CN107545225B - Method and device for detecting violation behavior of vehicle-mounted driver and electronic equipment - Google Patents

Method and device for detecting violation behavior of vehicle-mounted driver and electronic equipment Download PDF

Info

Publication number
CN107545225B
CN107545225B CN201610540740.3A CN201610540740A CN107545225B CN 107545225 B CN107545225 B CN 107545225B CN 201610540740 A CN201610540740 A CN 201610540740A CN 107545225 B CN107545225 B CN 107545225B
Authority
CN
China
Prior art keywords
vehicle
steering wheel
violation
mounted driver
area
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201610540740.3A
Other languages
Chinese (zh)
Other versions
CN107545225A (en
Inventor
朱丽
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hangzhou Hikvision Digital Technology Co Ltd
Original Assignee
Hangzhou Hikvision Digital Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hangzhou Hikvision Digital Technology Co Ltd filed Critical Hangzhou Hikvision Digital Technology Co Ltd
Publication of CN107545225A publication Critical patent/CN107545225A/en
Application granted granted Critical
Publication of CN107545225B publication Critical patent/CN107545225B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Abstract

The embodiment of the invention discloses a method and a device for detecting violation behaviors of a vehicle-mounted driver and electronic equipment, relates to a traffic violation detection technology, and can improve the accuracy of detection of the violation behaviors of the vehicle-mounted driver. The method for detecting the violation of the vehicle-mounted driver comprises the following steps: recognizing a steering wheel in the shot image; determining an illegal behavior area of the vehicle-mounted driver according to the identified steering wheel; and detecting the violation of the vehicle-mounted driver by using the determined violation area. The device comprises a module for realizing the method. The invention is suitable for detecting the violation of the vehicle-mounted driver of the public transport means.

