Disclosure of Invention
The invention aims to provide a control method of an intelligent vehicle-mounted system based on face recognition, which realizes vehicle data management and safety control through a background server and a mobile phone terminal app aiming at the defects of the prior art.
In order to achieve the purpose, the control method of the intelligent vehicle-mounted system based on the face recognition comprises a background server, a mobile phone end app, a WIFI 485 conversion assembly module and an LED internal and external display screen in sequence from top to bottom;
the method for realizing vehicle data management and safety control through the background server comprises the following steps:
the background server comprises four modules of terminal equipment management, program list setting, driver management and line travel, wherein the terminal equipment management module is used for grouping the LED internal and external screens; the program list setting module is used for setting text display editing, font setting, playing effect setting and display speed of the LED screen; the driver management module is used for auditing and inputting driver information, including name, password, photo, job number, ID card number and driver license model; the line travel module manages an operation route and consists of a starting station, a terminal station, a midway station, a departure time and an arrival time; the driver information is recorded into a background database after being checked by an administrator, when a driver is connected with a background server through a mobile phone end app, the background server sends personal information of the driver to the mobile phone end app for identity check, the check is passed through a continuous use system, the driver successfully logs in the mobile phone end app and signs in through face recognition, when the driver signs in, whether the driver is a preset person or not can be judged, if the driver is not the facial feature of the preset driver, the system can upload illegal information to the background server for later follow-up, meanwhile, the background server can send a request for turning off a vehicle ignition device to a vehicle body control system to prevent the vehicle from being illegally started, and the warning information is transmitted to an in-vehicle screen of an LED internal and external display screen through a WIFI/485 assembly module, and when the driver signs in successfully, the background server can automatically send current line travel, program information, program list information and program list information, The driver information is mobile phone end app, the AP connected with the WIFI/485 assembly module through the mobile phone end app sends data to the LED inner and outer display screens, and meanwhile, a request for starting the vehicle ignition device is automatically sent to the vehicle body control system, so that the vehicle is normally started.
The mobile phone end app sequentially registers the face, compares the face and acquires data from a background server through an http protocol and transmits the data to the WIFI/485 assembly module, the WIFI/485 assembly module provides a WIFI AP (wireless access point) function, the mobile phone end app is connected with the AP as an STA (wireless station) and then transmits the data to the WIFI/485 assembly module through a WIFI network, the data are finally converted into RS485 format data to be transmitted to the LED inner and outer display screens inside and outside the WIFI/485 assembly module, the LED inner and outer display screens display driver and prompt language information, and the LED outer screen display line information in the LED inner and outer display screens is connected with the WIFI/485 assembly module through an RS485 interface.
The method for realizing face recognition through the mobile phone terminal app comprises the following steps:
when the mobile phone app is opened, the mobile phone app enters a login interface, and a login name and a password are required to be filled in, the login name and the password are verified and added in advance through a driver management column of the background server, and a list which is not in a verification list does not allow a user to login the system. After the login name and the password are confirmed to be correct, the user clicks login, at the moment, the mobile phone app automatically checks the user name and the password from the background server by using an http protocol, the user enters a mobile phone app system after the user passes the check, the user needs to carry out face acquisition and comparison after entering the system, and the face detection and tracking, quality control, face acquisition and face comparison functions are realized by calling a Baidu face offline recognition SDK.
Firstly, starting face acquisition and registration, after SDK initialization, setting parameters such as exposure, fuzziness, face shielding threshold, detection angle, face detection quality and the like of an acquired picture, calling a method RgbDetectActivity to detect in real time to acquire a registered face, wherein the Activity designates a certain area on a screen, a user is required to be at the center of the screen, the position of the Activity is moderate from a camera, the face faces the acquisition camera, and if the posture of the user does not meet the requirement or the distance is too far, the user is prompted to adjust the corresponding position until a picture meeting the requirement is acquired. And when one frame of photo meeting the conditions is successfully collected, uploading the photo to a position of a driver management photo corresponding to the background server, downloading the photo to a local photo album of the mobile phone after the photo is successfully uploaded, comparing the subsequent photos, and finishing the collection and registration process.
And step two, performing face comparison and identification, wherein the face comparison and identification are performed when the user signs on duty and off duty each time, the identification is consistent, the sign-in success is prompted, and otherwise, the prompt fails. The human face comparison and identification is to calculate the similarity of two human faces by extracting the characteristics of the human faces so as to judge whether the human faces are the same person or not and give a similarity score within the Activity range of a mobile phone screen. And acquiring a face by using RgbDetectActivity in the SDK, calling syncFeature extraction features after the face is successfully acquired, comparing the extracted features with photos in a mobile phone album, successfully signing in through display, giving a higher similarity score, and otherwise, continuously identifying until the extracted features pass through, and finishing the face comparison identification process.
The method for controlling the LED display through the mobile phone end app and the WIFI/485 assembly comprises the following steps:
after the driver successfully signs in through the mobile phone app, the mobile phone app controls the route travel of the display screen outside the vehicle of the LED inside and outside, and the display area of the screen inside the vehicle displays program information and driver information. After receiving the information, the driver firstly checks whether the data is correct or not, and if the data is incorrect, the data can be manually modified. After the data is confirmed, the current network system is switched by clicking the setting option, and the current network system is connected to an AP (wireless access point) in an STA (wireless station) mode, wherein the AP is a wireless network formed by a WIFI/485 assembly module, and for the convenience of operation, a driver is advised to use a mobile phone to store AP information in advance, so that the AP can be automatically connected after the network is switched, and the AP can be in data communication with the WIFI/485 assembly module after being connected to the AP. The WIFI/485 assembly module is simply a black box for converting WIFI data into RS485 data, and supports a complete TCP/IP protocol stack. The mobile phone end app encodes data to be displayed and then directly transmits the encoded data to the WIFI/485 assembly module through the network socket, and the WIFI/485 assembly module automatically converts the data into RS485 format data after receiving the data and transmits the RS485 format data to the LED inner and outer display screens.
