversion model to obtain the length of the cigarette combustion cone within a specified time.

[0029] The fourth step is to process the length of the combustion cone to obtain the distance between the cigarette combustion line and the end of the combustion cone;

[0030] The fifth step is to obtain the maximum value of the distance between the cigarette burning line and the end of the combustion cone when each cigarette is inhaled by comparing the distance between the cigarette burning line and the end of the combustion cone.

[0031] The distance between the cigarette's combustion line and the end of the combustion cone during the first puff is denoted as S1, the distance during the second puff is denoted as S2, the distance during the third puff is denoted as S3, ... and the distance during the nth puff is denoted as Sn.

[0032] Furthermore, the specific steps for performing frame processing on the video data to obtain the image corresponding to the moment the cigarette was smoked include:

[0033] The first step is to divide the recorded data into a video frame sequence, and then determine the image frame to be processed and the associated image frames preceding the image frame to be processed in the video frame sequence.

[0034] The second step is to add an object frame based on the icon object in the image frame to be processed, and adjust the object frame to obtain an initial image frame containing the target object frame.

[0035] The third step is to identify the initial key points of the target object in the initial image frame based on the target object bounding box, and to determine the associated key points of the target object in the associated image frame.

[0036] The fourth step is to correct the initial key points in the initial image frame based on the associated key points to obtain a target image frame containing the target key points.

[0037] Furthermore, the specific steps for sharpening the target frame image include:

[0038] The first step is to select A data blocks centered on the data block where the noise reduction point is located in the target frame image. Each data block includes B×B pixel units. The data block containing the noise reduction point is the current data block, and the remaining A-1 data blocks are the surrounding data blocks. A and B are both integers, and 1 < B; 9 ≤ A ≤ 121.

[0039] The second step is to obtain the texture intensity value of the noise reduction point;

[0040] The third step is to determine the noise reduction parameter h of the noise reduction point based on the texture intensity value; where h∈[Hlow,Hhigh], Hlow and Hhigh represent the minimum and maximum values ​​of the set noise reduction parameter h, respectively;

[0041] The fourth step is to determine the denoising weights W(k) of the A-1 surrounding data blocks based on the denoising parameter h; where k represents the k-th surrounding data block; and the denoising weight of the k-th surrounding data block is...

[0042] ;

[0043] in,

[0044] ; ;

[0045] b represents the similarity distance between the k-th surrounding data block and the current data block, and Hlow and Hhigh represent the minimum and maximum values ​​of the set noise reduction parameter h, respectively;

[0046] The fifth step is to calculate and output the pixel values ​​of the noise reduction points after noise reduction based on the noise reduction weights.

[0047] Step 6: Select noise reduction points sequentially by row and column, repeat the above process, and perform noise reduction on all pixel units in the target frame image, and output the noise-reduced image.

[0048] The specific steps for obtaining the texture intensity value of the noise reduction point include:

[0049] The first step is to calculate the similarity distance between the current data block and A-1 surrounding data blocks;

[0050] The second step is to obtain the texture intensity value of the noise reduction point based on the maximum similarity distance.

[0051] Furthermore, the specific steps for establishing a three-dimensional correspondence model and mapping the images captured by the high-definition camera to the actual objects include:

[0052] The first step is to fix the high-definition camera in a fixed position directly in front of the smoking machine;

[0053] The second step is to fix cigarettes of different lengths on the smoking machine and take pictures of the cigarettes using the high-definition camera.

[0054] The third step is to use the electronic measuring ruler to measure the length of the cigarette captured by the high-definition camera, and use the vernier caliper to measure the actual length of the cigarette. The length of the cigarette captured by the high-definition camera is compared with the actual length of the cigarette to calculate the internal and external parameters of the high-definition camera.

[0055] The fourth step is to use the intrinsic and extrinsic parameters of the high-definition camera to perform global binding and adjustment of the high-definition camera, and establish the three-dimensional corresponding model.

[0056] Furthermore, the specific steps for globally binding and adjusting the high-definition camera using its intrinsic and extrinsic parameters include:

[0057] The first step is to fix the internal parameters of the high-definition camera and the actual length of the cigarette, and then adjust the external parameters of the cigarette image captured by the high-definition camera.

[0058] The second step is to fix the external parameters of the cigarette image captured by the high-definition camera and the actual length of the cigarette, and then adjust the internal parameters of the high-definition camera.