Description

Method and device for detecting violation behavior of vehicle-mounted driver and electronic equipment
Technical Field
The invention relates to a traffic violation detection technology, in particular to a method and a device for detecting vehicle-mounted driver violation behaviors and electronic equipment.
Background
Safety is the most basic requirement of people for traveling, wherein the primary condition of safe traveling is that a vehicle-mounted driver is required to drive according to relevant regulations of traffic laws, such as fastening a safety belt, not making a call during driving, not smoking in a cab, and the like. Therefore, how to monitor the operation behavior of the vehicle-mounted driver in the driving process and detect whether the vehicle-mounted driver has dangerous driving behaviors such as illegal behaviors becomes an important research hotspot for safe travel.
With the rapid development of computer vision technology, embedded technology, network communication technology and camera technology, the automatic detection of vehicle-mounted driver violation by using an Intelligent Transportation System (ITS) has become a research hotspot in current Intelligent Transportation.
From the above analysis, it is obvious that dangerous driving behaviors are one of the most important reasons for car accidents, and how to effectively prevent the harm caused by the dangerous driving behaviors by using technical means is urgent. The vehicle-mounted driver of the public transport means is taken as a main safety responsible person who takes the public transport means when the vehicle-mounted driver and people go out, the vehicle-mounted driver needs to be restrained, and dangerous behaviors such as not fastening a safety belt, playing a mobile phone, smoking and the like are avoided when the vehicle-mounted driver is driven.
At present, detection for violation behaviors of a vehicle-mounted driver mainly comprises: the safety belt detects, connects and makes a call and detects and smoke and detect, and wherein, the safety belt detects and adopts the Hough transform to carry out sharp detection, specifically as follows: the method comprises the steps of firstly detecting the head position of a vehicle-mounted driver according to a shot video image, expanding according to the head position to obtain an expanded area, determining the area where the safety belt possibly locates in the expanded area, and then detecting the safety belt in the determined area where the safety belt possibly locates by utilizing a probability Hough transformation algorithm.
And detecting the call receiving and making, namely detecting the position of the mobile phone in the shot video image by adopting an Adaboost iterative algorithm on the basis of detecting the position of the head of the vehicle-mounted driver, and judging that the vehicle is in a call receiving and making state when the vehicle-mounted driver is in the vicinity of the ear of the vehicle-mounted driver if the mobile phone is determined. However, in the method, the Adaboost iterative algorithm is adopted to detect the mobile phone, and due to the fact that under the conditions of different illumination, different camera exposure parameters, different phone receiving and making postures of a vehicle-mounted driver, different models of mobile phones and the like, a large number of samples of the mobile phone under different conditions need to be trained on the Adaboost detector, and if the samples are insufficient, the accuracy of a detection result is low.
The smoking detection is similar to the call receiving and making detection, the Adaboost iterative algorithm is adopted to detect the position of the head, and if the cigarette with the strip-shaped characteristic is near the mouth of the vehicle-mounted driver, the vehicle-mounted driver is judged to be in a smoking state. The method also has the technical problem of low accuracy of detection results.
When the method is applied to a public transport means scene, if the passenger flow is large and the number of passengers is large, a plurality of heads can be detected in a shot video image, so that interference is formed on the positioning of a vehicle-mounted driver, the accuracy of illegal behavior detection such as safety belt detection, call receiving and making detection, smoking detection and the like is further influenced, and the accuracy of illegal behavior detection is low.
Disclosure of Invention
In view of this, embodiments of the present invention provide a method, an apparatus, and an electronic device for detecting a vehicle-mounted driver violation, which can improve accuracy of detecting the vehicle-mounted driver violation, so as to solve a problem that, in an existing method for detecting a vehicle-mounted driver violation, accuracy of detecting the vehicle-mounted driver violation is low in a scene where a plurality of heads are detected in a captured video image.
In a first aspect, an embodiment of the present invention provides a method for detecting an illegal behavior of a vehicle-mounted driver, including:
recognizing a steering wheel in the shot image;
determining an illegal behavior area of the vehicle-mounted driver according to the identified steering wheel;
and detecting the violation of the vehicle-mounted driver by using the determined violation area.
With reference to the first aspect, in a first implementation manner of the first aspect, the recognizing, in the captured image, a steering wheel includes:
carrying out edge detection on the image to obtain an edge detection image;
and detecting the outline of the steering wheel by utilizing a probability Hough transformation algorithm in the edge detection image.
With reference to the first implementation manner of the first aspect, in a second implementation manner of the first aspect, the performing edge detection on the image to obtain an edge-detected image includes:
and converting the image into a gray-scale image, and processing the gray-scale image by using a Canny edge detection algorithm to obtain an edge detection image.
With reference to the first implementation manner of the first aspect, in a third implementation manner of the first aspect, the detecting a steering wheel profile by using a probabilistic Hough transform algorithm includes:
extracting arcs in the edge detection image, detecting the extracted arcs by using a probability Hough transform algorithm, projecting the arcs into the same Hough space, counting the voting number in the Hough space, and selecting the point with the largest voting number as a circle center;
calculating the distance from the circle center to the arc to obtain a radius;
and if the calculated radius meets the preset steering wheel radius threshold, the circle center is the circle center of the steering wheel, and the radius is the radius of the steering wheel.
With reference to the first aspect, in a fourth implementation manner of the first aspect, the violation area is an area from a shoulder to a thigh of the vehicle-mounted driver, and determining the violation area of the vehicle-mounted driver according to the identified steering wheel includes:
and translating the identified steering wheel or the circumscribed rectangle of the steering wheel upwards by a preset first distance threshold value to obtain the area from the shoulder to the thigh of the vehicle-mounted driver.
With reference to the fourth implementation manner of the first aspect, in a fifth implementation manner of the first aspect, the detecting the violation by the vehicle-mounted driver is seat belt detection, and the seat belt detection includes:
determining a safety belt detection area according to the identified area from the shoulder part to the thigh part of the vehicle-mounted driver;
at the determined seat belt detection region, it is detected whether the vehicle-mounted driver is wearing a seat belt.
With reference to the fifth implementation manner of the first aspect, in a sixth implementation manner of the first aspect, the detecting whether a seat belt is present at the determined seat belt detection region includes:
in the determined safety belt detection area, acquiring a self-adaptive Canny threshold according to the image brightness;
according to the acquired Canny threshold, Canny edge detection is carried out on the safety belt detection area to obtain an edge image in the safety belt detection area;
in the obtained edge image in the safety belt detection area, detecting a straight line segment in the edge image in the safety belt detection area by adopting a probability Hough transformation algorithm;
calculating the slope of the straight line segments and the distance between every two straight line segments according to the detected straight line segments, and if the slopes of the two straight line segments are approximately the same and the distance between the two straight line segments is smaller than a preset safety belt distance threshold, determining that a vehicle-mounted driver fastens a safety belt; and if the slopes of the two straight line segments are not approximately the same and the distance between the two straight line segments is not smaller than a preset safety belt distance threshold value, determining that the vehicle-mounted driver does not fasten the safety belt.
With reference to the first aspect, in a seventh implementation manner of the first aspect, the violation area is a head area of the vehicle-mounted driver, and determining the violation area of the vehicle-mounted driver according to the identified steering wheel includes:
reducing the identified steering wheel according to a preset proportion to obtain a reduced steering wheel, and translating the reduced steering wheel upwards by a preset second distance threshold to obtain a head area of the vehicle-mounted driver; or reducing the recognized circumscribed rectangle of the steering wheel according to a preset proportion to obtain a reduced circumscribed rectangle, and translating the reduced circumscribed rectangle upwards by a preset second distance threshold to obtain the head area of the vehicle-mounted driver.
With reference to the seventh implementation manner of the first aspect, in an eighth implementation manner of the first aspect, the detecting the violation behavior of the vehicle-mounted driver is call incoming and outgoing detection, where the call incoming and outgoing detection includes:
in the determined head area of the vehicle driver, whether a mobile phone is present is detected.
With reference to the eighth implementation manner of the first aspect, in a ninth implementation manner of the first aspect, the detecting whether a mobile phone is present in the determined head area of the vehicle-mounted driver includes:
extracting hand texture features contained in a head area of a vehicle-mounted driver according to a preset hand texture feature library;
and inputting the extracted hand texture features into a preset classification model, confirming that the vehicle-mounted driver makes a call if the output result of the classification model shows that the hand features exist, and confirming that the vehicle-mounted driver does not make a call if the output result of the classification model shows that the hand features do not exist.
With reference to the seventh implementation manner of the first aspect, in a tenth implementation manner of the first aspect, the detecting the violation by the vehicle-mounted driver is smoking detection, where the smoking detection includes:
determining a smoking detection area according to the steering wheel and the head area of the vehicle-mounted driver;
and detecting whether the smoking phenomenon exists in the determined smoking detection area.
With reference to the tenth implementation manner of the first aspect, in an eleventh implementation manner of the first aspect, the detecting whether a smoking phenomenon exists in the determined smoking detection area includes:
detecting an incomplete profile contained in the steering wheel profile;
a first connecting line connecting the starting point of the incomplete contour with the center of a circle, and a second connecting line connecting the ending point of the incomplete contour with the center of a circle;
if the angle between the first connecting line and the second connecting line exceeds a preset angle threshold value, extracting hand texture features containing cigarettes in the incomplete contour;
inputting the extracted hand texture features containing the cigarettes into a preset smoking distinguishing and classifying model, confirming that the vehicle-mounted driver has smoking phenomena if the output result of the smoking distinguishing and classifying model shows that the cigarettes exist in the hand features, and confirming that the vehicle-mounted driver does not have smoking phenomena if the output result of the smoking distinguishing and classifying model shows that the cigarettes do not exist in the hand features.
With reference to the tenth implementation manner of the first aspect, in a twelfth implementation manner of the first aspect, the detecting whether a smoking phenomenon exists in the determined smoking detection area includes:
