CN-121970936-A - Electronic cigarette preparation method and system based on image analysis

Abstract

The invention provides an electronic cigarette preparation method and system based on image analysis, and relates to the technical field of electronic cigarette preparation, wherein the method comprises the steps of obtaining a plurality of continuous X-ray images shot by an atomizer at different tobacco tar injection speeds; generating an internal oiling video of the atomizer at each tobacco tar injection speed by using a variational self-encoder based on a plurality of continuous X-ray images shot by the atomizer at each tobacco tar injection speed, determining the similarity of the internal oiling information of the atomizer at each tobacco tar injection speed and the internal oiling information of the atomizer at different tobacco tar injection speeds by using an oiling information determining model based on the internal oiling video of the atomizer at each tobacco tar injection speed, determining a target oiling speed based on the internal oiling information of the atomizer at each tobacco tar injection speed and the similarity of the internal oiling information of the atomizer at different tobacco tar injection speeds, and oiling the electronic cigarette based on the target oiling speed.

Inventors

• YUAN HUI

Assignees

• 苑卉

Dates

Publication Date

20260505

Application Date

20260403

Priority Date

20251030

Claims (10)

1. 1. The electronic cigarette preparation method based on image analysis is characterized by comprising the following steps of: acquiring a plurality of continuous X-ray images shot by the atomizer at different tobacco tar injection speeds; generating an internal oiling video of the atomizer at each tobacco-oil injection speed by using a variational self-encoder based on a plurality of continuous X-ray images shot by the atomizer at each tobacco-oil injection speed; determining the similarity of the internal oiling information of the atomizer at each tobacco tar injection speed and the internal oiling information of the atomizer at different tobacco tar injection speeds by using an oiling information determination model based on the internal oiling video of the atomizer at each tobacco tar injection speed; Determining a target oiling speed based on the internal oiling information of the atomizer at each tobacco tar injection speed and the similarity of the internal oiling information of the atomizer at different tobacco tar injection speeds; And oiling the electronic cigarette based on the target oiling speed.
2. 2. The method for preparing an electronic cigarette based on image analysis according to claim 1, wherein the oiling information determination model is a long-short-term neural network model.
3. 3. The method for preparing an electronic cigarette based on image analysis according to claim 1, wherein determining the target injection speed based on the similarity of the internal injection information of the atomizer at each of the tobacco tar injection speeds and the internal injection information of the atomizer at the different tobacco tar injection speeds comprises: Constructing a knowledge graph, wherein the knowledge graph comprises a plurality of nodes and a plurality of edges between the nodes, each node represents a tobacco tar injection speed, the edges between the nodes are the similarity of internal oiling information of the atomizer at the tobacco tar injection speed, and the node characteristics of each node comprise the internal oiling information of the atomizer at the tobacco tar injection speed; and processing the knowledge graph based on the graph self-encoder to determine the target oiling speed.
4. 4. The method for preparing an electronic cigarette based on image analysis according to claim 1, wherein the internal oiling information of the atomizer at each tobacco tar injection speed comprises oiling uniformity, bubble formation information, oiling filling degree and oiling sealing degree.
5. 5. An electronic cigarette preparation system based on image analysis, comprising: the acquisition module is used for acquiring a plurality of continuous X-ray images shot by the atomizer at different tobacco tar injection speeds; the generation module is used for generating an internal oiling video of the atomizer at each tobacco tar injection speed by using a variation self-encoder on the basis of a plurality of continuous X-ray images shot by the atomizer at each tobacco tar injection speed; the information determining module is used for determining the similarity of the internal oiling information of the atomizer at each tobacco tar injection speed and the internal oiling information of the atomizer at different tobacco tar injection speeds by using an oiling information determining model based on the internal oiling video of the atomizer at each tobacco tar injection speed; the oiling speed determining module is used for determining a target oiling speed based on the internal oiling information of the atomizer at each tobacco tar injection speed and the similarity of the internal oiling information of the atomizer at different tobacco tar injection speeds; and the oiling module is used for oiling the electronic cigarette based on the target oiling speed.
6. 6. The electronic cigarette manufacturing system based on image analysis of claim 5, wherein the oiling information determination model is a long-short term neural network model.
7. 7. The electronic cigarette manufacturing system based on image analysis of claim 5, wherein the determining a target oil injection rate based on the similarity of the internal oil injection information of the atomizer at each of the tobacco tar injection rates and the internal oil injection information of the atomizer at the different tobacco tar injection rates comprises: Constructing a knowledge graph, wherein the knowledge graph comprises a plurality of nodes and a plurality of edges between the nodes, each node represents a tobacco tar injection speed, the edges between the nodes are the similarity of internal oiling information of the atomizer at the tobacco tar injection speed, and the node characteristics of each node comprise the internal oiling information of the atomizer at the tobacco tar injection speed; and processing the knowledge graph based on the graph self-encoder to determine the target oiling speed.
8. 8. The electronic cigarette manufacturing system based on image analysis of claim 5, wherein the internal oiling information of the atomizer at each tobacco tar injection speed comprises oiling uniformity, bubble formation information, oiling filling level, oiling sealing level.
9. 9. An electronic device comprising a processor, a memory, and a computer program, wherein the computer program is stored in the memory and configured to be executed by the processor to implement the image analysis-based e-cigarette manufacturing method of any one of claims 1 to 4.
10. 10. A computer-readable storage medium having stored thereon a computer program, which when executed by a processor, implements the method for preparing an electronic cigarette based on image analysis as claimed in any one of claims 1 to 4.