[0059] The third step is to fix the actual length of the cigarette and adjust the internal and external parameters of the high-definition camera.

[0060] Furthermore, the specific steps for adjusting the cigarette measurement environment include:

[0061] The environment for measuring the cigarette combustion cone is adjusted to meet the cigarette detection conditions. The cigarette combustion cone is placed on a black background board, and a light is used to illuminate the cigarette combustion cone. The brightness and intensity of the light are adjusted to ensure that the image captured by the high-definition camera has the best clarity.

[0062] The cigarette testing conditions are a constant temperature and humidity environment with a temperature of 20-25℃ and a relative humidity of 40%-70%.

[0063] Furthermore, the smoking machine is a single-channel smoking machine, which is used to hold the cigarette and ignite and inhale the cigarette, with the inhalation parameters being autonomously adjustable.

[0064] Compared with existing technologies, the beneficial effects of the method for detecting the maximum length of a cigarette combustion cone provided by this invention are as follows: By adjusting the cigarette measurement environment, the accuracy of the cigarette measurement data is ensured; by fixing the cigarette on a smoking machine and installing a high-definition camera at the front of the smoking machine, the high-definition camera records the cigarette combustion process during smoking, providing a dynamic picture for analysis; by establishing a cigarette combustion cone length inversion model and analyzing the recorded data, the maximum distance between the cigarette combustion line and the end of the combustion cone is obtained; by repeating the above operations under the same environment, the accuracy of the data is ensured, thus solving the problem that traditional methods for measuring the length of the cigarette combustion cone, which involve a dynamic process, lead to insufficient accuracy in the measurement results. Attached Figure Description

[0065] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0066] Figure 1 This is a flowchart illustrating a specific method for detecting the maximum length of a cigarette combustion cone. Detailed Implementation

[0067] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0068] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0069] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0070] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0071] like Figure 1 The diagram shows a flowchart of a method for detecting the maximum length of a cigarette combustion cone provided by the present invention, wherein the method comprises the following steps:

[0072] S10: Adjust the cigarette measurement environment;

[0073] S20: Fix the cigarette to the smoking machine, install a high-definition camera in front of the smoking machine, record video while smoking the cigarette using the high-definition camera, and obtain the video data;

[0074] S30: Establish a cigarette combustion cone length inversion model, analyze video data, and obtain the maximum distance between the cigarette combustion line and the end of the combustion cone;

[0075] S40: The above operations of S10-S30 under the same conditions.

[0076] Light a cigarette, start the smoking machine, and press the shutter of the high-definition camera before taking the second puff to record the process of smoking the cigarette using the smoking machine.

[0077] Repeating the above operations under the same conditions refers to the same laboratory, the same testing instrument, and the same operator repeating the above process to test the same batch of cigarette samples within a short period of time.

[0078] The high-definition camera used is one with a resolution of 4K or higher.

[0079] The specific steps for establishing the cigarette combustion cone length inversion model in the above technical solution include:

[0080] The first step is to analyze the video recording data;

[0081] The second step is to establish a three-dimensional correspondence model, which maps the images captured by the high-definition camera to the actual objects.

[0082] The third step is to use an electronic measuring ruler to measure the length of the cigarette in the image when smoking, and input the length of the cigarette in the image into the corresponding three-dimensional model to obtain the actual length of the cigarette.

[0083] The fourth step involves collecting cigarette combustion data for a specified time period and the corresponding cigarette combustion cone length. The deep network training samples are then divided into a training set, a validation set, and a test set in a 6:2:2 ratio. The cigarette combustion data includes the cigarette length, temperature, wind direction, and wind speed within the image.

[0084] The fifth step is to use a convolutional neural network to establish a preliminary model for retrieving the length of the cigarette combustion cone;

[0085] The sixth step involves using a training set to train the prototype of the cigarette combustion cone length inversion model through a convolutional neural network to obtain the cigarette combustion cone length inversion model.

[0086] The seventh step is to use validation and test sets to validate and test the cigarette combustion cone length inversion model and optimize the cigarette combustion cone length inversion model.

[0087] Furthermore, in the above technical solution, the single-puff suction capacity of the smoking machine is 51ml, the suction time is 1.6s, and the interval between two puffs is 17s.