extracting a mouth region in a head region of a vehicle-mounted driver;
and extracting the texture features of the mouth area, confirming that the vehicle-mounted driver smokes if the extracted texture features comprise elongated rectangular areas, and confirming that the vehicle-mounted driver does not smoke if the extracted texture features do not comprise the elongated rectangular areas.
With reference to the first aspect or any one of the first to twelfth embodiments of the first aspect, in a thirteenth embodiment of the first aspect, the method further includes:
recognizing a head image in the shot image, matching the head image with a preset head image library to obtain a vehicle-mounted driver head image matched with the recognized head image, and obtaining vehicle-mounted driver information corresponding to the recognized head image according to the mapping relation between the vehicle-mounted driver head image in the head image library and the vehicle-mounted driver information;
and prompting and/or recording the violation behavior of the vehicle-mounted driver corresponding to the vehicle-mounted driver information according to the obtained vehicle-mounted driver information.
In a second aspect, an embodiment of the present invention provides an apparatus for detecting an illegal action of a vehicle-mounted driver, including: a steering wheel identification module, an illegal activity area determination module, and an illegal activity detection module, wherein,
the steering wheel identification module is used for identifying a steering wheel in the shot image;
the violation area determining module is used for determining the violation area of the vehicle-mounted driver according to the identified steering wheel;
and the violation behavior detection module is used for detecting the violation behavior of the vehicle-mounted driver by utilizing the determined violation behavior area.
With reference to the second aspect, in a first implementation manner of the second aspect, the steering wheel identification module includes: an edge detection unit, a contour detection unit, and a steering wheel acquisition unit, wherein,
the edge detection unit is used for carrying out edge detection on the image to obtain an edge detection image;
the contour detection unit is used for detecting the contour of the steering wheel by utilizing a probability Hough transformation algorithm in the edge detection image;
and the steering wheel acquiring unit is used for acquiring a steering wheel internally tangent to the profile of the steering wheel.
With reference to the first implementation manner of the second aspect, in a second implementation manner of the second aspect, the edge detection unit includes: a conversion subunit and an edge detection subunit, wherein,
the converting subunit is used for converting the image into a gray-scale image;
and the edge detection subunit is used for processing the gray-scale image by using a Canny edge detection algorithm to obtain an edge detection image.
With reference to the first implementation manner of the second aspect, in a third implementation manner of the second aspect, the contour detection unit includes: a circle center determining subunit, a radius determining subunit, and a contour determining unit, wherein,
the circle center determining subunit is used for extracting the circular arc in the edge detection image, detecting the extracted circular arc by using a probability Hough transformation algorithm, projecting the circular arc into the same Hough space, counting the voting number in the Hough space, and selecting the point with the largest voting number as a circle center;
the radius determining subunit is used for calculating the distance from the circle center to the arc to obtain a radius;
and the contour determining unit is used for determining the radius of the steering wheel if the calculated radius meets a preset steering wheel radius threshold, wherein the circle center is the circle center of the steering wheel, and the radius is the radius of the steering wheel.
With reference to the second aspect, in a fourth implementation manner of the second aspect, the violation area determination module includes:
the first shifting unit is used for translating the identified steering wheel or the circumscribed rectangle of the steering wheel upwards by a preset first distance threshold value to obtain the area from the shoulder to the thigh of the vehicle-mounted driver.
With reference to the fourth implementation manner of the second aspect, in a fifth implementation manner of the second aspect, the violation detection module includes: a seat belt detection region determination unit and a seat belt detection unit, wherein,
the safety belt detection area determining unit is used for identifying an area from the shoulder part to the thigh part of the vehicle-mounted driver and determining a safety belt detection area according to the identified area from the shoulder part to the thigh part of the vehicle-mounted driver;
and a seat belt detection unit for detecting whether the vehicle-mounted driver fastens the seat belt at the determined seat belt detection region.
With reference to the fifth embodiment of the second aspect, in a sixth embodiment of the second aspect, the seat belt detection unit includes: a threshold value obtaining subunit, an edge detecting subunit, a straight line segment detecting subunit and a safety belt detecting subunit, wherein,
the threshold acquisition subunit is used for acquiring a self-adaptive Canny threshold in the determined safety belt detection area according to the image brightness;
the edge detection subunit is used for carrying out Canny edge detection on the safety belt detection area according to the obtained Canny threshold value to obtain an edge image in the safety belt detection area;
the straight line segment detection subunit is used for detecting a straight line segment in the edge image in the safety belt detection area by adopting a probability Hough transformation algorithm in the obtained edge image in the safety belt detection area;
the safety belt detection subunit is used for calculating the slope of the straight line segments and the distance between every two straight line segments according to the detected straight line segments, and determining that a vehicle-mounted driver fastens a safety belt if the slopes of the two straight line segments are approximately the same and the distance between the two straight line segments is smaller than a preset safety belt distance threshold; and if the slopes of the two straight line segments are not approximately the same and the distance between the two straight line segments is not smaller than a preset safety belt distance threshold value, determining that the vehicle-mounted driver does not fasten the safety belt.
With reference to the second aspect, in a seventh implementation manner of the second aspect, the violation area determining module includes:
the second shifting unit is used for reducing the identified steering wheel according to a preset proportion to obtain a reduced steering wheel, and translating the reduced steering wheel upwards by a preset second distance threshold to obtain the head area of the vehicle-mounted driver; or the external rectangle is used for reducing the recognized external rectangle of the steering wheel according to a preset proportion to obtain a reduced external rectangle, and the reduced external rectangle is translated upwards by a preset second distance threshold to obtain the head area of the vehicle-mounted driver.
With reference to the seventh implementation manner of the second aspect, in an eighth implementation manner of the second aspect, the violation detection module includes: a head region recognition unit, and a mobile phone detection unit, wherein,
the head area identification unit is used for identifying a steering wheel and determining the head area of the vehicle-mounted driver according to the identified steering wheel;
and the mobile phone detection unit is used for detecting whether a mobile phone exists in the determined head area of the vehicle-mounted driver.
With reference to the eighth implementation manner of the second aspect, in a ninth implementation manner of the second aspect, the mobile phone detection unit includes: a hand texture feature extraction subunit and a mobile phone detection subunit, wherein,
the hand texture feature extraction subunit is used for extracting hand texture features contained in a head area of the vehicle-mounted driver according to a preset hand texture feature library;
and the mobile phone detection subunit is used for inputting the extracted hand texture features into a preset classification model, confirming that the vehicle-mounted driver makes a call if the output result of the classification model shows that the hand features exist, and confirming that the vehicle-mounted driver does not make a call if the output result of the classification model shows that the hand features do not exist.
With reference to the seventh implementation manner of the second aspect, in a tenth implementation manner of the second aspect, the violation detection module includes: a head region identification unit, a smoking detection region determination subunit, and a smoking detection unit, wherein,
the head area identification unit is used for identifying a steering wheel and determining the head area of the vehicle-mounted driver according to the identified steering wheel;
the smoking detection area determining subunit is used for determining a smoking detection area according to the steering wheel and the head area of the vehicle-mounted driver;
and the smoking detection unit is used for detecting whether smoking phenomenon exists in the determined smoking detection area.
With reference to the tenth embodiment of the second aspect, in an eleventh embodiment of the second aspect, the smoking detection unit comprises: a discontinuous detection subunit, a connection subunit, an extraction subunit and a smoking detection subunit, wherein,
a discontinuity detection subunit configured to detect an incomplete contour included in the steering wheel contour;
the connecting subunit is used for connecting a first connecting line between the starting point of the incomplete contour and the circle center and connecting a second connecting line between the ending point of the incomplete contour and the circle center;
the extraction subunit is used for extracting hand texture features containing cigarettes from the incomplete outline if the angle between the first connecting line and the second connecting line exceeds a preset angle threshold;
and the smoking detection subunit is used for inputting the extracted hand texture features containing the cigarettes into a preset smoking distinguishing and classifying model, confirming that the vehicle-mounted driver has smoking phenomena if the output result of the smoking distinguishing and classifying model shows that the cigarettes exist in the hand features, and confirming that the vehicle-mounted driver does not have smoking phenomena if the output result of the smoking distinguishing and classifying model shows that the cigarettes do not exist in the hand features.
With reference to the eleventh implementation manner of the second aspect, in a twelfth implementation manner of the second aspect, the smoking detection unit comprises: a mouth determining subunit and a smoking determining subunit, wherein,
a mouth determination subunit operable to extract a mouth region in a head region of the vehicle-mounted driver;
and the smoking determining subunit is used for extracting the texture features of the mouth area, confirming that the vehicle-mounted driver smokes if the extracted texture features comprise the elongated rectangular area, and confirming that the vehicle-mounted driver does not smoke if the extracted texture features do not comprise the elongated rectangular area.
With reference to the second aspect, any one of the first to twelfth embodiments of the second aspect, in a thirteenth embodiment of the second aspect, the apparatus further comprises: a vehicle-mounted driver information acquisition module and a violation processing module, wherein,
the vehicle-mounted driver information acquisition module is used for identifying a head image in the shot image, matching the head image with a preset head image library to obtain a vehicle-mounted driver head image matched with the identified head image, and obtaining vehicle-mounted driver information corresponding to the identified head image according to the mapping relation between the vehicle-mounted driver head image and the vehicle-mounted driver information in the head image library;
and the violation behavior processing module is used for prompting and/or recording violation behaviors of the vehicle-mounted driver corresponding to the vehicle-mounted driver information according to the obtained vehicle-mounted driver information.
