KR20190023362A - Device using mobile phone for warning of the use of the mobile phone while driving a car, and control method thereof - Google Patents

Abstract

The present invention relates to a technique for warning use of a mobile phone while driving a vehicle using a GPS receiver mounted on a mobile phone, a camera, a microphone, a touch screen, a vibration element, a speaker, a memory unit, and an operation control unit controlling the entire operation. Described are a device for warning use of a mobile phone while driving a vehicle using the mobile phone and a control method thereof. The device is configured to determine whether a user uses a the mobile phone while driving the vehicle by detecting a real-time position of the user with a GPS receiver and using an automated object recognition learning algorithm based on a feature of a frontal image photographed by a camera while operating the mobile phone at the same time and to determine whether a use state of the mobile phone is a calling state while driving the vehicle, a received text reading state, or a transmitting text writing state to generate a warning signal selectively for the mobile phone and vehicle according to a determination result.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a mobile phone use warning device and a control method thereof,

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a mobile phone use warning device and a control method thereof during operation of a vehicle using a mobile phone and, more particularly, to a mobile phone use warning device and a control method thereof, A navigation app with a DB (Database), and an operation control unit for controlling the entire operation, detects the real time position of the user, determines that the user uses the mobile phone while driving, And a method of controlling the mobile phone using the mobile phone using the mobile phone which selectively generates a warning signal to the mobile phone and the vehicle according to a result of determining whether the received character is in a reading state or a receiving character writing state will be.

With the popularization of mobile phones today, accidents caused by alcohol use, sleepy driving, and cell phone use are increasing, and the damage to oneself and others is increasing. According to a Korean research institute for civilian traffic culture, for example, when driving at 40 km / h and using mobile phones, the distance to stop is 1.5 times longer than the stopping distance in drinking and driving state of blood alcohol 0.05%, which is more dangerous than drinking driving .

According to the results of analyzing the use of mobile phone while driving at Korea Traffic Safety Authority, the first problem is that the ability to cope with the situation is considerably low. For example, the distance for stopping the signal while driving at 60 km per hour is about 23 m Second, it takes about 0.7 seconds for steady state, 1.6 seconds for normal state, and 1 second for normal state, while it is about 41m for medium and 60m for longer than normal. The response rate is more than two times slower. Third, the anterior main rate is also about 90% in normal condition, 57% in case of talking, 24% in case of exchanging text messages and 17% in case of manipulating navigation Respectively. As described above, when using a mobile phone during driving, attention and ability to cope with all of them are significantly reduced. In some cases, even though it may cause more dangerous situations than drinking driving, most people use it without any consciousness.

In recent years, techniques for recognizing the above-mentioned risk and preventing such a risk have been proposed. For example, Korean Patent Laid-open Publication No. 10-2017-0016579 discloses a technique for detecting right and left faceball regions in a face image captured by a camera, And then a straight line component of the mobile phone structure is detected in this area to alert the call status during operation [Patent Document 1]. In addition, Korean Patent No. 10-0796714 discloses a technology for limiting an incoming / outgoing call by determining a driving state by a GPS receiver and a hand-free set built in a mobile phone [Patent Document 2]. In another method, Korean Patent No. 10-0638326 discloses a technique in which a specific sensor is installed in a vehicle to allow a mobile phone base station to transmit a voice signal indicating that the driver is currently driving the mobile phone base station to the other party to restrict a call [Patent Document 3].

However, in the case of Korean Patent Laid-Open Publication No. 10-2017-0016579 using a camera as described above, it is necessary to perform a complicated process in which a face is detected by detecting a face area using a known technique, And the camera keeps track of the face continuously, it is not technically possible to accurately detect the image due to the driver's posture or movement, to check whether the mobile phone is in operation while driving, to check the received character or to write a text message for transmission There were many difficulties. That is, an accident caused by the use of a mobile phone during operation occurs instantly. Therefore, an accurate image is quickly detected, and a call activity of the driver or an operation of the mobile phone is immediately determined based on the detected image image. The technique can be a major obstacle to accuracy problems in the detection phase and the time lag from the determination. Also, in the case of Korean Patent No. 10-0796714 using the hand preset, a hand preset must be installed in the vehicle to restrict the call during driving. If the user disconnects the handset from the hand preset, Korean Patent No. 10-0638326 using the mobile telephone base station means a preventive measure for controlling the incoming call of the mobile phone while driving, but it is also necessary to install a specific sensor capable of networking with the mobile phone station separately in the vehicle It is not possible to detect and warn even an operation for reading a received text message of a mobile phone or an operation for writing a character for sending, in addition to the above-mentioned conventional technologies.

[1] Korean Patent Laid-Open Publication No. 10-2017-0016579 (entitled: Call Detection and Warning Method During Operation and Its System). [2] Korean Patent No. 10-0796714 (entitled: Cellular phone that restricts currency during driving). [3] Korean Patent No. 10-0638326 (entitled " Mobile Phone Reception Control Method and Device During Operation ").

[1] Ojala, T., Pietikainen, M. and Harwood, D., A Comparative Study of Texture Measures with Classification Based on Feature Distributions. Pattern Recognition 29 (1): 51-59, 1999). [2] Ojala, T., Pietikainen, M. and Maenpaa, T., Multiresolution Gray-scale and Rotation Invariant Texture Classification with Local Binary Patterns. IEEE Trans. Pattern Analysis and Machine Intelligence 24 (7): 971-987,2002). [3] S. Liao, X. Zhu, Z. Lei, L. Zhang and S. Z. Li, "Learning Multi-scale Block Local Binary Patterns for Face Recognition," ICB 2007. [4] http://opencv.willowgarage.com/wiki/, Jan. 2011. [5] Ramin Zabih and John Woodfill, "Non-parametric Local Transforms for Computing Visual Correspondence," ECCV 1994. [6] Bernhard Froba and Andreas Ernst, "Face Detection with the Modified Census Transform", FG 2004. [7] B. Froba, and A. Ernst, "Face Detection with the Modified Census Transform", In Proceedings of the Sixth IEEE Conference on Automatic Face and Gesture Recognition, pp. 91-96, May, 2004. [8] T. Song, Y. Lee, M. Kim, B. Ku, and H. Ko, "Fusion Methods of License Plate Detection and Super Resolution for Improving License Plate Recognition" , Vol. 16, No. 4, pp. 53-60, April, 2011.  [9] Y. Freund and RE Schapire, "A Decision-Theoretic Generalization of On-Line Learning and an Application to Boosting", Journal of The Computer Society and Systems, vol. , pp. 119-139, March, 1997. [10] R. E. Schapire and Y. Singer, "Improved Boosting Algorithms Using Confidence-rated Predictions" Machine Learning, vol. 37, pp. 297-336, December, 1999. [11] J. Friedman, T. Hastie and R. Tibshirani, "Additive Logistic Regression: a Statistical View of Boosting", Annals of statistics, vol. 28, no. 2, pp. 377-407, August, 2000.

The mobile phone use warning device and the control method thereof using the mobile phone according to the present invention can be applied to a navigation application equipped with a GPS receiver, a camera, a microphone, a touch screen, a vibration device, a speaker, a memory, And an arithmetic control unit for controlling the entire operation and an arithmetic control unit for controlling the whole operation are used to detect a call, a reading of a received text message or a character writing state during a vehicle operation, And an object of the present invention is to provide a mobile phone use warning device and a control method thereof for driving a vehicle using a mobile phone configured to generate a warning signal and a deceleration signal selectively in a vehicle.

