CN211653673U - Cigarette lighter for collecting cigarette specification and human face big data information - Google Patents
Cigarette lighter for collecting cigarette specification and human face big data information Download PDFInfo
- Publication number
- CN211653673U CN211653673U CN202020163077.1U CN202020163077U CN211653673U CN 211653673 U CN211653673 U CN 211653673U CN 202020163077 U CN202020163077 U CN 202020163077U CN 211653673 U CN211653673 U CN 211653673U
- Authority
- CN
- China
- Prior art keywords
- cigarette
- module
- lighter
- big data
- face
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Image Analysis (AREA)
Abstract
The utility model relates to the technical field of cigarette lighters, in particular to a cigarette lighter for collecting cigarette specifications and human face big data information, which comprises a processing unit, and a memory, a cigarette lighting module, a human face recognition module, a cigarette case recognition module and a networking module which are respectively connected with the processing unit; the utility model is arranged in smoking places, attracts smokers to use through the cigarette lighting module, collects face information and cigarette case information of the smokers and associates the face information and the cigarette case information when in use, and truly knows real passenger classification and consumer figures of different cigarette brands to guide marketing strategies; the real data support can be provided for the smoking crowd for further analyzing the single-item cigarette products; real data support is provided for the establishment of systems such as cigarette production in the production link of tobacco enterprises, marketing activities in the marketing link of tobacco enterprises, tobacco control organization regulations and the like; the non-cigarette product can be accurately recommended to the consumer according to the price level of the cigarette smoked by the consumer, the individual characteristics of the consumer and the like.
Description
Technical Field
The utility model relates to a cigar lighter technical field specifically is a cigar lighter for gathering cigarette specification and people's face big data information.
Background
In many smoking places, such as airport terminal buildings, high-speed rail waiting rooms, meeting center smoking rooms and the like, for convenience of smokers, particularly smokers who do not carry lighters or matches, the cigarette lighting device can be placed and deployed. The cigarette lighter on the existing market is used for safety consideration and convenience for smokers, and a plurality of trigger switches are adopted, namely, the smokers trigger the cigarette lighter to work through buttons and other forms, such as electric arc energization and tungsten wire heating. Although the existing cigarette lighter realizes and is convenient for smokers to light cigarettes, the existing cigarette lighter does not record and sense the users using the device, and does not have any sensing capability on the attributes of the users and the products of lighting the cigarettes. With the development of technologies such as artificial intelligence, big data, internet of things and the like, the intelligent upgrading of the terminal becomes a trend, and the intelligent level of the original cigarette lighter and the new requirements of big data acquisition and equipment management are difficult to meet.
SUMMERY OF THE UTILITY MODEL
In view of this, the utility model aims at providing a cigar lighter for gathering cigarette specification and people's face big data information adopts intelligent vision perception trigger technology through the trigger device part that changes the cigar lighter, replaces the current cigar lighter adopt trigger modes such as button, touching response, distance response.
In order to achieve the above purpose, the utility model adopts the following technical scheme:
the utility model discloses a cigar lighter for gathering cigarette specification and big data information of human face, including the processing unit and respectively with memory, some cigarette module, face identification module, cigarette case identification module and the networking module of processing unit connection;
the cigarette lighting module provides a cigarette lighting function for smokers;
the face recognition module is used for collecting face information of a smoker right in front of the cigarette lighter;
the cigarette case identification module is used for collecting cigarette case information of smokers during smoking;
the processing unit is used for processing and associating the collected face information and the cigarette case information and storing the associated information into the memory;
and the networking module is used for networking the cigarette lighter so as to realize data uploading.
Further, the utility model discloses a cigar lighter still includes interactive screen and light filling lamp, interactive screen is used for broadcast operation suggestion and advertisement, the light filling lamp is used for illuminating the face of smoker and is convenient for gather facial image information.
Further, the cigarette lighting module comprises a power supply, a transformer, a time relay, a switch and an electric arc lighter which are sequentially connected.
Further, the face recognition module is a first RGB camera.