[0088] Furthermore, in the above technical solution, the specific steps for analyzing the video recording data to obtain the maximum distance between the cigarette combustion line and the end of the combustion cone include:

[0089] The first step is to process the video data into frames to obtain the target frame image corresponding to the time when the cigarette is smoked.

[0090] The second step is to sharpen the target frame image;

[0091] The third step is to use an electronic ruler to obtain the length of the cigarette in the image, and then input it into the cigarette combustion cone length inversion model to obtain the length of the cigarette combustion cone within a specified time.

[0092] The fourth step is to process the length of the combustion cone to obtain the distance between the cigarette combustion line and the end of the combustion cone;

[0093] The fifth step is to obtain the maximum value of the distance between the cigarette burning line and the end of the combustion cone when each cigarette is inhaled by comparing the distance between the cigarette burning line and the end of the combustion cone.

[0094] The distance between the cigarette's combustion line and the end of the combustion cone during the first puff is denoted as S1, the distance during the second puff is denoted as S2, the distance during the third puff is denoted as S3, ... and the distance during the nth puff is denoted as Sn.

[0095] Furthermore, in the above technical solution, the specific steps for performing frame processing on the video recording data to obtain the image corresponding to the moment the cigarette was smoked include:

[0096] The first step is to divide the video data into a video frame sequence, and then determine the image frame to be processed and the associated image frames preceding the image frame to be processed in the video frame sequence.

[0097] The second step is to add an object frame based on the icon object in the image frame to be processed, and adjust the object frame to obtain an initial image frame containing the target object frame.

[0098] The third step is to identify the initial key points of the target object in the initial image frame based on the target object bounding box, and to determine the associated key points of the target object in the associated image frame.

[0099] The fourth step is to correct the initial key points in the initial image frame based on the associated key points to obtain the target image frame containing the target key points.

[0100] Furthermore, in the above technical solution, the specific steps for sharpening the target frame image include:

[0101] The first step is to select A data blocks in the target frame image, centered on the data block where the noise reduction point is located. Each data block consists of B×B pixel units. The data block containing the noise reduction point is the current data block, and the remaining A-1 data blocks are the surrounding data blocks. A and B are both integers, and 1 < B; 9 ≤ A ≤ 121.

[0102] The second step is to obtain the texture intensity value of the noise reduction point;

[0103] The third step is to determine the noise reduction parameter h of the noise reduction point based on the texture intensity value; where h∈[Hlow,Hhigh], Hlow and Hhigh represent the minimum and maximum values ​​of the set noise reduction parameter h, respectively;

[0104] The fourth step is to determine the denoising weights W(k) of the A-1 surrounding data blocks based on the denoising parameter h; where k represents the k-th surrounding data block; and the denoising weight of the k-th surrounding data block is...

[0105] ;

[0106] in,

[0107] ; ;

[0108] b represents the similarity distance between the k-th surrounding data block and the current data block, and Hlow and Hhigh represent the minimum and maximum values ​​of the set noise reduction parameter h, respectively;

[0109] The fifth step is to calculate and output the pixel values ​​of the noise reduction points based on the noise reduction weights.

[0110] The sixth step involves selecting noise reduction points row by row and column by column, repeating the above process to denoise all pixel units in the target frame image, and then outputting the denoised image.

[0111] The specific steps for obtaining the texture intensity value of the noise reduction point include:

[0112] The first step is to calculate the similarity distance between the current data block and A-1 surrounding data blocks;

[0113] The second step is to obtain the texture intensity value of the noise reduction point based on the maximum similarity distance.

[0114] Furthermore, in the above technical solution, the specific steps for establishing a three-dimensional correspondence model and mapping the images captured by the high-definition camera to the actual objects include:

[0115] The first step is to fix the high-definition camera in a fixed position directly in front of the smoking machine;

[0116] The second step is to fix cigarettes of different lengths on the smoking machine and take pictures of the cigarettes using a high-definition camera.

[0117] The third step is to use an electronic measuring ruler to measure the length of the cigarette captured by the high-definition camera, and use a vernier caliper to measure the actual length of the cigarette. The length of the cigarette captured by the high-definition camera is then compared with the actual length of the cigarette to calculate the internal and external parameters of the high-definition camera.

[0118] The fourth step is to use the internal and external parameters of the high-definition camera to perform global binding and adjustment of the high-definition camera and establish a three-dimensional corresponding model.

[0119] Furthermore, in the above technical solution, the specific steps for globally binding and adjusting the high-definition camera using its internal and external parameters include:

[0120] The first step is to fix the internal parameters of the high-definition camera and the actual length of the cigarette, and then adjust the external parameters of the cigarette image captured by the high-definition camera.

[0121] The second step is to fix the external parameters of the cigarette image captured by the high-definition camera and the actual length of the cigarette, and then adjust the internal parameters of the high-definition camera.

[0122] The third step is to fix the actual length of the cigarette and adjust the internal and external parameters of the high-definition camera.

[0123] Furthermore, in the above technical solution, the specific steps for adjusting the cigarette measurement environment include:

[0124] The environment for measuring the cigarette combustion cone is adjusted to meet the cigarette detection conditions. The cigarette combustion cone is placed on a black background board, and a light is used to illuminate the cigarette combustion cone. The brightness and intensity of the light are adjusted to ensure that the image captured by the high-definition camera has the best clarity.

[0125] The cigarette testing conditions are a constant temperature and humidity environment of 20-25℃ and 40%-70% relative humidity.

[0126] Furthermore, in the above technical solution, the smoking machine is a single-channel smoking machine, which is used to hold cigarettes and ignite and smoke them, with the smoking parameters being autonomously adjustable.

[0127] Specifically, the principle of this invention is as follows: adjust the cigarette measurement environment; fix the cigarette on the smoking machine, install a high-definition camera in front of the smoking machine, and record video during the smoking process; establish a cigarette combustion cone length inversion model, analyze the video data, and obtain the maximum distance between the cigarette combustion line and the end of the combustion cone; repeat the above operation under the same environment.

[0128] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for detecting the maximum length of a cigarette combustion cone, characterized in that, It has the following steps: S10: Adjust the cigarette measurement environment; S20: Fix the cigarette to the smoking machine, install a high-definition camera at the front of the smoking machine, and record video during the process of smoking the cigarette using the high-definition camera to obtain video data; S30: Establish a cigarette combustion cone length inversion model, analyze the video data, and obtain the maximum distance between the cigarette combustion line and the end of the combustion cone; S40: Repeat the operations of S10-S30 above under the same conditions; The specific steps for establishing the cigarette combustion cone length inversion model include: The first step is to analyze the recorded data; The second step is to establish a three-dimensional correspondence model, which maps the images captured by the high-definition camera to the actual objects. The third step is to use an electronic measuring ruler to measure the length of the cigarette in the image when the cigarette is being smoked, and input the length of the cigarette in the image into the corresponding three-dimensional model to obtain the actual length of the cigarette. The fourth step is to establish training samples by collecting cigarette combustion data for a specified time period and the corresponding cigarette combustion cone length. The training samples are then divided into a training set, a validation set, and a test set in a 6:2:2 ratio. The cigarette combustion data includes the cigarette length, temperature, wind direction, and wind speed within the image. The fifth step is to use a convolutional neural network to establish a preliminary model for retrieving the length of the cigarette combustion cone; The sixth step involves using the training set to train the prototype of the cigarette combustion cone length inversion model through a convolutional neural network to obtain the cigarette combustion cone length inversion model. Step 7: Use the validation set and the test set to validate and test the cigarette combustion cone length inversion model, and optimize the cigarette combustion cone length inversion model; The specific steps for analyzing the video data to obtain the maximum distance between the cigarette combustion line and the end of the combustion cone include: The first step is to perform frame segmentation on the video data to obtain the target frame image corresponding to the time when smoking a cigarette; The second step is to perform a sharpening process on the target frame image; The third step is to use the electronic ruler to obtain the length of the cigarette in the target frame image, and then input the length of the cigarette in the target frame image into the cigarette combustion cone length inversion model to obtain the length of the cigarette combustion cone within a specified time. The fourth step is to subtract the distance from the cigarette burning line to the end of the cigarette near the butt from the length of the combustion cone to obtain the distance between the cigarette burning line and the end of the combustion cone. The fifth step is to obtain the maximum value of the distance between the cigarette burning line and the end of the combustion cone when each cigarette is inhaled by comparing the distance between the cigarette burning line and the end of the combustion cone.

2. The method for detecting the maximum length of a cigarette combustion cone according to claim 1, characterized in that, The smoking machine has a single-puff suction capacity of 48-52ml, a suction time of 1.5-1.7s, and an interval of 16-18s between two puffs.