According to an embodiment of the present invention, a mobile phone having a GPS receiver, a camera, a microphone, a touch screen, a vibration device, a speaker, a memory unit, a navigation app with a geographic information DB, and an operation control unit The mobile phone use warning device detects the position of the user by using the GPS receiver, detects an image of the front face when the mobile phone is operated using the camera, detects a voice signal in the communication using the microphone, A sensing device for sensing a received character reading or a writing character writing signal using the sensing device; A reference value necessary for judging whether the vehicle is moving to a vehicle according to a signal detected from the sensing device, whether a call is being made during operation or whether the touch screen is being operated, an automated analysis algorithm for object identification, A memory unit for storing a value; A geographic information DB storing the speed limit or safe driving information of the vehicle to the road or the area; Each of the signals detected by the sensing device is compared with the reference value, the analysis algorithm for object recognition, and the feature value of the object candidate group stored in the memory unit to determine whether the user is driving, is in a call, And determines whether or not the mobile phone is in a busy state, a received character reading state, or a writing character writing state according to a result of the determination. And compares the information on the real-time position and the traveling speed detected by the GPS receiver with the information stored in the geographical information DB to output driving information on the road and an identification signal indicating overspeed An operation control unit; A signal driver for outputting a corresponding drive signal in accordance with each of the control signal and the identification signal according to the determination result of the operation control unit; When receiving a driving signal corresponding to a control signal in a call, character reading or text writing state during operation from the signal driving unit, generates a vibration alarm sound, and, in response to a driving signal corresponding to an identification signal obtained by comparing the GPS receiver and the geographic information DB Warning means for temporarily stopping a warning voice message, warning screen or cellular phone operation on the cellular phone; Wireless transmission means for wirelessly transmitting control signals and identification signals output from the operation control unit; And vehicle control means for activating the blinking signal of the vehicle when receiving the control signal from the operation control unit by the wireless transmission means and decelerating the vehicle upon receiving the identification signal whose moving speed exceeds the limit speed.

The mobile phone use warning device during operation of a vehicle using a mobile phone according to an embodiment of the present invention is characterized in that the sensing device includes a GPS sensing part for sensing a real time position of a user by a GPS receiver, And a screen for detecting operation of the touch screen of the mobile phone for checking the received text message or inputting the received text message, A sensing unit; The memory unit is a reference value for determining whether the mobile phone is used during operation, and is a reference value for a moving distance for determining whether the user is moving to the vehicle, a number of operations of the touch screen for determining a character reading or character writing state A reference value for a voice frequency for determining a call state, a reference value for a duration of a call state, a character read and a character write state; The arithmetic and control unit may include a movement judging unit for judging whether or not the user moves by comparing an arithmetic average of the real time position detected by the GPS sensor with a reference value for determining the movement distance stored in the memory unit, The camera of the image sensing unit is actuated to capture an image of the front face of the mobile phone when the mobile phone is operating, and the captured image image is processed by the characteristic value of the handle candidate candidate stored in the memory unit and the analysis algorithm, A pattern recognition unit for recognizing whether the mobile phone user belongs to a pattern feature and determining whether the mobile phone user is operating; a voice recognition unit for detecting a voice signal of the mobile phone and comparing the voice signal with a reference potential value for voice frequency stored in the memory unit, From the screen sensing unit, An operation judging unit for judging whether or not an operation signal according to a judgment result of the communication judging unit and the operation judging unit lasts for a predetermined time and judges whether the user is in a busy state, And outputs the first, second, and third control signals; and a travel information determination unit that outputs an identification signal corresponding to a comparison result between the GPS receiver and the geographic information DB ; Wherein the warning means includes a warning sound generating device for generating a vibration alarm sound at a predetermined interval when the driving signal corresponding to each state is inputted from the signal driving portion, a warning screen generating device for generating a warning screen on the touch screen, And an operation cut-off device for forcibly stopping the operation of the cellular phone according to the identification signal from the information determination section; The vehicle control unit includes a vehicle control unit for controlling the vehicle, an emergency blinker for generating a blink signal when the first, second, and third control signals corresponding to the respective states are input from the operation control unit, And a vehicle speed reducer that decelerates the vehicle when an identification signal exceeding the limit speed is input.

Preferably, according to the mobile phone use warning device during driving of a vehicle using a mobile phone according to an embodiment of the present invention, the object recognition analysis algorithm may be implemented using an LBP (Local Binary Pattern) or an MCB (Modified Census Transform) AdaBoost) learning algorithm.

More preferably, according to the mobile phone use warning device during operation of a vehicle using a cellular phone according to an embodiment of the present invention, the signal transmission means between the operation control section and the vehicle control means is a radio transmission device and a radio reception device, An infrared (IR) transceiver, a Bluetooth for short range communication, or a ZigBee module.

A vehicle using a mobile phone having a GPS receiver, a camera, a microphone, a touch screen, a vibration device, a speaker, a memory unit, a navigation app with a geographic information DB, and an operation control unit for controlling the entire operation, according to another embodiment of the present invention A method of controlling a mobile phone use warning device during operation comprises: S1) receiving a position signal of a user's mobile phone by a GPS receiver at all times at the start of the warning device; Operating the user to use the mobile phone (S2); If the mobile phone operates in the step S2, the operation control unit measures the moving distance per unit time by the GPS receiver, compares the arithmetic average value with the reference value of the moving distance stored in the memory unit, If the speed is lower than the predetermined speed, the process returns to step S2. If the speed exceeds the speed, it is determined that the vehicle is moving to the vehicle (S3). If it is determined that the vehicle is moving to the vehicle, the arithmetic and control unit controls the cellular phone to simultaneously photograph the front face of the vehicle, and uses the characteristic values of the candidate handle group stored in the memory unit and the automatic analysis algorithm for object recognition, If it is determined that the feature of the image captured in front is included in the inherent characteristic of the steering wheel, and if it does not belong to the inherent characteristic of the steering wheel, the process returns to the step S2. Step S4; If it is determined that the user is directly driving in the step, the operation control unit detects the voice signal by the microphone and determines whether the sensed value is within the reference value range of the voice frequency stored in the memory unit. If not, (S2), and if the voice signal is sensed, determining (S5) that the mobile phone user is in a talking state during operation; If it is determined that the communication state is established in the step, the operation control unit compares the communication duration with the reference value of the duration stored in the memory unit. If the duration is less than the reference value, the operation returns to step S2, A step (S6) of generating a first control signal for alerting the cellular phone and the vehicle during a driving operation; If no voice signal is detected in step S5, the number of scrolls is sensed from the touch screen. If there is no detected signal, the user repeats his / her own steps. If the number of scrolls is greater than the reference value (S7) when the number of times exceeds a predetermined number of times, the mobile phone user judges to be an Internet search or a character reading state; If it is determined in the step that the character is in the character reading state, the operation control unit compares the duration of the scroll operation with the reference value of the duration stored in the memory unit. If the duration is not more than the reference value, A step (S8) of generating a second control signal for warning a character reading action during operation of the mobile phone and the vehicle when the time is exceeded; If the voice signal is not detected in step S5, the operation control unit detects the number of multi-touches from the touch screen, compares the detected number of times with the reference value for the number of multi-touches stored in the memory unit, (Step S9); if it exceeds the predetermined number of times, it is determined to be a character writing state (step S9); If it is judged in the step S9 that the character is written, the operation control unit compares the duration of the multi-touch operation with the reference value of the duration stored in the memory unit. If the duration is not more than the reference value, (S10) of generating a third control signal for warning a character writing operation during operation of the mobile phone and the vehicle when the duration is exceeded. In the steps (S6, S8, and S10) 2 and the third control signal are generated, the arithmetic and control unit causes the vibration alarm sound to be generated at regular intervals by using the vibration element or the speaker mounted on the mobile phone, and makes the emergency flasher of the vehicle operate.