In a third aspect, an embodiment of the present invention provides an electronic device, where the electronic device includes: the device comprises a shell, a processor, a memory, a circuit board and a power circuit, wherein the circuit board is arranged in a space enclosed by the shell, and the processor and the memory are arranged on the circuit board; a power supply circuit for supplying power to each circuit or device of the electronic apparatus; the memory is used for storing executable program codes; the processor executes a program corresponding to the executable program code by reading the executable program code stored in the memory, and is used for executing any one of the methods for detecting the violation behavior of the vehicle-mounted driver.
According to the method, the device and the electronic equipment for detecting the violation behavior of the vehicle-mounted driver, which are provided by the embodiment of the invention, the steering wheel is identified in the shot image; determining an illegal behavior area of the vehicle-mounted driver according to the identified steering wheel; the method has the advantages that the violation behaviors of the vehicle-mounted driver are detected by utilizing the determined violation behavior area, so that the accuracy of the violation behavior detection of the vehicle-mounted driver can be improved, and the problem that the accuracy of the violation behavior detection of the vehicle-mounted driver is low in the scene that a plurality of heads are detected in a shot video image in the existing method for detecting the violation behaviors of the vehicle-mounted driver is solved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flowchart illustrating a method for detecting violation of a vehicle-mounted driver according to an embodiment of the present invention;
FIG. 2 is a schematic structural diagram of a device for detecting violation of a vehicle-mounted driver according to a second embodiment of the present invention;
fig. 3 is a schematic structural diagram of an embodiment of an electronic device according to the present invention.
Detailed Description
Embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
It should be understood that the described embodiments are only some embodiments of the invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is a flowchart illustrating a method for detecting an illegal action of a vehicle-mounted driver according to an embodiment of the present invention, and as shown in fig. 1, the method of this embodiment may include:
step 101, identifying a steering wheel in a shot image;
in this embodiment, as an optional embodiment, the image is an image acquired by detecting the behavior of a vehicle-mounted driver with a vehicle-mounted camera installed at a predetermined position in the cab.
As another alternative embodiment, after detecting and acquiring the vehicle-mounted driver behavior to obtain an image, the vehicle-mounted camera sends the image to the image analysis server by using a preset communication module, and the image analysis server analyzes and processes each uploaded image.
In this embodiment, as an optional embodiment, recognizing the steering wheel in the captured image includes:
a11, carrying out edge detection on the image to obtain an edge detection image;
in this embodiment, as an optional embodiment, a Canny edge detection algorithm is used to perform edge detection on the image. The Canny edge detection algorithm is a multi-stage edge detection operator, is designed according to the optimal criterion of image edge detection, has good edge monitoring performance, and is increasingly widely applied to image processing. The Canny edge detection algorithm is not easily interfered by noise, and can detect a real weak edge; in order to accurately define the gradient value range of the edge, the Canny edge detection algorithm is implemented by two thresholds, which are respectively used for detecting a strong edge and a weak edge, and the weak edge is included in the output image when the weak edge and the strong edge are connected.
As an alternative embodiment, performing edge detection on the image to obtain an edge detection image includes:
and converting the image into a gray-scale image, and processing the gray-scale image by using a Canny edge detection algorithm to obtain an edge detection image.
In this embodiment, as an optional embodiment, after the edge detection image is obtained, filtering processing may be performed on the obtained edge detection image.
And A12, detecting the outline of the steering wheel in the edge detection image by utilizing a probability Hough transformation algorithm.
In this embodiment, of the Canny edges in the edge detection image, the Canny edge of the steering wheel is a circle and has a distinct feature.
In this embodiment, as an optional embodiment, a probability Hough transform algorithm is used to detect a circle in the Canny edge, and if a steering wheel exists, a fitting circle can be obtained.
In this embodiment, the steering wheel profile includes a center of a circle of the steering wheel and a radius of the steering wheel. As an alternative embodiment, the detecting the steering wheel profile using the probabilistic Hough transform algorithm includes:
extracting arcs in the edge detection image, detecting the extracted arcs by using a probability Hough transform algorithm, projecting the arcs into the same Hough space, counting the voting number in the Hough space, and selecting the point with the largest voting number as a circle center;
calculating the distance from the circle center to the arc to obtain a radius;
and if the calculated radius meets the preset steering wheel radius threshold, the circle center is the circle center of the steering wheel, and the radius is the radius of the steering wheel.
In this embodiment, as an alternative embodiment, the diameter of the steering wheel is generally 80 to 130 pixels.
In this embodiment, as an optional embodiment, the method may further include:
and A13, acquiring the steering wheel internally tangent to the profile of the steering wheel.
In this embodiment, the steering wheel is a square frame externally connected to the steering wheel. After the probability Hough transformation algorithm is adopted to detect the fitting circle, the circle center of the fitting circle in the edge detection image and the diameter of the fitting circle can be obtained, and the external square frame of the steering wheel internally tangent to the profile of the steering wheel can be obtained. Thus, the subsequent processing is facilitated.
Step 102, determining an illegal behavior area of a vehicle-mounted driver according to the identified steering wheel;
in this embodiment, as an alternative embodiment, since the circular shape of the steering wheel has obvious features, the driver is the only person who operates the steering wheel, and has a more definite positional relationship with the steering wheel. Therefore, by locating the steering wheel, the in-vehicle driver area including the violation area can be determined. Therefore, when the number of passengers is large in the public transport means, the number of detected head images is possibly large through detection of shot images, so that the positioning of the vehicle-mounted driver is influenced, but in the public transport means, only one steering wheel has obvious characteristics, so that the area of the vehicle-mounted driver can be determined according to the position area of the steering wheel, the accuracy of the positioned vehicle-mounted driver can be improved, and the accuracy of the detection of the illegal behaviors of the vehicle-mounted driver is improved.
In this embodiment, different violation behaviors correspond to different violation behavior regions. For example, for detection of seat belt violation, the corresponding violation area is the area from the shoulder to the thigh of the vehicle-mounted driver; for the detection of the illegal behaviors of calling and receiving, the corresponding illegal behavior area is the head area of the vehicle-mounted driver; for detection of smoking violations, the corresponding violation area includes: steering wheel and vehicle driver head area. The steering wheel can be zoomed according to the preset setting, and then translated upwards by a corresponding distance to obtain each violation area.
And 103, detecting the violation of the vehicle-mounted driver by using the determined violation area.
In this embodiment, detecting an illegal behavior of a vehicle-mounted driver includes: seat belt detection, call on/off detection, and smoke detection, wherein,
the violation area is an area from the shoulder to the thigh of the vehicle-mounted driver, and the determining of the violation area of the vehicle-mounted driver according to the identified steering wheel comprises the following steps:
and translating the identified steering wheel or the circumscribed rectangle of the steering wheel upwards by a preset first distance threshold value to obtain the area from the shoulder to the thigh of the vehicle-mounted driver.
The seat belt detection includes:
b11, determining a safety belt detection area according to the identified area from the shoulder part to the thigh part of the vehicle-mounted driver;
in this embodiment, the safety belt detection area (the area from the shoulder to the thigh of the vehicle-mounted driver) is obtained by translating the steering wheel in the upward direction by the preset distance according to the recognized steering wheel, that is, the square frame externally connected to the steering wheel.
B12, detecting whether the vehicle driver is belted or not at the determined belt detection area.
In this embodiment, as an optional embodiment, the detecting whether there is a seat belt at the determined seat belt detection region includes:
b111, acquiring a self-adaptive Canny threshold value in the determined safety belt detection area according to the image brightness;
b112, according to the obtained Canny threshold, carrying out Canny edge detection on the safety belt detection area to obtain an edge image in the safety belt detection area;
in this embodiment, as an optional embodiment, filtering processing, for example, mean filtering processing may be performed on the edge image in the seat belt detection region.
B113, detecting a straight line segment in the edge image in the safety belt detection area by adopting a probability Hough transform algorithm in the obtained edge image in the safety belt detection area;
b114, calculating the slope of the straight line segments and the distance between every two straight line segments according to the detected straight line segments, and if the slopes of the two straight line segments are approximately the same and the distance between the two straight line segments is smaller than a preset safety belt distance threshold, determining that a vehicle-mounted driver fastens a safety belt; and if the slopes of the two straight line segments are not approximately the same and the distance between the two straight line segments is not smaller than a preset safety belt distance threshold value, determining that the vehicle-mounted driver does not fasten the safety belt.
In this embodiment, after the straight line is detected, the detected line segment is filtered according to the characteristics of the types of the commonly used seat belts, the straight line not belonging to the seat belt is filtered, a filtered straight line detection result is obtained, and a final conclusion whether the seat belt is fastened or not is obtained according to the filtered straight line detection result. For example, if the slopes of the two straight line segments are approximately the same, that is, the difference between the slopes of the two straight line segments is within a preset slope error range, and the distance between the two straight line segments is not less than a preset safety belt distance threshold, it is indicated that safety belt characteristics exist in the safety belt detection area, and it is determined that the vehicle-mounted driver has fastened the safety belt and does not have an illegal behavior.
As an alternative embodiment, the violation area is a head area of the vehicle-mounted driver, and determining the violation area of the vehicle-mounted driver according to the identified steering wheel includes:
and reducing the identified steering wheel according to a preset proportion to obtain a reduced steering wheel, and translating the reduced steering wheel upwards by a preset second distance threshold to obtain the head area of the vehicle-mounted driver.
The incoming and outgoing call detection comprises:
c11, determining the head area of the vehicle driver according to the recognized steering wheel;
in this embodiment, based on the obtained steering wheel, the steering wheel is reduced (to the size of the head area) according to a preset ratio, and the reduced steering wheel is translated upward by a preset second distance threshold, so as to obtain an area for detecting the call incoming and outgoing of the vehicle-mounted driver, that is, the head area of the vehicle-mounted driver.