3. The method for detecting the maximum length of a cigarette combustion cone according to claim 2, characterized in that, The specific steps for performing frame processing on the video data to obtain the image corresponding to the moment of smoking a cigarette include: The first step is to divide the recorded data into a video frame sequence, and then determine the image frame to be processed and the associated image frames preceding the image frame to be processed in the video frame sequence. The second step is to add an object frame based on the icon object in the image frame to be processed, and adjust the object frame to obtain an initial image frame containing the target object frame. The third step is to identify the initial key points of the target object in the initial image frame based on the target object bounding box, and to determine the associated key points of the target object in the associated image frame. The fourth step is to correct the initial key points in the initial image frame based on the associated key points to obtain a target image frame containing the target key points.

4. The method for detecting the maximum length of a cigarette combustion cone according to claim 3, characterized in that, The specific steps for sharpening the target frame image include: The first step is to select A data blocks centered on the data block where the noise reduction point is located in the target frame image. Each data block includes B×B pixel units. The data block containing the noise reduction point is the current data block, and the remaining A-1 data blocks are the surrounding data blocks. A and B are both integers, and B > 1 ≤ A ≤ 121. The second step is to obtain the texture intensity value of the noise reduction point; The third step is to determine the noise reduction parameter h of the noise reduction point based on the texture intensity value; where h∈[Hlow,Hhigh], Hlow and Hhigh represent the minimum and maximum values ​​of the set noise reduction parameter h, respectively; The fourth step is to determine the denoising weights W(k) of the A-1 surrounding data blocks based on the denoising parameter h; where k represents the k-th surrounding data block; and the denoising weight of the k-th surrounding data block is... ； in, ； ； b represents the similarity distance between the k-th surrounding data block and the current data block, and Hlow and Hhigh represent the minimum and maximum values ​​of the set noise reduction parameter h, respectively; The fifth step is to calculate and output the pixel values ​​of the noise reduction points after noise reduction based on the noise reduction weights. Step 6: Select noise reduction points sequentially by row and column, repeat the above process, and perform noise reduction on all pixel units in the target frame image, and output the noise-reduced image.

5. The method for detecting the maximum length of a cigarette combustion cone according to claim 4, characterized in that, The specific steps for establishing a three-dimensional correspondence model and matching the images captured by the high-definition camera with the actual objects include: The first step is to fix the high-definition camera in a fixed position directly in front of the smoking machine; The second step is to fix cigarettes of different lengths on the smoking machine and take pictures of the cigarettes using the high-definition camera. The third step is to use the electronic measuring ruler to measure the length of the cigarette captured by the high-definition camera, and use the vernier caliper to measure the actual length of the cigarette. The length of the cigarette captured by the high-definition camera is compared with the actual length of the cigarette to calculate the internal and external parameters of the high-definition camera. The fourth step is to use the intrinsic and extrinsic parameters of the high-definition camera to perform global binding and adjustment of the high-definition camera, and establish the three-dimensional corresponding model.

6. The method for detecting the maximum length of a cigarette combustion cone according to claim 5, characterized in that, The specific steps for globally binding and adjusting the high-definition camera using its internal and external parameters include: The first step is to fix the internal parameters of the high-definition camera and the actual length of the cigarette, and then adjust the external parameters of the cigarette image captured by the high-definition camera. The second step is to fix the external parameters of the cigarette image captured by the high-definition camera and the actual length of the cigarette, and then adjust the internal parameters of the high-definition camera. The third step is to fix the actual length of the cigarette and adjust the internal and external parameters of the high-definition camera.

7. The method for detecting the maximum length of a cigarette combustion cone according to claim 6, characterized in that, The specific steps for adjusting the cigarette measurement environment include: The environment for measuring the cigarette combustion cone is adjusted to meet the cigarette detection conditions. The cigarette combustion cone is placed on a black background board, and a light is used to illuminate the cigarette combustion cone. The brightness and intensity of the light are adjusted to ensure that the image captured by the high-definition camera has the best clarity.

8. The method for detecting the maximum length of a cigarette combustion cone according to claim 7, characterized in that, The smoking machine is a single-channel smoking machine, which is used to hold the cigarette and ignite and smoke it, with the smoking parameters being automatically adjustable.

Images