According to another embodiment of the present invention, when the first, second, and third control signals are generated in steps S6, S8, and S10, The information about the real-time position and the moving speed detected from the GPS receiver is compared with the information on the driving speed and the speed limit on the corresponding road stored in the geographical information DB and a warning corresponding to the driving information on the corresponding road And generates a warning screen at the same time as a voice message indicating that the speed limit is exceeded when the moving speed exceeds the speed limit, and gradually decelerates the vehicle. In step S10, A warning message is generated at the same time as a voice message indicating that the speed limit is exceeded when the moving speed detected by the moving speed exceeds the limit speed, The step of pausing forced to work further included.

According to the present invention, unlike the conventional method in which a separate camera is installed inside a vehicle to photograph a driver, the use of a mobile phone can be immediately detected during operation using a camera mounted on the mobile phone, , Selectively detects the received text message reading or text writing state, and causes the cellular phone and the vehicle to generate a warning signal corresponding to each state step by step, thereby preventing a traffic accident Can be prevented in advance.

The present invention can also be implemented by appropriately changing or adding software without adding a separate hardware configuration by using various functional elements mounted on a mobile phone such as a GPS receiver, a camera, a screen screen, a speaker, There is an advantage that the effect of the present invention can be achieved at a low cost.

1 is a block diagram for explaining the overall configuration of a mobile phone use warning device during driving of a vehicle using a mobile phone according to the present invention.
FIG. 2 is a block diagram for explaining a detailed configuration of the mobile phone use warning device during vehicle operation of FIG. 1;
3 is a block diagram for explaining a configuration of an operation control unit of the mobile phone use warning device during vehicle driving according to the present invention shown in FIG.
4 is a diagram for explaining a pattern recognition process for object recognition according to the present invention.
FIG. 5 is a view for explaining an object recognition algorithm of the LBP scheme used according to an embodiment of the present invention.
FIG. 6 is a diagram for explaining an MCT-based object recognition algorithm used in accordance with another embodiment of the present invention.
7 is a view for explaining a warning screen of a mobile phone use warning device during driving of a vehicle according to an embodiment of the present invention.
8 is a flowchart illustrating a method of controlling a mobile phone use warning device during driving of a vehicle according to an embodiment of the present invention.

Hereinafter, a mobile phone use warning device and a control method thereof during operation of a vehicle using a mobile phone according to the present invention will be described with reference to the accompanying drawings.

1 is a view for explaining the entire operation of the mobile phone use warning device 100 during driving of a vehicle according to the present invention. (10) and vehicle control means (20) mounted on the vehicle. 1B is a block diagram schematically showing the overall configuration of the detection / warning means 10 and the vehicle control means 20. As shown in FIG.

First, the configuration of the detection / warning means 10 will be described. The detection device 11 detects the position of the user in real time using a GPS (Global Positioning System: not shown) receiver mounted in the commercial mobile phone 30 Detects a frontal image using a camera (not shown) mounted on the cellular phone during operation of the cellular phone during operation, detects a voice signal during a call using a microphone (not shown) mounted on the cellular phone, (Not shown) to detect the operation of the mobile phone, such as writing or reading a text message.

The memory unit 13 stores various reference values for determining the use of a mobile phone during operation, information about analysis algorithms for object recognition, and feature values of object candidates required for analysis. The geographical information DB 17 Database) contains information on vehicle speed limits and safe driving on roads and areas.

The operation control unit 12 compares the detected information with the information stored in the memory unit 13 in accordance with the signal detected by the sensing device 11 to determine whether the user is driving the vehicle directly, And judges whether the user is in a call state, whether the user is reading the Internet search or received text message, or is writing a call character according to the determination result, And transmits control signals corresponding to the respective determination states to the vehicle control means 20 through the signal driving unit 14 and the wireless transmission device 16, respectively. In addition, it compares the information about the real-time position and the moving speed detected from the GPS receiver with the driving information on the corresponding road stored in the geographic information DB 17, and outputs an identification signal according to the comparison result to the signal driver 14 and the vehicle control (20), respectively.

The signal driving unit 14 is configured to perform a driving operation corresponding to an identification signal on the road in accordance with a control signal according to each determination result of the operation control unit 12 and a comparison result with the GPS receiver and the geographic information DB 17 And transmits it to the warning means 15.

The warning means 15 may generate a vibration alarm sound or warning voice message by a vibration element or a speaker (not shown) mounted on the mobile phone, Causing a warning screen and forcibly stopping the operation of the mobile phone.

The vehicle control means 20 includes a radio receiving apparatus 21, a vehicle control unit 22, an emergency blinking light 23 and a vehicle speed reducer 24, The emergency decelerator 23 and the vehicle decelerator 24 are selectively activated when the control signal and the identification signal corresponding to the respective determination states received from the vehicle control unit 22 are received by the vehicle control unit 22.

FIG. 2 is a detailed block diagram of a mobile phone use warning device during vehicle driving according to the present invention shown in FIG. 1, and the operation of the present invention will be described in detail.

First, the sensing device 11 of the sensing / warning means 10 includes a GPS sensing unit 11a for measuring the position of the user by the GPS receiver of the cellular phone, an image sensing unit 11a for sensing a front- A voice detection unit 11c for detecting a voice call signal by a user by means of a microphone of the mobile phone, and a voice recognition unit 11c for controlling the touch screen of the mobile phone in order to check or input the received text message, And a screen detection unit 11d for detecting that the image is to be displayed.

First, the GPS sensing unit 11a detects the position of the user in real time using the GPS receiver even when the mobile phone is not operating. For reference, briefly explaining the principle of GPS, a satellite that is orbiting the earth is arranged to receive radio waves from four satellites in any part of the earth. If there is a radio receiver, the radio waves received from these satellites are analyzed In recent years, unlike conventional GPS, which has a positioning error of about 20m, DGPS (Differential Global Positioning System) or CDGPS Carrier Differential Global Positioning System (GPS) have been developed to enable more sophisticated positioning of civilian GPS receivers.

When the user operates the mobile phone according to the incoming and outgoing signals, the image sensing unit 11b simultaneously operates the camera to photograph an image of the front face of the mobile phone. At this time, the camera will be described later, If it is determined that the average moving speed per unit time measured in real time from the GS sensing unit 11a is not being moved by the GS sensing unit 11a and it is determined that the mobile unit is moving by the moving means such as a vehicle, When the front face is photographed and the feature of the photographed front image is recognized as an image of a part of the car handle, it is determined that the user is using the cellular phone while driving. The detection of the GPS signal and the image image together by the GPS sensing unit 11a and the image sensing unit 11b means that even if the value measured by the GPS sensing unit 11a is determined to be moving by the vehicle or the like, In fact, since the mobile phone user himself may not be driving directly but may be moving to the public transportation means, it is determined whether the person himself / herself is driving according to the detection result of the image sensing unit 11b.

The voice sensing unit 11c senses a voice call signal by the microphone of the mobile phone. The screen sensing unit 11d operates the touch screen of the mobile phone to check the Internet, input the received text message, Hereinafter, the detection method in the screen sensing unit 11d will be described in detail.

Prior to the description, the touch screen is widely used as a monitor input device for a variety of household appliances such as a mobile phone, a navigation device, and a refrigerator, a washing machine, etc. The touch screen is largely classified into a resistance film type, an infrared (IR) , Ultrasonic method, and electrostatic capacity method. In the embodiment of the present invention, a capacitive method widely used in a cellular phone is employed. A capacitance method is briefly described. A transparent electrode (film) and a liquid crystal are attached to a screen surface to apply a voltage. When a human hand of a conductor is brought into contact with an electrode, a parasitic capacitance ), And operates on the principle that the character or symbol corresponding to the virtual keyboard on the screen is displayed according to the coordinate value at the contact position.