Optionally, the identified circumscribed rectangle of the steering wheel may be reduced according to a preset proportion to obtain a reduced circumscribed rectangle, and the reduced circumscribed rectangle is translated upwards by a preset second distance threshold to obtain the head area of the vehicle-mounted driver.
In this embodiment, the second distance threshold may be determined according to the statistical distance between the head of each vehicle-mounted driver and the steering wheel.
C12, detecting whether there is a mobile phone in the determined head area of the vehicle driver.
In the embodiment, when a user makes and receives a call, the user needs to hold the mobile phone by hand, so that the head position of the vehicle-mounted driver has obvious hand characteristics, and the hand of the user can be used as a detection target.
In this embodiment, as an optional embodiment, the detecting whether there is a mobile phone in the determined head area of the vehicle-mounted driver includes:
c111, extracting hand texture features contained in a head area of the vehicle-mounted driver according to a preset hand texture feature library;
and C112, inputting the extracted hand texture features into a preset classification model, confirming that the vehicle-mounted driver makes a call if the output result of the classification model shows that the hand features exist, and confirming that the vehicle-mounted driver does not make a call if the output result of the classification model shows that the hand features do not exist.
In this embodiment, the features of the target are extracted first, and then the target is sent to a classifier for classification. As an alternative embodiment, the texture features include: local Binary Pattern (LBP) features, Histogram of Oriented Gradient (HOG) features, Haar features, EGL features, and the like. The classification model is a detection classification model for training call receiving and making, and comprises the following steps: support Vector Machine (SVM) classifiers, neural network classifiers, and the like.
As an alternative embodiment, the violation area is a steering wheel or a head area of a vehicle driver, and the smoking detection comprises:
d11, determining the head area of the vehicle driver according to the recognized steering wheel;
d12, determining a smoking detection area according to the steering wheel and the head area of the vehicle-mounted driver;
in the embodiment, two smoking detection areas exist, namely a steering wheel, and if smoking phenomenon occurs in the steering wheel, cigarettes can be clamped in the hands of a vehicle-mounted driver; secondly, if the vehicle-mounted driver smokes in the head area, the cigarettes can be contained in the mouth of the vehicle-mounted driver. Thus, the smoking detection area comprises: steering wheel and vehicle driver head area.
And D13, detecting whether the smoking phenomenon exists in the determined smoking detection area.
In this embodiment, the smoking detection area is a steering wheel, and detecting whether a smoking phenomenon exists in the determined smoking detection area includes:
d111, detecting an incomplete contour contained in the steering wheel contour;
d112, a first connecting line connecting the starting point of the incomplete contour with the center of a circle, and a second connecting line connecting the ending point of the incomplete contour with the center of a circle;
d113, if the angle between the first connecting line and the second connecting line exceeds a preset angle threshold value, extracting hand texture features containing cigarettes in the incomplete contour;
and D114, inputting the extracted hand texture features containing the cigarettes into a preset smoking judgment classification model, confirming that the vehicle-mounted driver has smoking phenomena if the output result of the smoking judgment classification model shows that the cigarettes exist in the hand features, and confirming that the vehicle-mounted driver does not have smoking phenomena if the output result of the smoking judgment classification model shows that the cigarettes do not exist in the hand features.
In this embodiment, for the steering wheel, it may be verified whether the detected circle is complete according to the method for detecting a circle described above, and if the continuous incomplete area exceeds 40 degrees, it may be directly determined that the continuous incomplete area is the position of the hand. For example, if a continuous incomplete region is detected on the edge of a circle, a first connecting line connecting the starting point and the center of the continuous incomplete region, and a second connecting line connecting the ending point and the center of the continuous incomplete region, no Hough circle is detected between the first connecting line and the second connecting line as the continuous incomplete region, and if the angle between the first connecting line and the second connecting line exceeds a preset angle threshold value, for example, 40 degrees, it is determined that the hand feature is present in the continuous incomplete region.
And then classifying the continuous incomplete area by using a pattern recognition method, extracting LBP characteristics and the like, and then classifying by using a neural network to finally obtain whether the steering wheel has smoke or not.
As another alternative, the smoking detection area is a head area of a vehicle driver, and the detecting whether the smoking phenomenon exists in the determined smoking detection area comprises:
d211, extracting a mouth area in the head area of the vehicle-mounted driver;
and D212, extracting the texture features of the mouth area, confirming that the vehicle-mounted driver smokes if the extracted texture features comprise the elongated rectangular area, and confirming that the vehicle-mounted driver does not smoke if the extracted texture features do not comprise the elongated rectangular area.
In this embodiment, for the head area of the vehicle-mounted driver, in this embodiment, in a scene where the cigarette is contained in the mouth, the cigarette has an obvious contour feature, and therefore, in this embodiment, contour extraction is directly performed on the head area of the vehicle-mounted driver, and then whether a long and thin rectangular area exists is detected, and if so, it indicates that the vehicle-mounted driver has a smoking phenomenon.
As an alternative embodiment, the method further comprises:
and 104, recognizing a head image in the shot image, matching the head image with a preset head image library to obtain a vehicle-mounted driver head image matched with the recognized head image, and obtaining vehicle-mounted driver information corresponding to the recognized head image according to the mapping relation between the vehicle-mounted driver head image in the head image library and the vehicle-mounted driver information.
In this embodiment, specific in-vehicle driver information can be determined by head image matching.
As another alternative embodiment, the method further comprises:
and 105, prompting and/or recording violation behaviors of the vehicle-mounted driver corresponding to the vehicle-mounted driver information according to the obtained vehicle-mounted driver information.
In this embodiment, after detecting that the vehicle-mounted driver has the violation, the vehicle-mounted driver may be correspondingly processed according to a preset processing policy. For example, prompt information is sent to the vehicle-mounted driver to prompt the vehicle-mounted driver for violation.
In the embodiment, the position area of the vehicle-mounted driver is determined according to the position area of the steering wheel by checking the steering wheel in the image; and then, defining an illegal behavior area and detecting whether the vehicle-mounted driver has illegal behaviors. For example, according to the steering wheel, a rough area of the head of the vehicle driver is determined, the position of the hand is checked for the area, and then whether to make or not a call is detected; alternatively, the hand position is detected in the steering wheel and in the head area of the vehicle driver, and whether or not to smoke is determined. Therefore, the position area of the vehicle-mounted driver is positioned according to the steering wheel, the problem of positioning of the vehicle-mounted driver of the public transport means can be effectively solved, and the positioning result of the vehicle-mounted driver is prevented from being influenced by too many passengers in the public transport means; furthermore, the illegal behavior area is positioned according to the vehicle-mounted driver area with accurate positioning, abnormal illegal behaviors such as no safety belt fastening, call receiving and making, smoking and the like can be automatically detected, a relatively ideal abnormal illegal behavior detection effect is obtained, the accuracy of detecting the illegal behaviors of the vehicle-mounted driver is improved, for example, the method for detecting the head area of the vehicle-mounted driver and further detecting whether the vehicle-mounted driver receives or makes a call can effectively reduce false detection; in addition, in the public transport system, the embodiment establishes an effective supervision system for helping the public transport system, reduces the workload of public transport supervision, reduces the labor cost, improves the working efficiency, supervises and urges the vehicle-mounted drivers of public transport means to standardize own behaviors, plays a great role in maintaining the safety of the broad travelers, can effectively avoid traffic accidents caused by abnormal illegal behaviors of the vehicle-mounted drivers, and provides safety guarantee for traveling.
Fig. 2 is a schematic structural diagram of a second apparatus for detecting an illegal action of a vehicle-mounted driver according to an embodiment of the present invention, and as shown in fig. 2, the apparatus of this embodiment may include: a steering wheel identification module 21, an infraction area determination module 22, and an infraction detection module 23, wherein,
a steering wheel recognition module 21 for recognizing a steering wheel in the photographed image;
in this embodiment, the image is an image acquired by detecting the behavior of a vehicle-mounted driver with a vehicle-mounted camera installed at a predetermined position in the cab.
As an alternative embodiment, the steering wheel identifying module 21 includes: an edge detection unit, a contour detection unit, and a steering wheel acquisition unit (not shown in the figure), wherein,
the edge detection unit is used for carrying out edge detection on the image to obtain an edge detection image;
in this embodiment, the Canny edge detection algorithm is used to perform edge detection on the image.
In this embodiment, as an optional embodiment, the edge detecting unit includes: a conversion subunit and an edge detection subunit, wherein,
the converting subunit is used for converting the image into a gray-scale image;
and the edge detection subunit is used for processing the gray-scale image by using a Canny edge detection algorithm to obtain an edge detection image.
In this embodiment, as an optional embodiment, after the edge detection image is obtained, filtering processing may be performed on the obtained edge detection image.
The contour detection unit is used for detecting the contour of the steering wheel by utilizing a probability Hough transformation algorithm in the edge detection image;
in this embodiment, as an optional embodiment, the contour detection unit includes: a circle center determining subunit, a radius determining subunit, and a contour determining unit, wherein,
the circle center determining subunit is used for extracting the circular arc in the edge detection image, detecting the extracted circular arc by using a probability Hough transformation algorithm, projecting the circular arc into the same Hough space, counting the voting number in the Hough space, and selecting the point with the largest voting number as a circle center;
the radius determining subunit is used for calculating the distance from the circle center to the arc to obtain a radius;
and the contour determining unit is used for determining the radius of the steering wheel if the calculated radius meets a preset steering wheel radius threshold, wherein the circle center is the circle center of the steering wheel, and the radius is the radius of the steering wheel.
And the steering wheel acquiring unit is used for acquiring a steering wheel internally tangent to the profile of the steering wheel.
The violation area determining module 22 is configured to determine a violation area of the vehicle-mounted driver according to the identified steering wheel;
in this embodiment, as an optional embodiment, the violation area determining module 22 includes:
and the first displacement unit (not shown in the figure) is used for translating the identified steering wheel or the circumscribed rectangle of the steering wheel upwards by a preset first distance threshold value to obtain the area from the shoulder to the thigh of the vehicle-mounted driver.
As another alternative embodiment, the violation area determination module 22 includes:
the second shifting unit is used for reducing the identified steering wheel according to a preset proportion to obtain a reduced steering wheel, and translating the reduced steering wheel upwards by a preset second distance threshold to obtain the head area of the vehicle-mounted driver; or the external rectangle is used for reducing the recognized external rectangle of the steering wheel according to a preset proportion to obtain a reduced external rectangle, and the reduced external rectangle is translated upwards by a preset second distance threshold to obtain the head area of the vehicle-mounted driver.
And the illegal behavior detection module 23 is configured to detect the illegal behavior of the vehicle-mounted driver by using the determined illegal behavior region.
In this embodiment, detecting an illegal behavior of a vehicle-mounted driver includes: seat belt detection, call answering detection, and smoke detection.
As an alternative embodiment, the violation detection module 23 includes: a seat belt detection region determining unit, and a seat belt detection unit (not shown in the drawings), wherein,
the safety belt detection area determining unit is used for identifying an area from the shoulder part to the thigh part of the vehicle-mounted driver and determining a safety belt detection area according to the identified area from the shoulder part to the thigh part of the vehicle-mounted driver;
and a seat belt detection unit for detecting whether the vehicle-mounted driver fastens the seat belt at the determined seat belt detection region.
In this embodiment, as an optional embodiment, the seat belt detection unit includes: a threshold value obtaining subunit, an edge detecting subunit, a straight line segment detecting subunit and a safety belt detecting subunit, wherein,
the threshold acquisition subunit is used for acquiring a self-adaptive Canny threshold in the determined safety belt detection area according to the image brightness;
the edge detection subunit is used for carrying out Canny edge detection on the safety belt detection area according to the obtained Canny threshold value to obtain an edge image in the safety belt detection area;
the straight line segment detection subunit is used for detecting a straight line segment in the edge image in the safety belt detection area by adopting a probability Hough transformation algorithm in the obtained edge image in the safety belt detection area;
the safety belt detection subunit is used for calculating the slope of the straight line segments and the distance between every two straight line segments according to the detected straight line segments, and determining that a vehicle-mounted driver fastens a safety belt if the slopes of the two straight line segments are approximately the same and the distance between the two straight line segments is smaller than a preset safety belt distance threshold; and if the slopes of the two straight line segments are not approximately the same and the distance between the two straight line segments is not smaller than a preset safety belt distance threshold value, determining that the vehicle-mounted driver does not fasten the safety belt.
As another alternative embodiment, the violation detection module 23 includes: a head region recognition unit, and a mobile phone detection unit, wherein,
the head area identification unit is used for identifying a steering wheel and determining the head area of the vehicle-mounted driver according to the identified steering wheel;
in this embodiment, the steering wheel is reduced according to the obtained steering wheel in a preset ratio, and the reduced steering wheel is translated upward by a preset second distance threshold, so as to obtain an area for detecting whether the vehicle-mounted driver makes or receives a call.
And the mobile phone detection unit is used for detecting whether a mobile phone exists in the determined head area of the vehicle-mounted driver.
In this embodiment, as an optional embodiment, the mobile phone detecting unit includes: a hand texture feature extraction subunit and a mobile phone detection subunit, wherein,
the hand texture feature extraction subunit is used for extracting hand texture features contained in a head area of the vehicle-mounted driver according to a preset hand texture feature library;
in this embodiment, as an optional embodiment, the texture feature includes: local binary pattern features, histogram of oriented gradient features, Haar features, EGL features, and the like. The classification model comprises: support vector machine classifiers, neural network classifiers, and the like.
And the mobile phone detection subunit is used for inputting the extracted hand texture features into a preset classification model, confirming that the vehicle-mounted driver makes a call if the output result of the classification model shows that the hand features exist, and confirming that the vehicle-mounted driver does not make a call if the output result of the classification model shows that the hand features do not exist.
As still another alternative embodiment, the violation detection module 23 includes: a head region identification unit, a smoking detection region determination subunit, and a smoking detection unit, wherein,
the head area identification unit is used for identifying a steering wheel and determining the head area of the vehicle-mounted driver according to the identified steering wheel;
the smoking detection area determining subunit is used for determining a smoking detection area according to the steering wheel and the head area of the vehicle-mounted driver;
and the smoking detection unit is used for detecting whether smoking phenomenon exists in the determined smoking detection area.
In this embodiment, the smoking detection unit includes: a discontinuous detection subunit, a connection subunit, an extraction subunit and a smoking detection subunit, wherein,
a discontinuity detection subunit configured to detect an incomplete contour included in the steering wheel contour;
the connecting subunit is used for connecting a first connecting line between the starting point of the incomplete contour and the circle center and connecting a second connecting line between the ending point of the incomplete contour and the circle center;
the extraction subunit is used for extracting hand texture features containing cigarettes from the incomplete outline if the angle between the first connecting line and the second connecting line exceeds a preset angle threshold;
and the smoking detection subunit is used for inputting the extracted hand texture features containing the cigarettes into a preset smoking distinguishing and classifying model, confirming that the vehicle-mounted driver has smoking phenomena if the output result of the smoking distinguishing and classifying model shows that the cigarettes exist in the hand features, and confirming that the vehicle-mounted driver does not have smoking phenomena if the output result of the smoking distinguishing and classifying model shows that the cigarettes do not exist in the hand features.
As another alternative embodiment, the smoking detection area is a head area of a vehicle driver, and the smoking detection unit includes: a mouth determining subunit and a smoking determining subunit, wherein,
a mouth determination subunit operable to extract a mouth region in a head region of the vehicle-mounted driver;
and the smoking determining subunit is used for extracting the texture features of the mouth area, confirming that the vehicle-mounted driver smokes if the extracted texture features comprise the elongated rectangular area, and confirming that the vehicle-mounted driver does not smoke if the extracted texture features do not comprise the elongated rectangular area.
In this embodiment, as an optional embodiment, the apparatus may further include: an in-vehicle driver information acquisition module 24, and an violation handling module 25, wherein,
the vehicle-mounted driver information acquisition module 24 is configured to identify a head image in the captured image, match the head image with a preset head image library to obtain a vehicle-mounted driver head image matched with the identified head image, and obtain vehicle-mounted driver information corresponding to the identified head image according to a mapping relationship between the vehicle-mounted driver head image and the vehicle-mounted driver information in the head image library;
and the violation behavior processing module 25 is configured to prompt and/or record a violation behavior of the vehicle-mounted driver corresponding to the vehicle-mounted driver information according to the obtained vehicle-mounted driver information.
The apparatus of this embodiment may be used to implement the technical solution of the method embodiment shown in fig. 1, and the implementation principle and the technical effect are similar, which are not described herein again.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. The term "comprising", without further limitation, means that the element so defined is not excluded from the group consisting of additional identical elements in the process, method, article, or apparatus that comprises the element.
All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments.
In particular, as for the apparatus embodiment, since it is substantially similar to the method embodiment, the description is relatively simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof.
In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
The embodiment of the invention also provides electronic equipment, and the electronic equipment comprises the device in any one of the embodiments.
Fig. 3 is a schematic structural diagram of an embodiment of an electronic device of the present invention, which can implement the processes of the embodiments shown in fig. 1-2 of the present invention, and as shown in fig. 3, the electronic device may include: the device comprises a shell 31, a processor 32, a memory 33, a circuit board 34 and a power circuit 35, wherein the circuit board 34 is arranged inside a space enclosed by the shell 31, and the processor 32 and the memory 33 are arranged on the circuit board 34; a power supply circuit 35 for supplying power to each circuit or device of the electronic apparatus; the memory 33 is used for storing executable program codes; the processor 32 executes a program corresponding to the executable program code by reading the executable program code stored in the memory 33, so as to execute the method for detecting the violation of the vehicle-mounted driver according to any of the foregoing embodiments.
The specific execution process of the above steps by the processor 32 and the steps further executed by the processor 32 by running the executable program code may refer to the description of the embodiment shown in fig. 1-2 of the present invention, and are not described herein again.
The electronic device exists in a variety of forms, including but not limited to:
(1) a mobile communication device: such devices are characterized by mobile communications capabilities and are primarily targeted at providing voice, data communications. Such terminals include: smart phones (e.g., iphones), multimedia phones, functional phones, and low-end phones, among others.
(2) Ultra mobile personal computer device: the equipment belongs to the category of personal computers, has calculation and processing functions and generally has the characteristic of mobile internet access. Such terminals include: PDA, MID, and UMPC devices, etc., such as ipads.
(3) A portable entertainment device: such devices can display and play multimedia content. This type of device comprises: audio, video players (e.g., ipods), handheld game consoles, electronic books, and smart toys and portable car navigation devices.
(4) A server: the device for providing the computing service comprises a processor, a hard disk, a memory, a system bus and the like, and the server is similar to a general computer architecture, but has higher requirements on processing capacity, stability, reliability, safety, expandability, manageability and the like because of the need of providing high-reliability service.
(5) And other electronic equipment with data interaction function.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.
For convenience of description, the above devices are described separately in terms of functional division into various units/modules. Of course, the functionality of the units/modules may be implemented in one or more software and/or hardware implementations of the invention.
The above description of the embodiments will make clear to those skilled in the art that the present invention can be implemented
The invention can be implemented by means of software plus a necessary general-purpose hardware platform. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which may be stored in a storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the embodiments or some parts of the embodiments.
The above description is only for the specific embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (29)