Description

Electronic cigarette preparation method and system based on image analysis Technical Field The invention relates to the technical field of electronic cigarette preparation, in particular to an electronic cigarette preparation method and system based on image analysis. Background In recent years, with the rise of health consciousness and the limitation of conventional tobacco products, electronic cigarettes are rapidly rising as a substitute. However, the manufacturing process of electronic cigarettes still faces many challenges, especially in the oiling step inside the atomizer. Traditional oiling methods rely mainly on experience or simple mechanical control, and it is difficult to accurately adjust the oiling speed. Rapid oiling may result in a non-uniform distribution of the tobacco tar inside the atomizer, causing some areas to be too wet and other areas to be dry. Such non-uniformity can affect the efficiency of the heating element and thus the atomization effect and mouthfeel. Air is easily introduced in the high-speed oiling process to form bubbles. These bubbles can affect the flowability of the tobacco tar and can cause intermittent problems during atomization, such as "dry burn" or poor taste during smoking. Too slow oiling speed can significantly reduce the overall efficiency of the production line, extending the production cycle of each product, and thus increasing manufacturing costs. How to accurately determine the oil filling speed of an atomizer is a current problem to be solved. Disclosure of Invention The invention mainly solves the technical problem of how to accurately determine the oiling speed of the atomizer. According to a first aspect, the invention provides an electronic cigarette preparation method based on image analysis, which comprises the steps of obtaining a plurality of continuous X-ray images shot on a atomizer at different tobacco tar injection speeds, generating an internal oiling video of the atomizer at each tobacco tar injection speed by using a variable self-encoder based on the plurality of continuous X-ray images shot on the atomizer at each tobacco tar injection speed, determining similarity of the internal oiling information of the atomizer at each tobacco tar injection speed and the internal oiling information of the atomizer at different tobacco tar injection speeds by using an oiling information determination model based on the internal oiling video of the atomizer at each tobacco tar injection speed, determining a target oiling speed based on the similarity of the internal oiling information of the atomizer at each tobacco tar injection speed and the internal oiling information of the atomizer at different tobacco tar injection speeds, and oiling electronic cigarettes based on the target speed. In one possible implementation, the oiling information determination model is a long-short term neural network model. In one possible implementation manner, the determining the target oil injection speed based on the similarity of the internal oil injection information of the atomizer at each of the tobacco tar injection speeds and the internal oil injection information of the atomizer at the different tobacco tar injection speeds includes: Constructing a knowledge graph, wherein the knowledge graph comprises a plurality of nodes and a plurality of edges between the nodes, each node represents a tobacco tar injection speed, the edges between the nodes are the similarity of internal oiling information of the atomizer at the tobacco tar injection speed, and the node characteristics of each node comprise the internal oiling information of the atomizer at the tobacco tar injection speed; and processing the knowledge graph based on the graph self-encoder to determine the target oiling speed. In one possible implementation, the internal oiling information of the atomizer at each tobacco tar injection speed includes oiling uniformity, bubble formation information, oiling filling degree, and oiling sealing degree. According to a second aspect, the present invention provides an electronic cigarette preparation system based on image analysis, comprising: the acquisition module is used for acquiring a plurality of continuous X-ray images shot by the atomizer at different tobacco tar injection speeds; the generation module is used for generating an internal oiling video of the atomizer at each tobacco tar injection speed by using a variation self-encoder on the basis of a plurality of continuous X-ray images shot by the atomizer at each tobacco tar injection speed; the information determining module is used for determining the similarity of the internal oiling information of the atomizer at each tobacco tar injection speed and the internal oiling information of the atomizer at different tobacco tar injection speeds by using an oiling information determining model based on the internal oiling video of the atomizer at each tobacco tar injection speed; the oiling speed determining mod