On the basis of the above principle, in the case of checking the Internet search or received text message first, a drag operation of moving a finger from one point of the touch screen of the mobile phone to another point to stop the finger, , The content is confirmed by a scroll operation in the up and down and left and right directions by a flick operation in which the user touches one point of the touch screen and quickly moves in any direction while releasing the touch. For example, if a simple internet inquiry or received text message is short, the message window can be opened and the contents can be confirmed by a one or two scrolling operation. If the contents are long, And the contents are checked. On the other hand, when a text message is to be input for a call, a character or a symbol of a virtual keyboard displayed on the screen with a finger must be inputted by multi-touching several times. As described above, the checking of the Internet search and the received text message can be performed by a scroll operation such as dragging and flicking several times in one or two times depending on the contents. On the other hand, Touch operation is repeated at least ten to several times. Therefore, in the present invention, an operation signal using the principle of the touch screen, i.e., the number of touches in which the capacitance is changed, is detected to determine whether the text message is read, And determines whether the mobile phone is operated or not.

3, the operation control section 12 includes a movement judging section 12a for judging whether the user is moving to the vehicle, a driving judging section 12b for judging whether the vehicle is directly driving, An operation judging unit 12d for judging whether or not the touch screen of the mobile phone is operated, a duration time determining unit 12e for judging the duration of the call or the operation of the mobile phone, And a travel information determining section 12f for determining whether the safe running information or the speed limit is exceeded. The operation determination unit 12d further includes a character reading determination unit 12R for sensing a scroll operation signal of the touch screen and determining whether to read the received text message, a character determining unit Writing determination unit 12W. The operation control unit 12 configured as described above performs a control operation according to the signals sensed by the sensing units of the sensing device 11 described above and outputs control signals and identification signals according to the determined results from the determining units Hereinafter, the control operation will be described in detail.

The memory unit 13 stores various reference values and analysis algorithms necessary for the control operation of the operation control unit 12. The memory unit 13 stores information on the position of the mobile phone received in real time from the GPS detection unit 11a For example, a reference value for determining that the average speed arithmetically averaged at intervals of 5 minutes exceeds 4 km / hour to be moved to the vehicle, a reference potential value for a voice frequency generally known in the band of 200 to 3,500 Hz, For example, a reference value for Internet search or character reading, a reference value for character writing, a number of times of multi-touch exceeding 10 times, For example, a reference value for the determination of the duration of 20 seconds or less, and a feature value for the candidate group feature of the car handle, as will be described later, Etc., etc. required analysis algorithm to recognize objects are stored.

Therefore, the movement judging section 12a of the arithmetic and control section 12 measures the position detected by the GPS detecting section 11a at intervals of 5 minutes per unit time and outputs the arithmetic average value to the reference It is determined that the user is moving in a predetermined radius or on foot while the vehicle speed is less than or equal to 4 km / hour. If the vehicle speed is greater than 4 km / hour, it is determined that the user is moving to the vehicle. For reference, in the embodiment of the present invention, when the unit time is set to a short time of 1 to 2 minutes, the unit time is set to an interval of 5 minutes. If the vehicle is temporarily stopped at a traffic light stand- The average walking speed of a typical adult is about 4 km / hour, and the walking distance is calculated based on the measured walking distance, It is to be understood that the present invention is not limited to the above-described embodiments.

When the average value of the movement distance per unit time measured by the GPS sensing unit 11a is less than 4 km / hour by the movement determination unit 12a of the operation control unit 12, it is determined that the user stays within a predetermined radius, If the detection unit 11b does not operate the camera, if it exceeds 4km / hour, it is determined that the vehicle is moving by the vehicle, and the image sensing unit 11b operates the camera simultaneously with the operation of the mobile phone, . Here, the front image refers to an image of a direction of a camera behind the mobile phone.

When the image of the front face of the mobile phone is photographed according to the operation of the image sensing unit 11b, the operation determination unit 12b of the operation control unit 12 determines the feature value of the candidate of the car handle stored in the memory unit 13, Recognizes that the image taken on the front surface of the mobile phone belongs to the characteristic of the pattern unique to the car handle using the object recognition analysis algorithm, and determines that the mobile phone user is operating.

In order to describe a method of recognizing the use of a mobile phone while driving by using the object recognition algorithm, characteristics of an inherent pattern of an automobile handle will be described first. As shown in FIG. 5, one of the most common external features of the automobile handle 40 is that it is a circle having a predetermined diameter. When the automobile handle 40 has a small outer diameter (Ho), it varies from 37 cm to 38 cm , Medium size is 42cm ~ 47cm, and buses are 49cm ~ 51cm in the case of large. One of the other characteristics is that a connected central portion C1 on the steering shaft is provided with a pair of horizontal connecting portions C2 connected to the inner diameter side Hi of the handle in the left and right direction and one or two vertical connecting portions C2, (Ra) on the upper side and two or three small space portions (Rb to Rd) on the lower side with respect to the circular handle (40). The reason why the plurality of space portions Ra to Rd are formed is that even if the steering wheel is rotated in the clockwise or counterclockwise direction during operation, the operator can easily confirm the instrument panel or the instrument panel It is to secure space. A logo TM of each manufacturer is generally attached to the center of the center portion C1 and a plurality of function buttons for car turning are attached to one side of the pair of the horizontal and vertical connection portions C2 and C3 And the overall handle color is usually a dark color of black, which is a common feature.

The present invention detects the use of a mobile phone while driving by using the object recognition algorithm, and it can be generalized when the user operates the mobile phone to read or write a text message from the outside while driving, The camera is operated at an arbitrary position spaced about several centimeters above the car handle 40. At this time, a video image in various areas B1 to B9 is taken as shown in FIG. 5 according to the position of the mobile phone. For example, if the cellular phone is placed in the lower left area of the handle and operated, the image of the area B6 will be captured. If the camera is placed on the central area C1 and operated, Will be shot. In other words, any one of the areas B1 to B9 will be photographed according to whether the user is a left-handed person or a right-handed person or according to a custom.

Therefore, if the mobile determining unit 12a determines that the mobile phone user is moving to the vehicle, the operation determining unit 12b operates the camera of the image sensing unit 11b simultaneously with the cellular phone operation The image of the front face is immediately photographed, and the characteristic values of the inherent pattern of the automobile handle candidate group in the multiple areas B1 to B9 of the handles stored in the memory unit 12 and the characteristic values of the well known LBP (Local Binary Pattern ) Or MCT (Modified Census Transform) to recognize whether the feature of the image of the front of the mobile phone belongs to a characteristic feature of the vehicle handle, thereby determining that the user of the mobile phone is directly driving the mobile phone . 5, the area of the handles photographed by the camera is shown as nine areas B1 to B9, but this is for illustrative purposes only. Actually, a larger number of candidate regions are subdivided so that the feature values therefrom are stored in the memory portion 13 Of course).

Hereinafter, in order to help understand the principle of the object recognition algorithm used in the method of recognizing the feature of the automobile handle according to the present invention, the pattern recognition for object recognition will be briefly described.

First, pattern recognition belongs to the fields of cognitive science and artificial intelligence. Cognitive science uses computer, artificial intelligence, neurobiology, linguistics, etc. Artificial intelligence is a computer program that implements human learning ability and reasoning ability artificially to understand specific objects or understand syntactic patterns such as language. Here, the pattern means a feature, and the feature can be defined as a set of features constituting a pattern. Pattern recognition is a process of collecting and analyzing specific feature data, and these data are called a class. A common set of attributes or features that occur on the same data is called a set of patterns. The important thing here is the classification task that assigns a specific object to a specific class. Using the feature vector thus formed, a template for a target pattern to be compared is created and a pattern to be recognized is subjected to a normalization process according to a template configuration condition, and a correlation or a distance is measured to find a similarity, And performs recognition. A normalized template describing this category is called a training process. It is widely used as a pattern recognition method because of its simple algorithm and fast calculation speed.

As shown in FIGS. 4A and 4B, pattern recognition is generally divided into four steps. The first step is the data collection step of FIG. 4A. In this step, The data must be accurately determined to obtain the results. The second step is a feature selection step of FIG. 4B, which is a very important step in the pattern recognition process in which a target pattern is sufficiently analyzed to determine which feature is required. If the feature is determined, the algorithm for constructing the model is selected in the model selection step of FIG. 4c. In the learning step of FIG. 4d, the feature set extracted from the collected data is utilized to recognize To complete the complete model.

In general, pattern recognition methods for recognizing facial and other objects include techniques using an Adaboost classifier based on Haar-like features extraction, Principal Component Analysis (PCA), and Elastic Bunch Graph (EBGM) Matching) or LDA (Linear Discriminant Analysis) are widely used. However, these algorithms are sensitive to the contrast of illumination and bending of objects, so there is a limitation in stable feature extraction. Therefore, the present invention adopts the LBP or MCT algorithm which can perform stable pattern recognition while being less sensitive to external factors such as illumination change or noise compared to other algorithms. The LBP or MCT algorithm is a well-known technique for recognizing various objects such as a face. In order to facilitate understanding of the present invention, these principles will be briefly described with reference to FIG. 5 to FIG.

First, referring to FIG. 5, the LBP-based object recognition algorithm used in accordance with an embodiment of the present invention will be described. LBP is first recognized by academia in 1996 by Ojala et al. [Non-Patent Document 1] Ojala proposed a method of extending existing 3 x 3 LBPs into arbitrary-sized circular shapes [Non-Patent Document 2]. In 1996, the operator was introduced to measure the values of 8 neighbors using the value of the central pixel. The basic idea is to divide the face image into small blocks and to obtain LBP histograms. And the distance between the histograms was used to distinguish the facial images. Since facial images can be viewed as a composition of fine textures, LBP can be used to describe the texture of facial features. That is, it is possible to extract functional vectors that describe facial features by dividing facial images.

Therefore, since the LBP operates on a pattern, it can efficiently create an image vector and remove the offset of the pixel value, so that it can efficiently operate on the illumination change of the image and the rotation of the image. The LBP operator It is possible to describe the gray-scale image in textured form. Therefore, LBP is widely used in many fields related to object recognition besides face recognition because it is a very powerful method for explaining the shape of digital image and is suitable for feature extraction. In other words, it is used to measure the similarity between images by dividing the image of the object into small regions, extracting a single feature vector in histogram form of the divided image, expressing the objects efficiently based on the single feature vector.

The LBP operator proposed by Ojala et al. Specifies the label of each pixel of the image considering the value of the center pixel and the binaries of the 3 × 3 boundary, and performs the operation in the clockwise or counterclockwise direction using the center value You can get a value of 0 or 1 in one label. If you have a larger value around it than the center value, you should assign 1 to it. Therefore, in the present invention, by recognizing the characteristics of the handle by using the above-mentioned principle, for example, the area B6 of FIG. 5 is photographed by the camera and the handle 40 of the middle left- (B) to 5 (d) and 5 (c) are obtained by coding the relative brightness change of the surrounding 3x3 region of each pixel as a value calculated for all the pixels of the image, As shown in the figure, when 1 is brighter than the center pixel and 0 when it is darker, 0 is used as an index value for the local region.

The calculation method for the label of each pixel in the image, the center value is represented by (x _c, y _c) comparing the concentration between the surrounding pixel (x _c, y _c) to extract a binary set. The decimal value of the 8-bit decimal value is represented by a decimal label.

L _c is the value on the gray (Gray Value) of the center pixel, and l _n represent the gray values of the eight surrounding neighbor pixels, for when defined as S (x) as a function the same as the equation (2).

As described above, according to the embodiment of the present invention, histograms are obtained for the index values calculated for the respective pixels, and then the histogram is used as a model of the feature for the corresponding image region. Among them, the LBP feature To recognize the handles using a learning algorithm such as AdaBoost.

For reference, a multi-scale block LBP method called MB-LBP [Non-Patent Document 3] was published by S. Liao et al. In 2007 because the original LBP extracts local characteristics too by using 3x3 pixels The LBP in OpenCV is based on MB-LBP [Non-Patent Document 4]. That is, the MB-LBP is a method of dividing an image area of an arbitrary size into 3 × 3 blocks and calculating an LBP index for each block average. If LBP is used instead of a pixel unit, And MB-LBP is calculated for different scales and image positions, it is possible to extract various LBP features for a single object. This is also known as an ADA for selecting LBP features Boost's automated learning algorithm is used.

FIG. 6 is a view for explaining an MCT-based object recognition algorithm used in accordance with another embodiment of the present invention. The MCT is a representative method of the object recognition field from the face detection field to the car license plate recognition field requiring precision, It has excellent performance as one of. MCT is an abbreviation of Modified Census Transform, originally based on the principle of Census Transform (CT) proposed by Zabih et al in 1994 [Non-Patent Document 5]. CT is a value obtained by encoding the brightness change of one pixel per pixel in the image to 0 when it is brighter than the center pixel and 1 when it is darker and connecting the result to a bit string. Therefore, if the peripheral region is assumed to be a 3x3 region, the CT result for a certain pixel is the same concept as the LBP described above, and the basic principle is the same. In the case of LBP, the histogram of the LBP result values is used as a texture feature for the image area. And CT was used to convert the input image into an image with no effect on the ambient brightness change. There is also a slight difference in the matching method. In LBP, the similarity between histograms is measured. In CT, the Hamming distance of the CT bit string between the corresponding pixels is used.

CT has since been extended to MCT by B. Froba et al. In 2004 [Non-patent Document 6], MCT encodes the difference in brightness from the average of the local region instead of encoding the difference in brightness from the central pixel By providing a value of 0 or 1 to the center pixel, it is possible to express a local binary pattern more abundant than the conventional CT. That is, the MCT is a method of integrating and representing a pattern of a local region of an image, which is robust against illumination change and has a high memory efficiency [Non-Patent Document 7]. In 2011, T. Song et al. Proposed a license plate recognition system by combining license plate detection algorithm and super resolution algorithm using the advantages of MCT [Non-Patent Document 8].

In the embodiment of the present invention, by using the MCT feature extraction method of the above-described principle, the correlation value of the pixel x at the specific position with the neighboring pixels characteristic of the MCT is expressed as a pattern without using the brightness value itself. The method of obtaining the MCT feature using the kernel is expressed by the following equation (3).

Where N'x is the set of adjacent pixels in the kernel, I (x) is the brightness of X, and I (x) is the average brightness of the pixels in the kernel. ζ (I (x), I (y)) is a comparison function that outputs 1 if I (y) is greater than I (x), 0 otherwise. Ⓧ converts a 9-digit binary array resulting from ζ (I (x), I (y)) into a decimal number as a decimal conversion operator.

In the present invention, the area B4 of FIG. 6 is photographed by the camera in the same manner as the LBP system of FIG. 5 by using the above principle, so that the handle 40 of the right lower middle region as shown in FIG. 6 (a) If an image of a part of the box 25 is obtained, the average brightness of nine surrounding pixels including itself at a specific pixel position X in the image is obtained, and the average brightness at each position as shown in FIG. 6 (b) Then, as shown in FIG. 6 (c), a pixel having an average brightness is converted to 1, otherwise, a binary value array is converted to a decimal number to extract a feature value at a pixel position X. FIG. In order to recognize the image, the feature of the steering wheel is detected through the known Adaboost training method. Since the feature values of the MCT have discrete characteristics, they can be obtained according to the mathematical design of the classifier using a lookup table do. Generally, the Adaboost training method can use a known method of creating a strong classifier that combines a plurality of weak classifiers with low classification performance proposed by Y. Freund et al. In 1997 [Non-Patent Document 9].

In addition, since the introduction of the first AdaBoost, several improved versions of the AdaBoost method have been proposed. In 1999, the Real AdaBoost (Real) method of optimizing the training method through Exponential Criterion Minimization by RE Schapire et al. Adaboost [10], and Gentle Adaboost by J. Friedman et al. [11]. These methods can also be used in the present invention, and these methods are well known and will not be described herein.

Using the analysis algorithm of the LBP or MCT-based AdaBoost as described with reference to FIGS. 5 and 6, the feature value of a plurality of handle candidate groups stored in the memory unit 13 is compared with the feature value of the handle candidate group stored in the memory unit 13 When the photographed image is recognized as a handle, the arithmetic and control unit 12 finally determines that the user uses the mobile phone while driving.

2 and 3, if it is determined that the user is using the mobile phone while driving, the communication determination unit 12c of the arithmetic and control unit 12 detects the communication sound signal by the voice sensing unit 11c , It is judged to be in the busy state if it belongs to the reference potential value of the voice frequency of 200 to 3,500 Hz band stored in the memory unit 13 and the communication state is maintained by the signboard unit 12e in the memory unit 13 A first control signal for warning a call during operation is transmitted to the signal driving unit 14 and to the vehicle control means 20 via the wireless transmission device 16 . The travel information determining unit 12f compares the information about the position and the traveling speed detected from the GPS receiver with the travel information of the corresponding road stored in the geographic information DB 17 and outputs an identification signal for the comparison result to the signal driver 14 and the vehicle control means 20, respectively.

However, if a voice is not detected from the voice sensing unit 11c and an operation signal of the touch screen of the mobile phone is sensed from the screen sensing unit 11d, the operation determination unit 12d of the operation control unit 12 determines that the number of scrolling operations is 10 If the number of multi-touches exceeds 10, it is determined to be in the state of writing a character to be sent. Then, the operation of the operation control unit 12 is judged to be the state of reading the text message, ), A second control signal for warning a character reading state during operation and a third control signal for warning a character writing state are supplied to the signal driver 14 To the vehicle control means 20 via the radio transmission device 16, respectively. Similarly, the travel information determining unit 12f compares the GPS receiver with the information stored in the geographic information DB 17, and transmits the corresponding identification signal to the signal driver 14 and the vehicle controller 20.

The signal driving unit 14 receives the first, second, and third control signals from the operation control unit 12 and the navigation information determination unit 12f by the GPS receiver and the geographic information DB 17 Generates a drive signal corresponding to the obtained identification signal, and drives the warning means (15) of the mobile phone.

The warning means 15 includes a warning sound generating device 15a constituted by a vibration element or a speaker mounted on the mobile phone, a warning screen generating device 15b of the mobile phone and an operation cut-off device 15c for forcibly stopping the operation of the mobile phone .

Accordingly, when the driving signals corresponding to the first, second, and third control signals for warning a busy state, a character reading state, or a character writing state during operation are input from the signal driver 14, the warning sound generator 15a Generates a vibration warning sound at regular intervals and, when the driving signal corresponding to the identification signal obtained by the GPS receiver and the geographic information DB 17 is input by the driving information determination unit 12f, the warning sound generating unit 15a The warning screen generating device 15b and the warning screen generating device 15b generate warnings for the corresponding areas such as the "child protection zone" or the "accident bunch zone" as shown in FIGS. 7A and 7B, according to the road information stored in the geographic information DB 17 If the moving speed exceeds the limit speed by providing the screen with a voice message or comparing the real-time moving speed from the GPS receiver with the limiting speed at the corresponding road stored in the geographic information DB 17, Indicated generate together was a warning screen exceeds the "speed limit", as the voice message. When the drive signal corresponding to the identification signal whose moving speed of the mobile phone user exceeds the limit speed is inputted by the driving information determining unit 12f is continued while the character writing state continues during the operation, Forcibly stops the operation of the motor. In order to maximize the visual effect to the driver, the warning screen causes the screen to blink at regular intervals. In addition, the warning sound generating device 15a for generating the vibration warning sound may gradually increase the intensity of the warning sound depending on the degree of danger for each action such as a call state, a character reading state, or a character writing state.

On the other hand, when the first, second, and third control signals from the arithmetic and control unit 12 are inputted, the vehicle control unit 22 operates the emergency blinker 23 to give a warning signal to the following vehicle, When the identification signal with the moving speed of the user exceeding the limit speed is inputted by the judging unit 12f, the vehicle is decelerated by the vehicle speed reducer 24 step by step, Prevent collision accidents caused by Here, since the state coping ability and the front main rate decrease gradually as the state of the call, the character reading state, or the character writing state decreases, the vehicle speed reducer 24 can gradually increase the deceleration rate of the vehicle according to each state.

In the above example, if the reference value of the duration for the busy state, the character reading state, or the character writing state is set to 20 seconds, for example, if a call comes during driving, It is a minimum time considering that it takes about 15 ~ 20 seconds to get the understanding that it is time to call again. The call state, the character reading state, and the character writing state also become more dangerous depending on the state, and therefore, the reference value of the duration can be set shorter in the actual implementation.

The radio transmitting device 16 for transmitting a control signal according to the determination result of the arithmetic and control unit 12 to the vehicle control means 30 is a radio communication device such as Bluetooth or ZigBee, And the wireless receiving device 21 in the vehicle control means 30 is also configured with a wireless communication module adapted to receive the wireless signals. The vehicle control unit 22 is integrally formed with an ECU (Electronic Control Unit) for controlling the engine of the vehicle, the automatic transmission, and the driving system, the braking system, and the steering system, It is of course possible to program the emergency flasher 23 and the vehicle speed reducer 24 to appropriately control according to the determination result.

8 is a flow chart for explaining the control operation of the operation control unit of the mobile phone use warning device during vehicle driving according to the embodiment of the present invention, and FIG. 2, FIG. 3, and FIG. 5 to FIG. 7 Will be described in detail.

First, in step S001 with the start of the mobile phone use warning device of the present invention, a real time position signal of the user's mobile phone is always received by the GPS receiver without operating the mobile phone. Next, in step S002, when the mobile phone is operated for the purpose of confirming or transmitting a call, an Internet search, or a text message, in step S003, the movement determining unit 12a of the arithmetic and control unit 12 determines, For example, the moving distance to the position of the mobile phone per unit time at intervals of 5 minutes is measured by the unit 11a, and the arithmetic average is compared with the reference value stored in the memory unit 13, and the comparison result is 4 km / hour It is determined that the user is within the predetermined radius and the process returns to step S002. If the answer is Yes , it is determined that the user is moving to the vehicle, and the process proceeds to the next step. In step S004, the operation determination unit 12b of the operation control unit 12 photographs the front face of the camera of the image sensing unit 11b simultaneously with the operation of the cellular phone, It is determined whether the feature of the image captured on the front surface of the mobile phone belongs to the characteristic feature of the vehicle handle using the feature value and the automatic learning algorithm of the LBP or MCT described in Figs. 5 and 6, the returns to step (S002) determines that no direct driving, then it is determined to be using the case of Yes, which recognized as belonging to the distinctive features of the steering wheel, mobile phone of the user and direct operation at step (S005) .

In step S006, when a voice signal is detected by the microphone of the voice detection unit 11c of the mobile phone, the call determination unit 12c of the operation control unit 12 determines that the detected value is 200 to 3,500 Hz band. If YES in step S007, it is determined in step S007 that the user is in a busy state during operation, and the process proceeds to the next step. In step S008, it is determined whether the call state exceeds 20 seconds by the dura- tion timer 12e of the arithmetic and control unit 12, and if not, it is determined that the call is terminated and the process returns to step S002 return and, in the case of Yes at which the call state does not go to each deliver a first control signal for warning the call behavior of the driver on the mobile phone of the warning means 15 and the vehicle control means 20, the warning means 15 A vibration alarm sound is generated at a constant interval by a warning sound generating device 15a such as a vibration element or a speaker of the vehicle control means 20 and an emergency blinking light 23 of the vehicle control means 20 is operated to send a warning flashing signal to the following vehicle, . In step S009, the traveling information determining unit 12f compares the information about the real-time position and the traveling speed detected from the GPS receiver with the traveling information and the traveling speed of the corresponding road stored in the geographic information DB 17, The warning screen as shown in FIGS. 7A and 7B corresponding to the running information of the road is generated by the warning sound generating device 15a and the warning screen device 15a together with the voice message in accordance with the comparison result The control unit 15 determines the speed limit on the road and, if it does not exceed the limit speed, it repeats its own step. If the speed limit is exceeded, the warning sound generator 15a and the warning screen unit 15a And causes the vehicle to decelerate the vehicle by the vehicle speed reducer 24 of the vehicle control means 20 at the same time as generating the warning screen with the voice message.

If it is determined in step S006 that the voice signal is not detected by the voice sensing unit 11c, the process proceeds to the next step and the character reading determination unit 12R of the operation determination unit 12d of the operation control unit 12 ) And the character writing determination unit 12W determine whether the user is searching the Internet or reading the received character according to whether or not the touch screen is operated from the screen sensing unit 11d, .

First, in step S010, the character reading determination unit 12R detects the number of scrolling operation signals of the touch screen from the screen sensing unit 11d. If there is no detected signal, repeating (S010) again, and, in the step when the If the number of scroll motion signal, compared to the reference value for the character reading state decision scroll count designation stored in the memory unit 13, for example less than 10 Yes ( S011), it is determined that the user is in the Internet search or character reading state, and the process proceeds to the next step. In step S012, it is determined whether the duration of the scroll operation signal is continued for more than 20 seconds, for example, by the duration timer unit 12e of the operation control unit 12. If the duration is not longer It is determined that the user has stopped reading the character and the operation returns to step S002. If the determination is Yes , if it is longer than 20 seconds, the warning means 15 informs the user of the second control signal To the vehicle control means 20 so that the alarm sound generator 15a of the warning means 15 generates a vibration alarm sound at regular intervals and the emergency flasher 23 of the vehicle control means 20 is operated It sends a warning flashing signal to the following vehicle and proceeds to the next step. In step S013, similarly to step S009, the travel information determining unit 12f compares the travel speed information and the speed limit on the corresponding roads of the GPS receiver and the geographical information DB 17 with each other, 7A and 7B corresponding to the running information of the road are generated together with the voice message by the warning sound generating device 15a and the warning screen device 15b, If the speed limit is not exceeded, the control unit 20 repeats its own step. If the speed limit is exceeded, the warning screen as shown in FIG. The vehicle decelerator 24 decelerates the vehicle.

Next, in step S014, the character writing determination unit 12R detects the number of multi-touch operation signals of the touch screen from the screen sensing unit 11d, and determines whether the number of detected signals is 10 If time more than the repeat step (S014) again, and the Yes that If the number of the multi-touch signal exceeds 10 times, as compared with the reference value for the character writing state determining scroll count designation stored in the memory 13 In step S015, it is determined that the user is writing a character to be printed, and the process proceeds to the next step. In step S016, as in the character reading step, it is determined whether the duration of the scroll operation signal continues for more than 20 seconds by the duration timer section 12e of the operation control section 12, It is determined that the user has stopped writing the character and the operation returns to the step S002. If the answer is Yes , the third control signal for warning the character writing operation during the operation is outputted to the warning means 15 The warning sound generator 15a of the warning means 15 generates a vibration alarm sound at a predetermined interval and the emergency flasher 23 of the vehicle control means 20 is operated It sends a warning flashing signal to the following vehicle and proceeds to the next step. In step S017, the travel information determining unit 12f compares the travel information and the speed limit on the corresponding roads of the GPS receiver and the geographic information DB 17 with each other, as in the above-described steps S009 and S0013) , A warning screen as shown in Figs. 7A and 7B corresponding to the driving information of the road is generated together with the voice message by the warning sound generating device 15a and the warning screen device 15b according to the comparison result Determines the limit speed on the road, and repeats its own step if it does not exceed the limit speed, and gradually decelerates the vehicle by the vehicle speed reducer (24) of the vehicle control means (20) The warning screen as shown in FIG. 7C is generated together with the voice message, and then the cellular phone operation forcibly is temporarily stopped by the power cutoff device 15c of the cellular phone.

As described above, according to the present invention, it is possible to detect and warn a call or a character reading or writing operation while driving through software addition or modification without changing hardware by using various sensing functions mounted on existing mobile phones, The object and effect of the present invention can be achieved.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, but, on the contrary, It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention.

100: Mobile phone use warning device 10: Detection and warning means
11: sensing device 12:
13: memory part 14: signal driving part
15: Warning device 16: Radio transmitting device
20: vehicle control means 21: wireless receiving device
22: vehicle control unit 23: emergency blinking light
24: vehicle speed reducer 30: mobile phone

Claims (8)

A vehicle driving system using a mobile phone including a GPS receiver, a camera, a microphone, a touch screen, a vibration device, a speaker, a memory unit, a navigation app provided with a geographic information DB, and an operation control unit In a mobile phone use warning device,
Detecting a position of the user using the GPS receiver, detecting an image on the front face when the mobile phone is operated using the camera, detecting a voice signal in the communication using the microphone, reading the received character using a touch screen, A sensing device for sensing a credit character writing signal;
A reference value necessary for judging whether the vehicle is moving to a vehicle according to a signal detected from the sensing device, whether a call is being made during operation or whether the touch screen is being operated, an automated analysis algorithm for object identification, A memory unit for storing a value;
A geographic information DB storing the speed limit or safe driving information of the vehicle to the road or the area;
Each of the signals detected from the sensing device is compared with the reference value, the analysis algorithm for object recognition, and the feature value of the object candidate group stored in the memory unit to determine whether the user is driving, And determines whether or not the cellular phone call or the operation state continues for a predetermined period of time and determines whether the user is in a busy state, a received character reading state, or a leaving character writing state according to the determination result And outputs the control signal corresponding to each state, and compares the information about the real-time position and the moving speed detected by the GPS receiver with the information stored in the geographic information DB, An arithmetic control unit;
A signal driver for outputting a corresponding drive signal in accordance with each of the control signal and the identification signal according to the determination result of the operation control unit;
When receiving a driving signal corresponding to a control signal in a call, character reading or text writing state during operation from the signal driving unit, generates a vibration alarm sound, and, in response to a driving signal corresponding to an identification signal obtained by comparing the GPS receiver and the geographic information DB Warning means for temporarily stopping a warning voice message, warning screen or cellular phone operation on the cellular phone;
Wireless transmission means for wirelessly transmitting control signals and identification signals output from the operation control unit; And
Vehicle control means for activating the blinking signal of the vehicle upon receipt of the control signal from the operation control section by the wireless transmission means and for decelerating the vehicle upon receipt of the identification signal whose movement speed exceeds the limit speed;
And a warning device for warning a user of the mobile phone while driving the vehicle using the mobile phone. The method according to claim 1,
The method according to claim 1,
The sensing device includes a GPS sensing unit for sensing a user's real-time position by a GPS receiver, an image sensing unit sensing an image of the front side taken by the camera when the mobile phone is operating, a voice sensing unit And a screen sensing unit for detecting operation of the touch screen of the mobile phone for searching the Internet or receiving a received text message or inputting a text message for outgoing;
The memory unit is a reference value for determining whether the mobile phone is used during operation, and is a reference value for a moving distance for determining whether the user is moving to the vehicle, a number of operations of the touch screen for determining a character reading or character writing state A reference value for a voice frequency for determining a call state, a reference value for a duration of a call state, a character read and a character write state;
The arithmetic and control unit may include a movement judging unit for judging whether or not the user moves by comparing an arithmetic average of the real time position detected by the GPS sensor with a reference value for determining the movement distance stored in the memory unit, The camera of the image sensing unit is actuated to capture an image of the front face of the mobile phone when the mobile phone is operating, and the captured image image is processed by the characteristic value of the handle candidate candidate stored in the memory unit and the analysis algorithm, A pattern recognition unit for recognizing whether the mobile phone user belongs to a pattern feature and determining whether the mobile phone user is operating; a voice recognition unit for detecting a voice signal of the mobile phone and comparing the voice signal with a reference potential value for voice frequency stored in the memory unit, From the screen sensing unit, a touch screen An operation judging unit for judging whether or not an operation signal according to a judgment result of the communication judging unit and the operation judging unit lasts for a predetermined time and judges whether the user is in a busy state, And outputs the first, second, and third control signals; and a travel information determination unit that outputs an identification signal corresponding to a comparison result between the GPS receiver and the geographic information DB ;
Wherein the warning means includes a warning sound generating device for generating a vibration alarm sound at a predetermined interval when the driving signal corresponding to each state is inputted from the signal driving portion, a warning screen generating device for generating a warning screen on the touch screen, And an operation cut-off device for forcibly stopping the operation of the cellular phone according to the drive signal according to the identification signal from the information determination section; And
Wherein the vehicle control unit comprises: a vehicle control unit for controlling the vehicle; an emergency blinker for generating a blink signal when the first, second, and third control signals corresponding to the respective states are inputted from the operation control unit; And a vehicle speed reducer that decelerates the vehicle when an identification signal exceeding the speed is input;
And a warning device for using the mobile phone while driving the vehicle using the mobile phone. 3. The method according to claim 1 or 2,
Wherein the object recognition analysis algorithm is one of an LBP (Local Binary Pattern) or an AdaBoost learning algorithm based on a MCT (Modified Census Transform). 3. The method according to claim 1 or 2,
The signal transmitting means between the arithmetic control unit and the vehicle control means is a radio transmitting apparatus and a radio receiving apparatus which use either an infrared (IR) transceiver, Bluetooth for short range communication or ZigBee module A device for warning of the use of a mobile phone while driving a vehicle using a mobile phone. Control of a mobile phone use warning device during driving of a vehicle using a mobile phone equipped with a GPS receiver, a camera, a microphone, a touch screen, a vibration device, a speaker, a memory unit, a navigation app provided with a geographic information DB, In the method,
(S1) of receiving a position signal of a user's cellular phone by the GPS receiver at all times simultaneously with the start of the warning device;
Operating the user to use the mobile phone (S2);
If the mobile phone operates in the step S2, the operation control unit measures the moving distance per unit time by the GPS receiver, compares the arithmetic average value with the reference value of the moving distance stored in the memory unit, If the speed is lower than the predetermined speed, the process returns to step S2. If the speed exceeds the speed, it is determined that the vehicle is moving to the vehicle (S3).
If it is determined that the vehicle is moving to the vehicle, the arithmetic and control unit controls the cellular phone to simultaneously photograph the front face of the vehicle, and uses the characteristic values of the candidate handle group stored in the memory unit and the automatic analysis algorithm for object recognition, If it is determined that the feature of the image captured in front is included in the inherent characteristic of the steering wheel, and if it does not belong to the inherent characteristic of the steering wheel, the process returns to the step S2. Step S4;
If it is determined that the user is directly driving in the step, the operation control unit detects the voice signal by the microphone and determines whether the sensed value is within the reference value range of the voice frequency stored in the memory unit. If not, (S2), and if the voice signal is sensed, determining (S5) that the mobile phone user is in a talking state during operation;
If it is determined that the communication state is established in the step, the operation control unit compares the communication duration with the reference value of the duration stored in the memory unit. If the duration is less than the reference value, the operation returns to step S2, A step (S6) of generating a first control signal for alerting the cellular phone and the vehicle during a driving operation;
If no voice signal is detected in step S5, the number of scrolls is sensed from the touch screen. If there is no detected signal, the user repeats his / her own steps. If the number of scrolls is greater than the reference value (S7) when the number of times exceeds a predetermined number of times, the mobile phone user judges to be an Internet search or a character reading state;
If it is determined in the step that the character is in the character reading state, the operation control unit compares the duration of the scroll operation with the reference value of the duration stored in the memory unit. If the duration is less than the reference value, A step (S8) of generating a second control signal for warning a character reading action during operation of the mobile phone and the vehicle when the time is exceeded;
If the voice signal is not detected in step S5, the operation control unit detects the number of multi-touches from the touch screen, compares the detected number of times with the reference value for the number of multi-touches stored in the memory unit, (Step S9); if it exceeds the predetermined number of times, it is determined to be a character writing state (step S9);
If it is judged in the step S9 that the character is written, the operation control unit compares the duration of the multi-touch operation with the reference value of the duration stored in the memory unit. If the duration is not more than the reference value, (S10) generating a third control signal for warning a character writing operation during operation of the mobile phone and the vehicle when the duration is exceeded;
When the first, second, and third control signals are generated in the steps S6, S8, and S10, the arithmetic and control unit causes the vibration alarm sound to be generated at regular intervals using the vibrator or the speaker mounted on the mobile phone And the emergency flashing lamp of the vehicle is activated. 6. The method of claim 5,
Wherein the object recognition analysis algorithm uses one of an LBP or MCT-based adaboard learning algorithm. The method according to claim 5 or 6,
When the first, second, and third control signals are generated in the steps S6, S8, and S10, information on the real-time position and the movement speed detected from the GPS receiver is transmitted to the navigation device Information and a speed limit, a warning screen corresponding to driving information on the road is generated together with a voice message according to the comparison result, and when the moving speed exceeds the speed limit, the speed limit is exceeded Generates a warning screen at the same time as the voice message and gradually decelerates the vehicle, and if the moving speed detected by the GPS receiver exceeds the limit speed in the step S10, Further comprising the step of forcibly stopping the operation of the mobile phone after the mobile phone is operated. How to control the phone use warning device. 8. The method of claim 7,
In step S3, the reference value for the moving distance stored in the memory unit for determining whether the user is moving to the vehicle is based on 4 km / hour for 5 minutes of unit time, and in step S6, S8, and S10, The reference value for the number of scrolls and the number of multi-touches for judging the reading state and the character writing state is based on 10 times, and the reference potential value for the voice frequency for determining the calling state corresponds to the range of 200 to 3,500 Hz And the reference value for the duration of the call operation, the scroll operation, and the multi-touch operation for determining the call state, the character read state, and the character write state is based on 20 seconds. Control method of mobile phone warning device during operation.

Images