1. A method of detecting a violation by a vehicle driver, comprising:
identifying a steering wheel in the shot image to obtain a steering wheel outline, wherein the steering wheel outline comprises a steering wheel circle center and a steering wheel radius;
determining an illegal behavior area of the vehicle-mounted driver according to the identified steering wheel, wherein the illegal behavior area comprises at least one of the following items: the region from the shoulder to the thigh of the vehicle-mounted driver, the head region of the driver and a steering wheel;
detecting the violation behavior of the vehicle-mounted driver by using preset characteristics appearing in the determined violation behavior area, wherein the preset characteristics comprise at least one of the following characteristics: hand texture features, mouth region texture features, straight line segment features, the violation including at least one of: seat belt violation, call receiving and making violation, smoking violation;
the detecting the violation of the vehicle-mounted driver by using the preset features appearing in the determined violation area comprises:
detecting an incomplete profile contained in the steering wheel profile;
a first connecting line connecting the starting point of the incomplete contour with the center of a circle, and a second connecting line connecting the ending point of the incomplete contour with the center of a circle;
if the angle between the first and second links exceeds a preset angle threshold, the incomplete contour is classified by a pattern recognition method to determine whether a cigarette is present in the steering wheel.
2. The method for detecting vehicle-mounted driver violation according to claim 1, wherein identifying a steering wheel in the captured image comprises:
carrying out edge detection on the image to obtain an edge detection image;
and detecting the outline of the steering wheel by utilizing a probability Hough transformation algorithm in the edge detection image.
3. The method for detecting the violation of the vehicle-mounted driver according to claim 2, wherein the performing edge detection on the image to obtain an edge detection image comprises:
and converting the image into a gray-scale image, and processing the gray-scale image by using a Canny edge detection algorithm to obtain an edge detection image.
4. The method of detecting an in-vehicle driver violation according to claim 2, wherein said detecting a steering wheel profile using a probabilistic Hough transform algorithm comprises:
extracting arcs in the edge detection image, detecting the extracted arcs by using a probability Hough transform algorithm, projecting the arcs into the same Hough space, counting the voting number in the Hough space, and selecting the point with the largest voting number as a circle center;
calculating the distance from the circle center to the arc to obtain a radius;
and if the calculated radius meets the preset steering wheel radius threshold, the circle center is the circle center of the steering wheel, and the radius is the radius of the steering wheel.
5. The method for detecting the violation of the vehicle-mounted driver according to claim 1, wherein the violation area is an area from a shoulder to a thigh of the vehicle-mounted driver, and the determining the violation area of the vehicle-mounted driver according to the identified steering wheel comprises:
and translating the identified steering wheel or the circumscribed rectangle of the steering wheel upwards by a preset first distance threshold value to obtain the area from the shoulder to the thigh of the vehicle-mounted driver.
6. The method for detecting vehicle driver violation according to claim 5, wherein said detecting vehicle driver violation is seat belt detection, said seat belt detection comprising:
determining a safety belt detection area according to the identified area from the shoulder part to the thigh part of the vehicle-mounted driver;
at the determined seat belt detection region, it is detected whether the vehicle-mounted driver is wearing a seat belt.
7. The method of detecting an in-vehicle driver violation according to claim 6, wherein said detecting whether a seat belt is present at the determined seat belt detection zone comprises:
in the determined safety belt detection area, acquiring a self-adaptive Canny threshold according to the image brightness;
according to the acquired Canny threshold, Canny edge detection is carried out on the safety belt detection area to obtain an edge image in the safety belt detection area;
in the obtained edge image in the safety belt detection area, detecting a straight line segment in the edge image in the safety belt detection area by adopting a probability Hough transformation algorithm;
calculating the slope of the straight line segments and the distance between every two straight line segments according to the detected straight line segments, and if the slopes of the two straight line segments are approximately the same and the distance between the two straight line segments is smaller than a preset safety belt distance threshold, determining that a vehicle-mounted driver fastens a safety belt; and if the slopes of the two straight line segments are not approximately the same and the distance between the two straight line segments is not smaller than a preset safety belt distance threshold value, determining that the vehicle-mounted driver does not fasten the safety belt.
8. The method for detecting vehicle driver violation according to claim 1, wherein said violation area is a vehicle driver head area; the determining of the violation area of the vehicle-mounted driver according to the identified steering wheel comprises:
reducing the identified steering wheel according to a preset proportion to obtain a reduced steering wheel, and translating the reduced steering wheel upwards by a preset second distance threshold to obtain a head area of the vehicle-mounted driver; or
And reducing the identified external rectangle of the steering wheel according to a preset proportion to obtain a reduced external rectangle, and translating the reduced external rectangle upwards by a preset second distance threshold to obtain the head area of the vehicle-mounted driver.
9. The method for detecting the violation of the vehicle driver according to claim 8, wherein the detecting the violation of the vehicle driver is a call incoming detection, and the call incoming detection comprises:
in the determined head area of the vehicle driver, whether a mobile phone is present is detected.
10. The method of detecting vehicle driver violations as claimed in claim 9, wherein said detecting whether a mobile phone is present in the determined vehicle driver head area comprises:
extracting hand texture features contained in a head area of a vehicle-mounted driver according to a preset hand texture feature library;
and inputting the extracted hand texture features into a preset classification model, confirming that the vehicle-mounted driver makes a call if the output result of the classification model shows that the hand features exist, and confirming that the vehicle-mounted driver does not make a call if the output result of the classification model shows that the hand features do not exist.
11. The method for detecting the violation of the vehicle driver according to claim 8, wherein the detecting the violation of the vehicle driver is smoking detection, and the smoking detection comprises:
determining a smoking detection area according to the steering wheel and the head area of the vehicle-mounted driver;
and detecting whether the smoking phenomenon exists in the determined smoking detection area.
12. The method of detecting an in-vehicle driver violation according to claim 11, wherein said detecting whether a smoking event is present in the determined smoking detection area comprises:
detecting an incomplete profile contained in the steering wheel profile;
a first connecting line connecting the starting point of the incomplete contour with the center of a circle, and a second connecting line connecting the ending point of the incomplete contour with the center of a circle;
if the angle between the first connecting line and the second connecting line exceeds a preset angle threshold value, extracting hand texture features containing cigarettes in the incomplete contour;
inputting the extracted hand texture features containing the cigarettes into a preset smoking distinguishing and classifying model, confirming that the vehicle-mounted driver has smoking phenomena if the output result of the smoking distinguishing and classifying model shows that the cigarettes exist in the hand features, and confirming that the vehicle-mounted driver does not have smoking phenomena if the output result of the smoking distinguishing and classifying model shows that the cigarettes do not exist in the hand features.
13. The method of detecting an in-vehicle driver violation according to claim 11, wherein said detecting whether a smoking event is present in the determined smoking detection area comprises:
extracting a mouth region in a head region of a vehicle-mounted driver;
and extracting the texture features of the mouth area, confirming that the vehicle-mounted driver smokes if the extracted texture features comprise elongated rectangular areas, and confirming that the vehicle-mounted driver does not smoke if the extracted texture features do not comprise the elongated rectangular areas.
14. The method of detecting an in-vehicle driver violation according to any of claims 1-13, further comprising:
recognizing a head image in the shot image, matching the head image with a preset head image library to obtain a vehicle-mounted driver head image matched with the recognized head image, and obtaining vehicle-mounted driver information corresponding to the recognized head image according to the mapping relation between the vehicle-mounted driver head image in the head image library and the vehicle-mounted driver information;
and prompting and/or recording the violation behavior of the vehicle-mounted driver corresponding to the vehicle-mounted driver information according to the obtained vehicle-mounted driver information.
15. An apparatus for detecting an in-vehicle driver violation, comprising: a steering wheel identification module, an illegal activity area determination module, and an illegal activity detection module, wherein,
the steering wheel identification module is used for identifying a steering wheel in the shot image to obtain a steering wheel outline, and the steering wheel outline comprises a steering wheel circle center and a steering wheel radius;
the violation area determining module is used for determining a violation area of the vehicle-mounted driver according to the identified steering wheel, and the violation area comprises at least one of the following items: the region from the shoulder to the thigh of the vehicle-mounted driver, the head region of the driver and a steering wheel;
the violation detection module is used for detecting the violation of the vehicle-mounted driver by using preset characteristics appearing in the determined violation area, wherein the preset characteristics comprise at least one of the following characteristics: hand texture features, mouth region texture features, straight line segment features, the violation including at least one of: seat belt violation, call receiving and making violation, smoking violation;
the violation detection module is specifically configured to:
detecting an incomplete profile contained in the steering wheel profile;
a first connecting line connecting the starting point of the incomplete contour with the center of a circle, and a second connecting line connecting the ending point of the incomplete contour with the center of a circle;
if the angle between the first and second links exceeds a preset angle threshold, the incomplete contour is classified by a pattern recognition method to determine whether a cigarette is present in the steering wheel.
16. The apparatus for detecting an in-vehicle driver violation according to claim 15, wherein said steering wheel identification module comprises: an edge detection unit, a contour detection unit, and a steering wheel acquisition unit, wherein,
the edge detection unit is used for carrying out edge detection on the image to obtain an edge detection image;
the contour detection unit is used for detecting the contour of the steering wheel by utilizing a probability Hough transformation algorithm in the edge detection image;
and the steering wheel acquiring unit is used for acquiring a steering wheel internally tangent to the profile of the steering wheel.
17. The apparatus for detecting an in-vehicle driver violation according to claim 16, wherein said edge detection unit comprises: a conversion subunit and an edge detection subunit, wherein,
the converting subunit is used for converting the image into a gray-scale image;
and the edge detection subunit is used for processing the gray-scale image by using a Canny edge detection algorithm to obtain an edge detection image.
18. The apparatus for detecting an in-vehicle driver violation according to claim 16, wherein said contour detection unit comprises: a circle center determining subunit, a radius determining subunit, and a contour determining unit, wherein,
the circle center determining subunit is used for extracting the circular arc in the edge detection image, detecting the extracted circular arc by using a probability Hough transformation algorithm, projecting the circular arc into the same Hough space, counting the voting number in the Hough space, and selecting the point with the largest voting number as a circle center;
the radius determining subunit is used for calculating the distance from the circle center to the arc to obtain a radius;
and the contour determining unit is used for determining the radius of the steering wheel if the calculated radius meets a preset steering wheel radius threshold, wherein the circle center is the circle center of the steering wheel, and the radius is the radius of the steering wheel.
19. The apparatus for detecting vehicle-mounted driver violation according to claim 15, wherein said violation area determination module comprises:
the first shifting unit is used for translating the identified steering wheel or the circumscribed rectangle of the steering wheel upwards by a preset first distance threshold value to obtain the area from the shoulder to the thigh of the vehicle-mounted driver.
20. The apparatus for detecting vehicle-mounted driver violation according to claim 19, wherein said violation detection module comprises: a seat belt detection region determination unit and a seat belt detection unit, wherein,
the safety belt detection area determining unit is used for identifying an area from the shoulder part to the thigh part of the vehicle-mounted driver and determining a safety belt detection area according to the identified area from the shoulder part to the thigh part of the vehicle-mounted driver;
and a seat belt detection unit for detecting whether the vehicle-mounted driver fastens the seat belt at the determined seat belt detection region.
21. The apparatus for detecting an in-vehicle driver violation according to claim 20, wherein said seat belt detection unit comprises: a threshold value obtaining subunit, an edge detecting subunit, a straight line segment detecting subunit and a safety belt detecting subunit, wherein,
the threshold acquisition subunit is used for acquiring a self-adaptive Canny threshold in the determined safety belt detection area according to the image brightness;
the edge detection subunit is used for carrying out Canny edge detection on the safety belt detection area according to the obtained Canny threshold value to obtain an edge image in the safety belt detection area;
the straight line segment detection subunit is used for detecting a straight line segment in the edge image in the safety belt detection area by adopting a probability Hough transformation algorithm in the obtained edge image in the safety belt detection area;
the safety belt detection subunit is used for calculating the slope of the straight line segments and the distance between every two straight line segments according to the detected straight line segments, and determining that a vehicle-mounted driver fastens a safety belt if the slopes of the two straight line segments are approximately the same and the distance between the two straight line segments is smaller than a preset safety belt distance threshold; and if the slopes of the two straight line segments are not approximately the same and the distance between the two straight line segments is not smaller than a preset safety belt distance threshold value, determining that the vehicle-mounted driver does not fasten the safety belt.
22. The apparatus for detecting vehicle-mounted driver violation according to claim 15, wherein said violation area determination module comprises:
the second shifting unit is used for reducing the identified steering wheel according to a preset proportion to obtain a reduced steering wheel, and translating the reduced steering wheel upwards by a preset second distance threshold to obtain the head area of the vehicle-mounted driver; or the external rectangle is used for reducing the recognized external rectangle of the steering wheel according to a preset proportion to obtain a reduced external rectangle, and the reduced external rectangle is translated upwards by a preset second distance threshold to obtain the head area of the vehicle-mounted driver.
23. The apparatus for detecting vehicle-mounted driver violation according to claim 22, wherein said violation detection module comprises: a head region recognition unit, and a mobile phone detection unit, wherein,
the head area identification unit is used for identifying a steering wheel and determining the head area of the vehicle-mounted driver according to the identified steering wheel;
and the mobile phone detection unit is used for detecting whether a mobile phone exists in the determined head area of the vehicle-mounted driver.
24. The apparatus for detecting vehicle-mounted driver violation according to claim 23, wherein said mobile phone detection unit comprises: a hand texture feature extraction subunit and a mobile phone detection subunit, wherein,
the hand texture feature extraction subunit is used for extracting hand texture features contained in a head area of the vehicle-mounted driver according to a preset hand texture feature library;
and the mobile phone detection subunit is used for inputting the extracted hand texture features into a preset classification model, confirming that the vehicle-mounted driver makes a call if the output result of the classification model shows that the hand features exist, and confirming that the vehicle-mounted driver does not make a call if the output result of the classification model shows that the hand features do not exist.
25. The apparatus for detecting vehicle-mounted driver violation according to claim 22, wherein said violation detection module comprises: a head region identification unit, a smoking detection region determination subunit, and a smoking detection unit, wherein,
the head area identification unit is used for identifying a steering wheel and determining the head area of the vehicle-mounted driver according to the identified steering wheel;
the smoking detection area determining subunit is used for determining a smoking detection area according to the steering wheel and the head area of the vehicle-mounted driver;
and the smoking detection unit is used for detecting whether smoking phenomenon exists in the determined smoking detection area.
26. The apparatus for detecting an in-vehicle driver violation according to claim 25, wherein said smoking detection unit comprises: a discontinuous detection subunit, a connection subunit, an extraction subunit and a smoking detection subunit, wherein,
a discontinuity detection subunit configured to detect an incomplete contour included in the steering wheel contour;
the connecting subunit is used for connecting a first connecting line between the starting point of the incomplete contour and the circle center and connecting a second connecting line between the ending point of the incomplete contour and the circle center;
the extraction subunit is used for extracting hand texture features containing cigarettes from the incomplete outline if the angle between the first connecting line and the second connecting line exceeds a preset angle threshold;
and the smoking detection subunit is used for inputting the extracted hand texture features containing the cigarettes into a preset smoking distinguishing and classifying model, confirming that the vehicle-mounted driver has smoking phenomena if the output result of the smoking distinguishing and classifying model shows that the cigarettes exist in the hand features, and confirming that the vehicle-mounted driver does not have smoking phenomena if the output result of the smoking distinguishing and classifying model shows that the cigarettes do not exist in the hand features.
27. The apparatus for detecting an in-vehicle driver violation according to claim 25, wherein said smoking detection unit comprises: a mouth determining subunit and a smoking determining subunit, wherein,
a mouth determination subunit operable to extract a mouth region in a head region of the vehicle-mounted driver;
and the smoking determining subunit is used for extracting the texture features of the mouth area, confirming that the vehicle-mounted driver smokes if the extracted texture features comprise the elongated rectangular area, and confirming that the vehicle-mounted driver does not smoke if the extracted texture features do not comprise the elongated rectangular area.
28. The apparatus for detecting an in-vehicle driver violation according to any of claims 15-27, further comprising: a vehicle-mounted driver information acquisition module and a violation processing module, wherein,
the vehicle-mounted driver information acquisition module is used for identifying a head image in the shot image, matching the head image with a preset head image library to obtain a vehicle-mounted driver head image matched with the identified head image, and obtaining vehicle-mounted driver information corresponding to the identified head image according to the mapping relation between the vehicle-mounted driver head image and the vehicle-mounted driver information in the head image library;
and the violation behavior processing module is used for prompting and/or recording violation behaviors of the vehicle-mounted driver corresponding to the vehicle-mounted driver information according to the obtained vehicle-mounted driver information.
29. An electronic device, characterized in that the electronic device comprises: the device comprises a shell, a processor, a memory, a circuit board and a power circuit, wherein the circuit board is arranged in a space enclosed by the shell, and the processor and the memory are arranged on the circuit board; a power supply circuit for supplying power to each circuit or device of the electronic apparatus; the memory is used for storing executable program codes; the processor executes a program corresponding to the executable program code by reading the executable program code stored in the memory, and is used for executing the method for detecting the vehicle-mounted driver violation behavior according to any one of the preceding claims 1-14.
CN201610540740.3A 2016-06-23 2016-07-06 Method and device for detecting violation behavior of vehicle-mounted driver and electronic equipment Active CN107545225B (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN2016104865832 2016-06-23
CN201610486583 2016-06-23