Compared with the prior art, the vehicle data management and safety control are realized through the background server, the face recognition is realized through the mobile phone APP terminal, the LED display is controlled through the mobile phone APP terminal and the WIFI/485 assembly module, the vehicle and personnel management efficiency is improved, the vehicle and personnel use safety is improved, the personnel use the vehicle conveniently, and the working efficiency is improved.
Detailed Description
The technical scheme in the embodiment of the invention is clearly and completely described below.
Example 1.
The control method of the intelligent vehicle-mounted system based on the face recognition described in the embodiment comprises a background server, a mobile phone end app, a WIFI to 485 assembly module and an LED internal and external display screen;
the background server comprises four modules of terminal equipment management, program list setting, driver management and line travel, wherein the terminal equipment management module is used for grouping the LED internal and external screens; the program list setting module is used for setting text display editing, font setting, playing effect setting and display speed of the LED screen; the driver management module checks and inputs driver information, wherein the driver information comprises a name, a password, a photo, a job number, an identity card number and a driver license model; the line travel module manages an operation route, and the operation route consists of a starting station, a terminal station, a midway station, departure time and arrival time; the driver information is required to be checked by a manager and then is recorded into a database of a background server, the driver is connected with the background server through a mobile phone end app, the background server sends personal information of the driver to the mobile phone end app for identity check, the check is passed through a continuous use system, the driver successfully logs in the mobile phone end app and signs in through face recognition, whether the current driver is a preset person or not can be judged when the driver signs in, when the face characteristics are inconsistent with the preset person, the mobile phone end app can upload illegal information to the background server, meanwhile, the background server can send a request for turning off a vehicle ignition device to a control system of a vehicle, meanwhile, the warning information is transmitted to an inner screen of the LED inner and outer display screen through a WIFI/485 assembly module, and when the driver signs in successfully, the background server issues current line travel, program list information and driver information to a mobile phone end app, and the AP connected with the WIFI/485 assembly module through the mobile phone end app sends data to the LED inner and outer display screens, and simultaneously sends a request for starting the vehicle ignition device to the control system of the vehicle, so that the vehicle is normally started.
The mobile phone end app sequentially performs face registration, face comparison and data acquisition from a background server through an http protocol and transmits the data to the WIFI/485 assembly module, the WIFI/485 assembly module provides a WIFI AP (wireless access point) function, the mobile phone end serves as an STA (wireless station) to be connected with the AP and then transmits the data to the WIFI/485 assembly module through a WIFI network, the data are finally converted into RS485 format data inside the WIFI/485 assembly module and then transmitted to the LED inner and outer display screens, and the LED inner and outer display screens are connected with the WIFI/485 assembly module through RS485 interfaces.
The mobile phone end app enters a login interface when opened, and requires to be filled with a login name and a password, the login name and the password are checked and added in advance through a driver management column of a background server, the login system is not allowed in a list in an audit list, the login name and the password are clicked for login after being confirmed to be correct, at the moment, the mobile phone end app automatically checks a user name and a password from the background server by using an http protocol, the user enters the mobile phone end app system after the user name and the password are checked, the user needs to acquire and compare the face after the user enters the system, and the functions of face detection and tracking, quality control, face acquisition and face comparison are realized by calling a hundred-degree face offline recognition SDK; firstly, starting face collection and registration, after SDK initialization, setting parameters such as exposure, fuzziness, face shielding threshold, detection angle, face detection quality and the like of collected pictures, calling a method RgbDetectActivity to detect in real time to obtain registered faces, wherein the Activity designates a certain area on a screen, requires that a user is at a central position of the screen, has a moderate distance from a camera, and faces the collected camera; and step two, face comparison and recognition are carried out, a user can carry out face comparison and recognition each time when signing on or off duty, the recognition is consistent, the sign-in success is prompted, otherwise, the sign-in failure is prompted, the face comparison and recognition is that the face features are extracted, the similarity of two faces is calculated, whether the two faces are the same person is judged, a similarity score is given in the Activity range of a mobile phone screen, the face is obtained by using RgbDetectActivities in the SDK, syncFeature extraction features are called after the face is obtained successfully, the features are extracted and compared with pictures in a mobile phone album, the recognition is successful through display, a higher similarity score is given, and otherwise, the recognition is carried out until the face comparison and recognition process is completed.
After the driver successfully signs in through the mobile phone app, the mobile phone app controls the route travel of the display screen outside the vehicle of the LED inside and outside, and the display area of the screen inside the vehicle displays program information and driver information. After receiving the information, a driver firstly checks whether the data are correct or not, if the data are incorrect, the data can be modified manually, after the data are confirmed to pass, the current network system can be switched and connected to an AP (wireless access point) in an STA (wireless station) mode, the AP is a wireless network formed by a WIFI/485 assembly module, for the convenience of operation, the driver is advised to use a mobile phone to store AP information in advance, the AP can be automatically connected after the network is switched, and the data can be communicated with the WIFI/485 assembly module after the AP is connected. The WIFI/485 assembly module is a black box used for converting WIFI data into RS485 data in short, and supports a complete TCP/IP protocol stack, the mobile phone end app directly transmits the data to be displayed to the WIFI/485 assembly module through a network socket after encoding, and the WIFI/485 assembly module automatically converts the data into RS485 format data after receiving the data and transmits the data to the LED inner and outer display screens.