Further, the cigarette case identification module includes second RGB camera and toper formation of image cover, the second RGB camera sets up the less one end of toper formation of image cover aperture.
Further, the processing unit is an RK3288 processor.
Further, the networking module includes but is not limited to a network card, a wireless network card, a GPRS module, a WIFI module and a 4G module.
The beneficial effects of the utility model are as follows: the utility model discloses a cigar lighter for gathering cigarette specification and big data information of face sets up in smoking place, attracts the smoker to use through the point cigarette module, gathers the face information and the cigarette case information of smoker and associates when using, really knows the real guest crowd classification of different cigarette brands and consumer portrays and guides the marketing strategy; the real data support can be provided for the smoking crowd for further analyzing the single-item cigarette products; real data support is provided for the establishment of systems such as cigarette production in the production link of tobacco enterprises, marketing activities in the marketing link of tobacco enterprises, tobacco control organization regulations and the like; accurate recommendation of non-cigarette products can be carried out on consumers according to the price level of cigarettes smoked by the consumers, the individual characteristics of the consumers and the like; the utility model is suitable for a smoking room that established the tobacco trade of the million ten thousand, other organizations etc..
Drawings
The invention will be further described with reference to the following figures and examples:
fig. 1 is a schematic structural view of the present invention;
fig. 2 is a schematic structural view of the cigarette lighting module of the present invention;
FIG. 3 is a flow chart of the image recognition algorithm of the present invention;
the system comprises a 1-interactive screen, a 2-face recognition module, a 3-cigarette lighting module, a 4-cigarette case recognition module and a 5-light supplement lamp.
Detailed Description
As shown in fig. 1: the cigarette lighter for collecting cigarette specifications and human face big data information comprises a processing unit, and a memory, a cigarette lighting module 3, a human face recognition module 2, a cigarette case recognition module 4 and a networking module which are respectively connected with the processing unit;
the cigarette lighting module 3 provides a cigarette lighting function for smokers;
the face recognition module 2 is used for collecting face information of a smoker right in front of the cigarette lighter;
the cigarette case identification module 4 is used for collecting cigarette case information of smokers during smoking;
the processing unit is used for processing and associating the collected face information and the cigarette case information and storing the associated information into the memory;
the networking module is used for networking the cigarette lighter so as to realize data uploading.
Be provided with interactive screen 1 on the cigar lighter front panel, be provided with cigarette case identification module in display screen bottom right side on the cigar lighter front panel, be provided with some cigarette module on the cigar lighter front panel in cigarette case identification module left side, be provided with the face identification camera under interactive screen 1 on the cigar lighter front panel, the inside processing unit that is provided with of cigar lighter, processing unit adopts the model to be RK3288 treater, RK3288 treater can use the tall and erect system of ann, cooperation networking module utilizes the network to establish data synchronization with the cloud terminal and is connected, RK3288 treater built-in serial circuits is connected with the inside relay of some cigarette module 3, cigarette case identification module and face identification camera all are connected with processing unit through data line. The processing terminal is connected with the cloud terminal through the WIFI module, cloud synchronization software is installed in a processor in the processing terminal, and the preset port is connected and controlled by the PC end and stores cigarette case information, identity information and face information.
In this embodiment, the utility model discloses a cigar lighter still includes interactive screen 1 and light filling lamp 5, and interactive screen 1 is used for broadcast operation suggestion and advertisement, and light filling lamp 5 is used for illuminating the face of smoker and is convenient for gather facial image information. Wherein light filling lamp 5 contains surrounding type light filling and relay control panel, and the relay control panel links to each other with processing unit, and intelligent control light filling switch and light filling are long.
In the embodiment shown in fig. 2, the ignition module 3 includes a power supply, a transformer, a time relay, a switch and an arc ignitor connected in sequence. The power is connected with the electric arc lighter, a switch is arranged between a PCB (printed circuit board) of the electric arc lighter and the power, a time relay is arranged between the power and the switch, the PCB is connected with a transformer, and the transformer is connected with an electric arc end ignition contact. The power supply reduces the voltage to the voltage (for example, 3.7V) required by the PCB through the transformer, and then outputs the voltage to the electric arc lighter to supply power to the electric arc lighter. The transformer described above converts a voltage of 12V to a voltage of 3.7V. A time relay is arranged between a power supply and a switch, and the on time and the off time of the time relay are set, so that the power supply can be switched on for ignition in the on time of the time relay, and the power supply cannot be switched on for ignition in the off time, so that a circuit is protected; for example, the on time is set to 20 minutes and the off time is set to 1 minute, so that the electronic igniter is turned off for 1 minute after operating for 20 minutes, and heat is dissipated within the off time of 1 minute to protect the circuit of the electronic igniter. Convert the voltage value that is fit for electric arc sparker through the transformer with the power, at the switch on state between transformer and electric arc sparker, whole return circuit does not circular telegram, and electric arc sparker is out of work, and when the switch was in the closure state, the transformer was the power supply of electric arc sparker, and electric arc sparker gets the electricity and begins work.
In this embodiment, the face recognition module 2 is a first RGB camera; the cigarette case identification module 4 comprises a second RGB camera and a conical imaging cover, and the second RGB camera is arranged at one end of the conical imaging cover with a smaller opening diameter; stably grab the cigarette case photo and transmit to the processing terminal.
The first RGB camera and the second RGB camera specifically adopt common CMOS RGB visible light cameras, can collect images in real time, and are provided with active light source equipment such as a light supplement lamp 5, so that a good image collecting effect is achieved. The CMOS camera adopts a self-adaptive wide dynamic ISP processor, so that the camera module can stably acquire image data with quality conforming to algorithm processing under different environmental conditions. Wherein the camera used for collecting the cigarette case image is arranged behind the cigarette case placing area and can shoot all cigarette case areas on the device. The face image acquisition camera needs to have a larger field angle, and can acquire face images which appear at different spatial positions due to different use habits.
In this embodiment, the networking module includes, but is not limited to, a network card, a wireless network card, a GPRS module, a WIFI module, a 4G module, and the like.
The utility model discloses still provide an algorithm that is used for face identification and cigarette case discernment specially, at first, processing unit contains two treater: the edge calculation processor is used for calling a cigarette case image recognition algorithm running on the edge calculation processor after the edge calculation processor acquires the image of the cigarette case image acquisition camera, so that the recognition of the cigarette case appearance image is realized. The results of the model are shown in figure 3,
the image classification by using the convolutional neural network has become a standard deep learning method, however, the cigarette specification recognition is trained by using a direct classification model, and the difficulties are as follows:
1) the images on the front and the back of the cigarette case are different;
2) the image samples of the cigarette packs sold in larger quantities and the image samples of the cigarette packs sold in smaller quantities are severely unbalanced, and only one sample can be used for training in a large part of the classes with not long time to market or few quantities. If the direct classification training method is adopted, the small samples cannot be effectively trained due to the fact that training samples are distributed seriously unevenly.
In order to solve the problem 1), the utility model provides a positive and negative side branch training method, in the training process, separately mark the positive and negative cigarette case of unified specification promptly, carry out the model training as two categories.
In order to solve problem 2), the utility model provides an adopt hierarchical categorised training mode. The classes with the training sample number smaller than a certain value are divided into a large class, the K classes with the training sample number larger than or equal to the certain value are independently used as classes to form a new K +1 class for training, and then the large class is classified.
The input image of the face recognition is a detected face area, and the extraction of face identity codes, gender recognition, age recognition and emotion recognition are realized by adopting the same network through the training of a deep neural network. The face identity code is a 128-dimensional floating point number, the floating point number is a face identity feature code, the face identity code has personal uniqueness, and when the cosine distance of the identity feature code is smaller than a threshold value T, two face images can be judged to belong to the same identity individual. The output sex is 1 and 0, 1 represents the sex of a male, 0 represents the sex of a female, the age is the segmental age, the five age stages are divided into 1-18, 18-25, 25-40, 40-60, 60 and above, and the emotion is four types of normal, joyful, anxious and sad emotion.
In training this model process, this utility model adopted the multitask learning training method, because the degree of difficulty of a plurality of tasks above the same face image simultaneous annotation is great, also hardly ensures the sample equilibrium simultaneously. The utility model discloses a following network architecture:
firstly, obtaining an optimal human face characteristic representation mode by adopting an identity training basic convolution frame with the largest markable quantity, then fixing parameters of a convolution layer, respectively training a gender branch network structure, an age branch network structure and an emotion branch network structure, achieving common convolution, and then realizing network structures of different tasks.
Cigarette case discernment and face recognition algorithm all operate on the marginal calculation chip that has the deep learning convolution neural network throughput, the utility model discloses tailor and optimize the model, can realize the real-time processing of image on the CPU marginal calculation chip of dual-core ARM Cortex-A53 or above performance promptly.
After the cigarette case and the face recognition are realized, a signal is sent to a central processing unit, the central processing unit judges the recognition result and sends the judgment result to a cigarette lighting module to realize the cigarette lighting function; meanwhile, the recognition result is pushed to a cloud receiving server in a network post mode. And the cloud big data server performs structured storage and management on the received information.
The utility model discloses an intelligent visual perception triggers technique possesses when triggering the cigar lighter and realizing normal work, can collect the information that the visual perception of the user that uses this cigar lighter arrived, include but not only be limited to user's identity, age, sex, dress and this user are used for cigarette brand and the specification of some cigarette at present. The processor of the cigarette lighter transmits the collected information to the cloud end by being connected with the internet, so that the collection of big data is realized.
In the construction process of sharing smoking rooms and smoking areas, the utility model can provide better cigarette lighting service for consumers, and can automatically identify the cigarette case by the cigarette case identification algorithm, thereby automatically acquiring the perfect product information of specification, price, taste and the like of the cigarettes smoked by the consumers; the face of a consumer is identified through a face identification algorithm, face characteristic information of the consumer, including desensitization information (ensuring user information safety) such as age, gender, emotion and dressing, is automatically acquired, and a processor of the cigarette lighter transmits the acquired information to a cloud end through connecting the internet, so that big data collection is realized; through an algorithm, the cigarette case information and the face information identified by the two items are subjected to information association, and the application value of the associated information is as follows:
(1) truly understanding the real customer group classification and consumer representation of different cigarette brands;
(2) guiding the marketing strategy;
(3) the real data support can be provided for the smoking crowd for further analyzing the single-item cigarette products;
(4) real data support can be provided for further analysis of the consumption fashion trend;
(5) real data support is provided for the establishment of systems such as cigarette production in the production link of tobacco enterprises, marketing activities in the marketing link of tobacco enterprises, tobacco control organization regulations and the like;
(6) accurate recommendation of non-cigarette products can be carried out on consumers according to the price level of cigarettes smoked by the consumers, the individual characteristics of the consumers and the like;
the utility model is suitable for a smoking room that established the tobacco trade of the million ten thousand, other organizations etc..
Finally, although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the present invention can be modified or replaced by other means without departing from the spirit and scope of the present invention, which should be construed as limited only by the appended claims.
Claims (7)
1. The utility model provides a cigar lighter for gathering cigarette specification and people's face big data information which characterized in that: the cigarette lighter comprises a processing unit, and a memory, a cigarette lighting module, a face recognition module, a cigarette case recognition module and a networking module which are respectively connected with the processing unit;
the cigarette lighting module provides a cigarette lighting function for smokers;
the face recognition module is used for collecting face information of a smoker right in front of the cigarette lighter;
the cigarette case identification module is used for collecting cigarette case information of smokers during smoking;
the processing unit is used for processing and associating the collected face information and the cigarette case information and storing the associated information into the memory;
and the networking module is used for networking the cigarette lighter so as to realize data uploading.
2. The cigar lighter for gathering cigarette specification and human face big data information according to claim 1, characterized in that: still include interactive screen and light filling lamp, interactive screen is used for broadcast operation suggestion and advertisement, the light filling lamp is used for illuminating the face of smoker and is convenient for gather facial image information.
3. The cigar lighter for gathering cigarette specification and human face big data information according to claim 1, characterized in that: the cigarette lighting module comprises a power supply, a transformer, a time relay, a switch and an electric arc lighter which are connected in sequence.
4. The cigar lighter for gathering cigarette specification and human face big data information according to claim 1, characterized in that: the face recognition module is a first RGB camera.
5. The cigar lighter for gathering cigarette specification and human face big data information according to claim 1, characterized in that: the cigarette case identification module includes second RGB camera and toper formation of image cover, the second RGB camera sets up the less one end of toper formation of image cover aperture.
6. The cigar lighter for gathering cigarette specification and human face big data information according to claim 1, characterized in that: the processing unit is an RK3288 processor.
7. The cigar lighter for gathering cigarette specification and human face big data information according to claim 1, characterized in that: the networking module comprises but is not limited to a network card, a wireless network card, a GPRS module, a WIFI module and a 4G module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202020163077.1U CN211653673U (en) | 2020-02-12 | 2020-02-12 | Cigarette lighter for collecting cigarette specification and human face big data information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202020163077.1U CN211653673U (en) | 2020-02-12 | 2020-02-12 | Cigarette lighter for collecting cigarette specification and human face big data information |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN211653673U true CN211653673U (en) | 2020-10-09 |
Family
ID=72685032
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202020163077.1U Active CN211653673U (en) | 2020-02-12 | 2020-02-12 | Cigarette lighter for collecting cigarette specification and human face big data information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN211653673U (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111126355A (en) * | 2020-02-12 | 2020-05-08 | 深圳爱莫科技有限公司 | Cigarette lighter for collecting cigarette specification and human face big data information |
-
2020
- 2020-02-12 CN CN202020163077.1U patent/CN211653673U/en active Active
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111126355A (en) * | 2020-02-12 | 2020-05-08 | 深圳爱莫科技有限公司 | Cigarette lighter for collecting cigarette specification and human face big data information |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105301997B (en) | Intelligent prompt method and system based on mobile robot | |
| CN108111948A (en) | The visual output that server at speech interface equipment provides | |
| CN110264299A (en) | Clothes recommended method, device and computer equipment based on recognition of face | |
| CN109379819A (en) | Control method, the program of communication terminal | |
| CN108292320A (en) | Information processing unit, information processing method and program | |
| US11018939B1 (en) | Determining product compatibility and demand | |
| CN107729380A (en) | Clothing matching method, terminal, terminal | |
| CN109805673A (en) | A kind of Intelligent mirror | |
| CN211653673U (en) | Cigarette lighter for collecting cigarette specification and human face big data information | |
| US20180158102A1 (en) | Advertising display system using smart film screen | |
| CN109361852A (en) | A kind of image processing method and device | |
| CN107198391A (en) | A kind of makeup instructs U.S. face mirror | |
| CN112446590B (en) | Comprehensive student management system, method, medium and terminal | |
| CN109376621A (en) | A kind of sample data generation method, device and robot | |
| CN108881975A (en) | Method, apparatus, terminal device and the storage medium that TV is reminded | |
| CN109683705A (en) | The methods, devices and systems of eyeball fixes control interactive controls | |
| CN104267817B (en) | Method of intelligently adjusting screen and smart screen | |
| CN113180427A (en) | Multifunctional intelligent mirror | |
| EP2228766A1 (en) | Data processing system, its computer program and data processing method | |
| CN111126355A (en) | Cigarette lighter for collecting cigarette specification and human face big data information | |
| JP2021513708A (en) | Methods and systems for classifying food | |
| CN109493583A (en) | Intelligent landscape management system method for building up based on context aware and image procossing | |
| CN211479207U (en) | Intelligent cigarette case identification access control equipment | |
| US10665082B1 (en) | Providing device power-level notifications | |
| CN114190823B (en) | Intelligent household robot and control method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Long Tao Inventor after: Chao Hang Inventor before: Yang Heng Inventor before: Long Tao Inventor before: Chao Hang |