Publications (2)

Publication Number Publication Date
CN107545225A CN107545225A (en) 2018-01-05
CN107545225B true CN107545225B (en) 2021-07-02

Family

ID=60966592

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610540740.3A Active CN107545225B (en) 2016-06-23 2016-07-06 Method and device for detecting violation behavior of vehicle-mounted driver and electronic equipment

Country Status (1)

Country Link
CN (1) CN107545225B (en)

Families Citing this family (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108319919B (en) * 2018-02-05 2022-10-21 安徽华信电动科技股份有限公司 Social public health record receiving information system and method based on action recognition
CN110163037B (en) * 2018-03-14 2022-03-04 北京航空航天大学 Method, device, system, processor and storage medium for monitoring driver state
CN108564034A (en) * 2018-04-13 2018-09-21 湖北文理学院 The detection method of operating handset behavior in a kind of driver drives vehicle
CN109145843A (en) * 2018-08-29 2019-01-04 上海萃舟智能科技有限公司 A kind of full vehicle information identification system of bayonet high definition camera and method
CN110046560B (en) * 2019-03-28 2021-11-23 青岛小鸟看看科技有限公司 Dangerous driving behavior detection method and camera
CN110503831A (en) * 2019-08-28 2019-11-26 上海眼控科技股份有限公司 A kind of method and apparatus identifying driver's illegal activities
CN110723621B (en) * 2019-10-11 2021-09-17 浙江新再灵科技股份有限公司 Device and method for detecting smoking in elevator car based on deep neural network
CN113033239B (en) * 2019-12-09 2023-07-07 杭州海康威视数字技术股份有限公司 Behavior detection method and device
CN113449581A (en) * 2020-03-24 2021-09-28 杭州海康威视数字技术股份有限公司 Target area detection method and device and electronic equipment
CN111753701B (en) * 2020-06-18 2023-08-15 百度在线网络技术(北京)有限公司 Method, device, equipment and readable storage medium for detecting violation of application program
CN111860280A (en) * 2020-07-15 2020-10-30 南通大学 Deep learning-based driver violation behavior recognition system
CN113053127B (en) * 2020-11-26 2021-11-26 江苏奥都智能科技有限公司 Intelligent real-time state detection system and method
CN112487990A (en) * 2020-12-02 2021-03-12 重庆邮电大学 DSP-based driver call-making behavior detection method and system

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102567703A (en) * 2010-12-10 2012-07-11 上海上大海润信息系统有限公司 Hand motion identification information processing method based on classification characteristic
CN103870806A (en) * 2014-02-21 2014-06-18 杭州奥视图像技术有限公司 Safety belt detection method combining with steering wheel detection
CN104112141A (en) * 2014-06-29 2014-10-22 中南大学 Method for detecting lorry safety belt hanging state based on road monitoring equipment

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103366506A (en) * 2013-06-27 2013-10-23 北京理工大学 Device and method for automatically monitoring telephone call behavior of driver when driving
CN104598934B (en) * 2014-12-17 2018-09-18 安徽清新互联信息科技有限公司 A kind of driver's cigarette smoking monitoring method

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102567703A (en) * 2010-12-10 2012-07-11 上海上大海润信息系统有限公司 Hand motion identification information processing method based on classification characteristic
CN103870806A (en) * 2014-02-21 2014-06-18 杭州奥视图像技术有限公司 Safety belt detection method combining with steering wheel detection
CN104112141A (en) * 2014-06-29 2014-10-22 中南大学 Method for detecting lorry safety belt hanging state based on road monitoring equipment

Also Published As

Publication number Publication date
CN107545225A (en) 2018-01-05

Similar Documents

Publication Publication Date Title
CN107545225B (en) Method and device for detecting violation behavior of vehicle-mounted driver and electronic equipment
CN106682602B (en) Driver behavior identification method and terminal
CN109460699B (en) Driver safety belt wearing identification method based on deep learning
CN105844128B (en) Identity recognition method and device
CN110858295B (en) Traffic police gesture recognition method and device, vehicle control unit and storage medium
US9721173B2 (en) Machine learning approach for detecting mobile phone usage by a driver
US20150161796A1 (en) Method and device for recognizing pedestrian and vehicle supporting the same
CN110826370B (en) Method and device for identifying identity of person in vehicle, vehicle and storage medium
CN111274881A (en) Driving safety monitoring method and device, computer equipment and storage medium
CN106650567B (en) Safety belt detection method and device
CN107798688B (en) Moving target identification method, early warning method and automobile rear-end collision prevention early warning device
CN102902957A (en) Video-stream-based automatic license plate recognition method
CN111178272A (en) Method, device and equipment for identifying driver behavior
CN111931642A (en) Safety belt wearing detection method and device, electronic equipment and storage medium
CN114373189A (en) Behavior detection method and apparatus, terminal device and storage medium
CN114202957A (en) Parking management method, management device, and storage medium
CN103854293A (en) Vehicle tracking method and device based on feature point matching
CN112818839A (en) Method, device, equipment and medium for identifying violation behaviors of driver
CN113283296A (en) Helmet wearing detection method, electronic device and storage medium
CN111325058B (en) Driving behavior detection method, device, system and storage medium
CN110634120B (en) Vehicle damage judgment method and device
CN107891807B (en) Automobile alarm method and vehicle-mounted alarm system
CN114170585B (en) Dangerous driving behavior recognition method and device, electronic equipment and storage medium
CN112633387A (en) Safety reminding method, device, equipment, system and storage medium
CN112528923A (en) Video analysis method and device, electronic equipment and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant