CN110782692A - Signal lamp fault detection method and system - Google Patents
Signal lamp fault detection method and system Download PDFInfo
- Publication number
- CN110782692A CN110782692A CN201911049565.8A CN201911049565A CN110782692A CN 110782692 A CN110782692 A CN 110782692A CN 201911049565 A CN201911049565 A CN 201911049565A CN 110782692 A CN110782692 A CN 110782692A
- Authority
- CN
- China
- Prior art keywords
- signal lamp
- fault
- lamp
- determining
- image
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/097—Supervising of traffic control systems, e.g. by giving an alarm if two crossing streets have green light simultaneously
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/09—Recognition of logos
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Traffic Control Systems (AREA)
Abstract
The embodiment of the application provides a signal lamp fault detection method and a system, comprising the following steps: receiving a fault detection instruction sent by user equipment, wherein the fault detection instruction carries identification information of a signal lamp; acquiring a monitoring video of the signal lamp from the electronic monitoring platform in real time according to the fault detection instruction; extracting a frame of monitoring image from the monitoring video every first preset time, and identifying the characteristic information of a signal lamp in the monitoring image through a preset image identification model; determining the fault type of the signal lamp according to the characteristic information; and sending fault information of the signal lamp to the user equipment, wherein the fault information comprises a fault type. The method and the system have the advantages of large fault detection coverage range and high accuracy, and can find and report the fault of the signal lamp for maintenance, thereby avoiding traffic disorder and even traffic accidents caused by untimely maintenance.
Description
Technical Field
The application relates to the technical field of image recognition, in particular to a signal lamp fault detection method and system.
Background
In road traffic, the traffic signal lamp plays an important role in maintaining the traffic order at road intersections and ensuring the traveling safety of pedestrians and vehicles. If the signal lamp fault is not found in time, traffic confusion and even traffic accidents caused by untimely signal lamp maintenance can occur.
For each intersection, because the signal lamps on each branch are controlled by the same signal machine, the current common signal lamp fault detection method is to detect the voltage value of the control channel of each signal lamp in the signal machine, and if the voltage value is not within the preset voltage range, the signal lamp controlled by the control channel is considered to be possibly extinguished.
However, the above method can only detect whether the signal lamp is turned off, but cannot detect faults such as the signal lamp being blocked by foreign matter, the traffic lights being simultaneously lit, the position of the signal lamp being shifted, the red light or the green light being lit for a long time, and the coverage of fault detection is small. In addition, the fault detection method is limited by the structural defects that the construction of signal lamps is not standard (for example, the signal lamps of all fork roads are controlled by one control channel), and the like, and the detection accuracy is low. Therefore, it is desirable to provide a signal lamp fault detection method with wide coverage and high detection precision.
Disclosure of Invention
The application provides a signal lamp fault detection method and system, which are used for solving the problems of small fault detection coverage range and low detection accuracy rate in the prior art.
In a first aspect, the present embodiment provides a signal lamp fault detection method, which is applied to a signal lamp fault detection system, and the method includes:
acquiring a monitoring video of the signal lamp from an electronic monitoring platform in real time according to the identification information of the signal lamp;
extracting a frame of monitoring image from the monitoring video every interval of first preset time, and identifying the characteristic information of a signal lamp in the monitoring image through a preset image identification model;
and determining the fault type of the signal lamp according to the characteristic information.
In a first implementation manner of the first aspect, identifying, by a preset image recognition model, feature information of a signal lamp in the monitoring image includes:
determining an identification area of the monitoring image according to a signal lamp area calibrated in advance in a reference image of the signal lamp; and determining the characteristic information of the signal lamp according to the image in the identification area.
In a second implementation manner of the first aspect, the determining the fault type of the signal lamp according to the characteristic information includes: determining a lighting fault according to the color of the lamp holder; or determining the lamp body offset fault according to the position information of the lamp panel; or determining abnormal faults in the lamp according to the color of the lamp holder; or determining to release the conflict fault according to the color of the lamp holder.
In a third implementation manner of the first aspect, determining a lighting fault of the signal lamp according to a color of a lamp holder specifically includes:
and in a second preset time, if the color of the lamp head of each sub-signal lamp in the signal lamp is red and green simultaneously, or red and yellow, or yellow and green, or red, green and yellow, or is continuously yellow, or the color of the lamp head cannot be identified, determining that the signal lamp has a lighting fault.
In a fourth implementation manner of the first aspect, determining a lamp body offset fault of the signal lamp according to the position information of the lamp panel specifically includes:
and comparing the position information with a reference position of the signal lamp in a reference image, and determining that the signal lamp has a lamp body deviation fault if the difference value between the position information and the reference position exceeds a preset range.
In a fifth implementation manner of the first aspect, determining an abnormal fault in the lamp according to the color of the lamp holder specifically includes:
for each sub-signal lamp in the signal lamp, judging whether a green light time period exists in the sub-signal lamp within a third preset time range according to the characteristic information within the third preset time range, wherein the green light time period comprises: a left opening and right opening time period, a left opening and right closing time period, a left closing and right opening time period and a left closing and right closing time period;
if the green light time period does not exist within a third preset time range, determining that the sub signal lamp has an abnormal fault when the sub signal lamp has a light;
if a left-close time period and a right-close time period exist and the green light time period is not between the preset time period lower limit and the preset time period upper limit, determining that the sub signal lamp has an abnormal fault when the sub signal lamp has the light;
if the non-left-off and right-off time period in the green light time period exists, determining green light time in the non-left-off and right-off time period, and if the green light time is greater than the upper limit of the time, determining that the sub-signal lamp has an abnormal fault when the sub-signal lamp exists;
the left-on state means that the minimum time of the green light time period is the initial time point of the current fault judgment cycle; the left closing means that the color of the lamp holder of the previous frame of the monitoring image at the minimum moment is non-green; the right opening means that the maximum time of the green light time period is the end time point of the fault judgment period; and the right closing means that the color of the lamp holder of the next frame of monitoring image at the maximum moment is non-green.
In a sixth implementation manner of the first aspect, determining to release the conflict fault according to the color of the lamp holder specifically includes:
determining the lamp head color of each sub-signal lamp of the signal lamp on each branch road of the road junction within a fourth preset time range; and if the colors of the lamp heads of the same sub-signal lamps on the adjacent branches are the same within the fourth preset time range, determining that the signal lamps have the passing conflict fault.
In a seventh implementation manner of the first aspect, acquiring, according to the identification information of the signal lamp, the monitoring video of the signal lamp from the electronic monitoring platform in real time includes:
and if the signal lamp has preset working time and the current moment is within the working time range, acquiring the monitoring video of the signal lamp from the electronic monitoring platform in real time according to the identification information of the signal lamp.
In an eighth implementation form of the first aspect, the method further includes: and sending the fault information to an operation and maintenance management platform, wherein the operation and maintenance management platform is used for managing and maintaining the signal lamp, and the fault information comprises a fault type and fault related information.
In a ninth implementation form of the first aspect, the image recognition model is determined by:
acquiring a plurality of sample images of different environmental parameters from the electronic monitoring platform, and labeling characteristic information of a signal lamp in each sample image, wherein each sample image comprises an image of the signal lamp;
forming a training sample library by the labeled sample images;
and generating an image recognition model according to a deep learning algorithm, and training the image recognition model to recognize the characteristic information of the signal lamp according to the training sample library.
In a second aspect, the present embodiment provides a traffic light fault detection system, including a fault detection server and a fault determination server, wherein,
the fault detection server is used for acquiring a monitoring video of the signal lamp from the electronic monitoring platform in real time according to the identification information of the signal lamp; extracting a frame of monitoring image from the monitoring video every interval of first preset time, and identifying the characteristic information of a signal lamp in the monitoring image through a preset image identification model;
and the fault judgment server is used for determining the fault type of the signal lamp according to the characteristic information.
The technical scheme provided by the application comprises the following beneficial technical effects:
according to the signal lamp fault detection method and system provided by the embodiment of the application, the monitoring video acquired by the electronic monitoring platform is further processed by using an image recognition technology, the characteristic information of the signal lamp is determined, and the fault type of the signal lamp is judged according to the characteristic information. The method and the system can detect the lighting fault of the signal lamp, can also detect the lamp body offset fault, the abnormal fault during the lamp and the conflict fault, and have the advantages of large coverage range of fault detection and high accuracy.
In addition, the fault detection method provided by the embodiment can detect the fault of the signal lamp in real time, and timely discover and report the fault, so that maintenance personnel can timely maintain the fault signal lamp, and traffic confusion and even traffic accidents caused by untimely maintenance are avoided.
Drawings
In order to more clearly explain the technical solution of the present application, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious to those skilled in the art that other drawings can be obtained according to the drawings without any creative effort.
Fig. 1 is a flowchart illustrating a method for determining an image recognition model according to the present embodiment;
fig. 2 is a flowchart of a signal lamp fault detection method shown in the present embodiment;
fig. 3 is a frame of monitoring image shown in the present embodiment;
fig. 4 is a reference image shown in the present embodiment;
fig. 5 is a schematic structural diagram of the signal lamp fault detection system shown in this embodiment;
fig. 6 is a schematic view of an application scenario of the signal lamp fault detection system shown in this embodiment.
Detailed Description
To make the objects, technical solutions and advantages of the exemplary embodiments of the present application clearer, the technical solutions in the exemplary embodiments of the present application will be clearly and completely described below with reference to the drawings in the exemplary embodiments of the present application, and it is obvious that the described exemplary embodiments are only a part of the embodiments of the present application, but not all the embodiments.
At present, in order to facilitate traffic management and maintain traffic order, electronic monitoring devices are arranged on each branch road of most of road junctions (such as three-branch road junctions, crossroads, five-branch road junctions and the like) and used for collecting monitoring videos and sending the monitoring videos to an electronic monitoring platform for unified management. The monitoring video collected by the electronic monitoring device generally includes images of signal lamps to monitor the movement of vehicles and pedestrians at each intersection under the indication of the corresponding signal lamps.
Based on this, the present embodiment provides a method and a system for detecting a fault of a signal lamp, which are used to detect a working condition of the signal lamp at an intersection in real time according to an image of the signal lamp in a monitoring video.
The operation of the signal lamp fault detection method and system provided by the embodiment all rely on an image recognition model to recognize the characteristic information of the signal lamp in the monitoring image, such as the position information of the lamp panel, the type of the lamp head, the color of the lamp head and the like. Therefore, the embodiment of the present application first describes the image recognition model in detail.
Referring to fig. 1, an image recognition model provided in the embodiment of the present application is determined through the following steps S101 to S103.
Step S101, obtaining a plurality of sample images of different environmental parameters from an electronic monitoring platform, and labeling characteristic information of a signal lamp in each sample image, wherein each sample image comprises an image of the signal lamp.
For example, a plurality of monitoring videos with different environmental parameters (such as sunny days, rainy days, foggy days, morning, evening, and the like) can be obtained from the electronic monitoring platform, and are subjected to preprocessing such as video coding and decoding, image denoising, enhancement, and the like, and some monitoring images are extracted from the videos to be used as sample images. Since the signal lights at each intersection may be different, for example, the shape of the signal light panel may be horizontal signal lights, vertical signal lights, the head of the signal light may be circular, arrow-shaped, and the like, and thus, there are various patterns of the signal lights in these sample images.
After the sample images are determined, labels can be manually marked in the sample images, wherein the labels include characteristic information of signal lamps in the sample images, and the characteristic information includes: position information (such as the position of a pixel point of a signal lamp in the whole monitored image), lamp panel types (such as horizontally arranged signal lamps and vertically arranged signal lamps), lamp holder types (such as round lamps and arrow lamps), lamp holder colors (such as red, green and yellow) and the like.
And S102, forming a training sample library by the labeled sample images.
And S103, generating an image recognition model according to a deep learning algorithm, and training the image recognition model to recognize the characteristic information of the signal lamp according to a training sample library.
For example, a Spatial Regularization Network (SRN) may be employed to learn the relationships between multiple labels of a sample image. Because the shape of the signal lamp is specified in national standards, the SRN learns the semantic and spatial relationship among labels only by using image-level supervision information, and achieves the aim of identifying the characteristic information of the signal lamp.
An image recognition model for recognizing the characteristic information of the signal lamp in the image can be obtained through the above steps S101 to S103.
Referring to fig. 2, an embodiment of the present application provides a method for detecting a failure of a traffic light, which is applied to a system for detecting a failure of a traffic light, and includes the following steps S201 to S205.
Step S201, the fault detection system receives a fault detection instruction sent by the user equipment, where the fault detection instruction carries identification information of a signal lamp.
The fault detection system provided by the embodiment can be accessed to the user equipment and executes corresponding operation according to the instruction of the user equipment. For example, a Personal Computer (PC) client may be disposed on the user equipment, and the PC client is used as a tool for human-computer interaction, and is capable of controlling the PC to generate a fault detection instruction according to a user instruction and sending the fault detection instruction to the fault detection system.
The fault detection instruction carries identification information of the signal lamp and can also carry a fault type to be detected, and the instruction is used for indicating a fault detection system to execute corresponding fault detection operation according to the identification information and the fault detection type. The identification information of the signal lamp is used for uniquely indicating the signal lamp, and may be a number, a name, an address, and the like of the signal lamp. The failure type includes at least one of a light-on failure, a lamp body shift failure, an abnormal failure at the time of lighting, and a release conflict failure.
It should be noted that, in one fault detection instruction, the number of signal lamps may be one or multiple, for example, all the traffic signal lamps on a certain line, all the traffic signal lamps in a certain area, and the like.
Step S202, the fault detection system acquires the monitoring video of the signal lamp from the electronic monitoring platform in real time according to the fault detection instruction.
Since the electronic monitoring platform stores monitoring videos shot by different electronic monitoring devices, and each monitoring video usually includes an image of a signal lamp at a specific branch, there is a one-to-one correspondence between the monitoring videos and the signal lamps. Therefore, the fault detection system can acquire the monitoring video of the signal lamp from the electronic monitoring platform in real time according to the identification information of the signal lamp and the corresponding relation between the monitoring video and the signal lamp.
At present, most of the signal lamps work continuously for 24 hours, but a small part of the signal lamps only work within a preset working time, for example, the signal lamps at school doorways can only work at the peak time of the people flow for students to school and school, such as the time periods of 7:30am-7:45am, 12:00pm-12:15pm, 13:50pm-14:00pm, 17:30pm-17:40pm and the like. Therefore, as an optional implementation manner, when the fault detection system acquires the monitoring video of the signal lamp, it may first determine whether the signal lamp is preset with a working time, and if the current time is within a preset working time range, acquire the monitoring video corresponding to the signal lamp from the electronic monitoring platform.
Step S203, the fault detection system extracts a frame of monitoring image from the monitoring video every first preset time, and identifies the characteristic information of the signal lamp in the monitoring image through a preset image identification model.
At present, a signal lamp (for example, the signal lamp in fig. 3) for controlling the passage of vehicles is generally provided with three sub-signal lamps including a left turn signal lamp, a straight going signal lamp and a right turn signal lamp. Each sub signal lamp is provided with a red lamp, a yellow lamp and a green lamp. Because the monitoring video comprises a plurality of frames of monitoring images, the fault detection system extracts one frame of monitoring image from the monitoring video at intervals of first preset time (such as 0.5S, 1S, 2S and the like), and identifies the characteristic information of the signal lamp in the monitoring image through a preset image identification model, wherein the characteristic information comprises the lamp head color of each sub-signal lamp.
Through a preset image recognition model, pixel-level segmentation can be carried out on a monitored image, the position of a signal lamp is accurately determined, feature extraction is carried out on the positioned position of the signal lamp, whether the signal lamp is the same target or not is confirmed through comparison of partial features of the target, and target detection is completed.
Taking the monitoring image exemplarily shown in fig. 4 as an example, in the monitoring image, the sub-signal lights from left to right are a left turn signal light, a straight going signal light, and a right turn signal light in this order. And the green light of the left turn signal lamp is on, the green light of the straight going signal lamp is on, and the red light of the right turn signal lamp is on. The position information of the signal lamp in the monitoring image can be determined to be (x, y) through an image recognition model, wherein x belongs to [ x ∈ [ [ x ]
1，x
2]，y∈[y
1，y
2]. The lamp panel type of the left turn signal lamp is a longitudinal installation type; the lamp holder is in the shape of an arrow, and the color of the lamp holder is green. The lamp panel type of the straight signal lamp is a longitudinal installation type, the lamp holder type is an arrow head shape, and the color of the lamp holder is green. The lamp panel type of the right turn signal lamp is a longitudinal installation type, the lamp holder type is an arrow head shape, and the color of the lamp holder is red.
As an alternative embodiment, each signal lamp has a preset reference image, and each reference image is labeled with a signal lamp region in advance. For example, in the reference image shown in fig. 4, the area selected by the black frame is the pre-marked signal light area. These reference images are stored in a reference image library of the fault detection system to be called by the fault detection system during image recognition.
When the image recognition model recognizes the characteristic information of the signal lamp in the monitored image, the signal lamp area pre-marked in the reference image of the signal lamp can be determined as the recognition area of the monitored image, the image in the recognition area is recognized, and the characteristic information of the signal lamp in the monitored image is determined, so that the recognition efficiency of the image is improved. And if the characteristic information of the signal lamp cannot be determined according to the identification area, identifying the whole monitoring image.
It should be noted that the reference image and the monitoring image have the same shooting conditions, including the same electronic monitoring device and the same shooting angle. The reference image is different from the monitoring image in that the monitoring image may be taken in different weather or time, such as rainy day, foggy day, night, etc., and the picture thereof may be unclear, whereas the reference image is taken in a good-sight condition and the picture thereof is clear.
And step S204, the fault detection system determines the fault type of the signal lamp according to the characteristic information.
The fault detection method provided by the embodiment can determine different types of faults such as a lighting fault, a lamp body deviation fault, an abnormal fault during lamp, a release conflict fault and the like according to the characteristic information. The determination of different fault types may use different characteristic information, and the determination process of different types of faults will be described in detail below.
(1) Determination of lighting failure
And the fault detection system determines the lighting fault of the signal lamp according to the color of the lamp holder in the characteristic information. Specifically, within a second preset time (e.g., 5S), if the color of the lamp head of any one of the sub-signal lamps (e.g., the left turn signal lamp, the straight-going signal lamp, or the right turn signal lamp shown in fig. 3) in the monitored image shows red and green at the same time, or red and yellow, or yellow and green, or red, green and yellow, or continues to be yellow (i.e., when the lighting time of the yellow lamp exceeds the preset yellow lamp, e.g., 3S), or the color of the lamp head cannot be identified, it is determined that there is a lighting fault in the signal lamp. The condition that the color of the lamp holder cannot be identified can further indicate that the signal lamp has extinguishment, shielding or offset faults.
(2) Determination of lamp body deflection fault
And the fault detection system determines the lamp body offset fault of the signal lamp according to the position information of the lamp panel in the characteristic information.
Specifically, the position information (x, y) of the traffic light in the monitored image is compared with the reference position (m, n) of the traffic light in the reference image, where m ∈ [ m ] m
1，m
2]，n∈[n
1，n
2]. If the difference between the position information and the reference position exceeds a predetermined range, e.g. | x
1-m
1| is ≧ Δ h, or | y
1-n
1And if | ≧ Δ h, determining that the lamp body offset fault exists in the signal lamp.
(3) Determination of abnormal lamp failure
And the fault detection system determines abnormal faults in the lamp according to the color of the lamp holder in the characteristic information.
In the fault determination process, different sub-signal lamps, such as a left turn signal lamp, a straight traveling signal lamp and a right turn signal lamp, are pre-configured with corresponding upper time length limits and lower time length limits. In a light period of the signal, each of the sub-signal lights comprises at least one of a green light time and a red light time, and may further comprise a yellow light time. The embodiment determines whether the monitored signal lamp has the abnormal fault when the lamp exists by determining whether the abnormal condition exists when the green lamp of each sub-signal lamp exists.
The fault detection system extracts a frame of monitoring image every first preset time interval, identifies the characteristic information of the signal lamp in the monitoring image and stores the characteristic information in a message queue mode. When the fault detection system judges the abnormal fault of the lamp, all the characteristic information of the signal lamp in a third preset time range (for example, 3min, 5min and the like) can be acquired, and whether the abnormal fault of the lamp exists in the signal lamp or not can be judged according to the characteristic information by taking the third preset time range as a fault detection period.
In the process of judging the abnormal fault during lighting, for each sub-signal lamp, whether a green lamp time period exists in the fault detection period is judged according to the color of the lamp holder in the fault detection period. Wherein, the green light time quantum is the time quantum that the lamp holder colour is green, includes: left-open right-open time period, left-open right-close time period, left-close right-open time period and left-close right-close time period. In this embodiment, the minimum time of the green light period is referred to as the left end, and the maximum time of the green light period is referred to as the right end. The left opening means that the left end is the initial time point of the current fault judgment cycle. Left closed means that the color of the lamp head of the previous monitoring image at the left end is non-green (such as red or yellow). The right opening means that the right end is the end time point of the current fault judgment cycle. And the right closed state means that the color of the lamp holder of the next monitoring image at the right end is non-green (such as red or yellow).
Secondly, if the green light time period does not exist in the fault detection period, the abnormal fault when the signal lamp has the light is determined. Taking the fault detection cycle as 3min as an example, if no green light time period exists within 3min, it is indicated that the green light of the sub-signal lamp is always in the off state in the time period, and the signal lamp has an abnormal fault when the lamp exists.
In the green light time period, only the left closing time period and the right closing time period are a complete green light time period, and the rest are incomplete green light time periods.
And if a left closing time period and a right closing time period exist in the fault detection period, and the duration of the green light time period is not between the lower duration limit and the upper duration limit, determining that the sub signal lamp has an abnormal fault when the sub signal lamp has the light.
If the non-left-off and right-off time period (namely, the left-open and right-open time period, the left-open and right-close time period or the left-closed and right-open time period) in the green light time period exists, and the green light time length in the non-left-open and right-close time period is greater than the upper time limit, determining that the sub-signal lamp has an abnormal fault when the sub-signal lamp is on. Because the green time in the non-left-off and right-off time periods may not be a complete green time period, when the green time is less than the lower time limit, it cannot be determined whether there is an abnormal fault when the lamp is present.
It should be noted that, because the traffic flows at different road junctions are different and the green time lengths of the signal lamps at different road junctions are different, the lower time length limit and the upper time length limit of the green light corresponding to the sub-signal lamps at different road junctions, which are the same, may be different, and are specifically determined according to the preset configuration in the fault detection system, which is not limited in this embodiment.
(4) Determination of a clear conflict fault
And the fault detection system determines to pass the conflict fault according to the color of the lamp holder in the characteristic information.
The clear conflict faults include straight conflict, straight left conflict and left conflict. The straight collision means that two straight signal lamps on adjacent branches of the road junction are both green; the direct left conflict means that the direct traffic light and the left turn traffic light on the adjacent turnout at the road junction are both green; the left-left conflict means that two left turn signal lamps on adjacent branches of the road junction are all green.
Specifically, in the process of performing the passing conflict fault detection, the fault detection system first determines whether each branch road of the detected road junction is provided with an electronic monitoring device. If yes, in steps S202-S203, the monitoring videos at each branch of the road junction are respectively obtained, the characteristic information of the signal lamp in each monitoring video is respectively determined, and a fourth preset time range (e.g., 5S, 10S, etc.) is used as a passing conflict detection period.
Specifically, after determining the feature information on each branch in the fourth preset time range, the fault detection system determines whether the signal lamp has a release conflict fault according to the color of the lamp holder in the feature information. In one example, if the color of the lamp heads of the same sub-signal lamps on two adjacent branches is the same within a fourth preset time range, the signal lamp is determined to have the release conflict fault. For example, when the left turn signal lamps of two adjacent branches are both green, the signal lamps have a release conflict fault.
In step S205, the fault detection system sends the fault information of the signal lamp to the user equipment, where the fault information includes a fault type.
In the present embodiment, the failure information includes a failure type and failure-related information. The fault related information may include a detection task description, a fault photo, a fault code, a fault intersection, an electronic monitoring device corresponding to a fault signal lamp, shooting time of a monitoring video corresponding to a fault, and the like.
After determining the failure information, the failure detection system sends it to the user equipment, and stores it in a local database, for example, a local FTP (File Transfer Protocol) server.
After receiving the fault information, the user device may display the fault information in a GIS (geographic information System) map, and/or a fault list for the user to view. Meanwhile, after the user checks the fault information, the user can send a detection stopping instruction of the fault signal lamp to the fault detection system through the PC client of the user equipment.
As an optional implementation manner, the fault detection system can also send the fault information to an operation and maintenance management platform, and the operation and maintenance management platform is used for managing and maintaining the signal lamp, so that the operation and maintenance manager can timely maintain the fault signal lamp.
In addition, the user equipment can also send a fault false alarm message of the signal lamp to the operation and maintenance management platform through the PC client, or carry out one-key guarantee operation and send fault information to the operation and maintenance management platform.
According to the signal lamp fault detection method provided by the embodiment of the application, the monitoring video acquired by the electronic monitoring platform is further processed by utilizing an image recognition technology, the characteristic information of the signal lamp is determined, and the fault type of the signal lamp is judged according to the characteristic information. The method can detect the lighting fault of the signal lamp, can also detect the lamp body offset fault, the abnormal fault during lighting and the conflict fault, and has the advantages of large fault detection coverage range and high accuracy.
In addition, the fault detection method provided by the embodiment can detect the fault of the signal lamp in real time, and timely discover and report the fault, so that maintenance personnel can timely maintain the fault signal lamp, and traffic confusion and even traffic accidents caused by untimely maintenance are avoided.
Referring to fig. 5, a schematic structural diagram of a signal lamp fault detection system exemplarily shown in this embodiment; and fig. 6 is a schematic view of an application scenario of the fault detection system.
The signal lamp fault detection system provided by the embodiment is used for executing the signal lamp fault detection method provided by the embodiment, and comprises a fault detection server and a fault determination server. The failure detection server is configured to perform the above steps S201 to S203, and the failure determination server is configured to perform the above steps S204 to S205.
The signal lamp fault detection system provided by the embodiment of the application can detect the lighting fault of the signal lamp, can also detect the lamp body offset fault and abnormal fault and the conflict fault during lamp release, and has the advantages of large coverage range of fault detection and high accuracy. Moreover, the system can detect the fault of the signal lamp in real time, and timely discover and report the fault, so that maintenance personnel can maintain the fault signal lamp in time, and traffic disorder and even traffic accidents caused by untimely maintenance are avoided.
All other embodiments, which can be derived by a person skilled in the art from the exemplary embodiments shown in the present application without inventive effort, shall fall within the scope of protection of the present application. Moreover, while the disclosure herein has been presented in terms of exemplary one or more examples, it is to be understood that each aspect of the disclosure can be utilized independently and separately from other aspects of the disclosure to provide a complete disclosure.
It should be understood that the terms "first," "second," "third," and the like in the description and in the claims of the present application and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used are interchangeable under appropriate circumstances and can be implemented in sequences other than those illustrated or otherwise described herein with respect to the embodiments of the application, for example.
Furthermore, the terms "comprises" and "comprising," as well as any variations thereof, are intended to cover a non-exclusive inclusion, such that a product or device that comprises a list of elements is not necessarily limited to those elements explicitly listed, but may include other elements not expressly listed or inherent to such product or device.
Finally, it should be noted that: the above embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present application.
Claims (11)
1. A signal lamp fault detection method is applied to a signal lamp fault detection system, and comprises the following steps:
acquiring a monitoring video of the signal lamp from an electronic monitoring platform in real time according to the identification information of the signal lamp;
extracting a frame of monitoring image from the monitoring video every interval of first preset time, and identifying the characteristic information of a signal lamp in the monitoring image through a preset image identification model;
and determining the fault type of the signal lamp according to the characteristic information.
2. The method of claim 1, wherein the identifying the characteristic information of the signal lamp in the monitoring image through a preset image recognition model comprises:
determining an identification area of the monitoring image according to a signal lamp area calibrated in advance in a reference image of the signal lamp; and determining the characteristic information of the signal lamp according to the image in the identification area.
3. The method of claim 1 or 2, wherein the characteristic information includes information on a location of a lamp head color and a lamp panel, and wherein determining the type of failure of the signal lamp based on the characteristic information includes:
determining a lighting fault according to the color of the lamp holder; alternatively, the first and second electrodes may be,
determining a lamp body offset fault according to the position information of the lamp panel; alternatively, the first and second electrodes may be,
determining abnormal faults in the lamp according to the color of the lamp holder; alternatively, the first and second electrodes may be,
and determining the conflict fault according to the color of the lamp holder.
4. The method according to claim 3, wherein the determining of the lighting failure according to the color of the lamp holder specifically comprises:
and in a second preset time, if the color of the lamp head of each sub-signal lamp in the signal lamp is red and green simultaneously, or red and yellow, or yellow and green, or red, green and yellow, or is continuously yellow, or the color of the lamp head cannot be identified, determining that the signal lamp has a lighting fault.
5. The method of claim 3, wherein the determining the lamp body deviation fault according to the position information of the lamp panel specifically comprises:
and comparing the position information with a reference position of the signal lamp in a reference image, and determining that the signal lamp has a lamp body deviation fault if the difference value between the position information and the reference position exceeds a preset range.
6. The method according to claim 3, wherein the determining of the abnormal fault in the lamp according to the color of the lamp holder specifically comprises:
for each sub-signal lamp in the signal lamp, judging whether a green light time period exists in the sub-signal lamp within a third preset time range according to the characteristic information within the third preset time range, wherein the green light time period comprises: a left opening and right opening time period, a left opening and right closing time period, a left closing and right opening time period and a left closing and right closing time period;
if the green light time period does not exist within a third preset time range, determining that the sub signal lamp has an abnormal fault when the sub signal lamp has a light;
if a left-close time period and a right-close time period exist and the green light time period is not between the preset time period lower limit and the preset time period upper limit, determining that the sub signal lamp has an abnormal fault when the sub signal lamp has the light;
if the non-left-off and right-off time period in the green light time period exists, determining green light time in the non-left-off and right-off time period, and if the green light time is greater than the upper limit of the time, determining that the sub-signal lamp has an abnormal fault when the sub-signal lamp exists;
the left-on state means that the minimum time of the green light time period is the initial time point of the current fault judgment cycle; the left closing means that the color of the lamp holder of the previous frame of the monitoring image at the minimum moment is non-green; the right opening means that the maximum time of the green light time period is the end time point of the fault judgment period; and the right closing means that the color of the lamp holder of the next frame of monitoring image at the maximum moment is non-green.
7. The method according to claim 3, wherein the determining of the clear conflict fault according to the color of the lamp head specifically comprises:
determining the lamp head color of each sub-signal lamp of the signal lamp on each branch road of the road junction within a fourth preset time range;
and if the lamp head colors of the same sub-signal lamps on the adjacent branches are the same within the fourth preset time range, determining that the signal lamps have the passing conflict fault.
8. The method of claim 1, wherein obtaining the monitoring video of the signal lamp from the electronic monitoring platform in real time according to the identification information of the signal lamp comprises:
and if the signal lamp has preset working time and the current moment is within the working time range, acquiring the monitoring video of the signal lamp from the electronic monitoring platform in real time according to the identification information of the signal lamp.
9. The method of claim 1, further comprising:
and sending fault information to an operation and maintenance management platform, wherein the operation and maintenance management platform is used for managing and maintaining the signal lamp, and the fault information comprises fault types and fault related information.
10. The method of claim 1, wherein the image recognition model is determined by:
acquiring a plurality of sample images of different environmental parameters from the electronic monitoring platform, and labeling characteristic information of a signal lamp in each sample image, wherein each sample image comprises an image of the signal lamp;
forming a training sample library by the labeled sample images;
and generating an image recognition model according to a deep learning algorithm, and training the image recognition model to recognize the characteristic information of the signal lamp according to the training sample library.
11. A signal lamp fault detection system is characterized by comprising a fault detection server and a fault judgment server, wherein,
the failure detection server is configured to,
acquiring a monitoring video of the signal lamp from an electronic monitoring platform in real time according to the identification information of the signal lamp;
extracting a frame of monitoring image from the monitoring video every interval of first preset time, and identifying the characteristic information of a signal lamp in the monitoring image through a preset image identification model;
the failure determination server is configured to,
and determining the fault type of the signal lamp according to the characteristic information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911049565.8A CN110782692A (en) | 2019-10-31 | 2019-10-31 | Signal lamp fault detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911049565.8A CN110782692A (en) | 2019-10-31 | 2019-10-31 | Signal lamp fault detection method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110782692A true CN110782692A (en) | 2020-02-11 |
Family
ID=69388115
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911049565.8A Pending CN110782692A (en) | 2019-10-31 | 2019-10-31 | Signal lamp fault detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110782692A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111428647A (en) * | 2020-03-25 | 2020-07-17 | 浙江浙大中控信息技术有限公司 | Traffic signal lamp fault detection method |
| CN111681442A (en) * | 2020-06-09 | 2020-09-18 | 贾若然 | Signal lamp fault detection device based on image classification algorithm |
| CN111882910A (en) * | 2020-06-15 | 2020-11-03 | 太原市高远时代科技有限公司 | High-accuracy traffic signal lamp fault detection method and system |
| CN111931724A (en) * | 2020-09-23 | 2020-11-13 | 北京百度网讯科技有限公司 | Signal lamp abnormity identification method and device, electronic equipment and road side equipment |
| CN112180285A (en) * | 2020-09-23 | 2021-01-05 | 北京百度网讯科技有限公司 | Method and device for identifying faults of traffic signal lamp, navigation system and road side equipment |
| CN112528787A (en) * | 2020-12-02 | 2021-03-19 | 博云视觉科技(青岛)有限公司 | Signal lamp fault detection method based on deep learning |
| CN112634621A (en) * | 2020-12-25 | 2021-04-09 | 北京百度网讯科技有限公司 | Traffic light data quality detection method, device, electronic equipment and medium |
| CN116704720A (en) * | 2023-08-03 | 2023-09-05 | 四川富基智能科技有限公司 | Intelligent traffic violation early warning system and method |
| CN111931724B (en) * | 2020-09-23 | 2024-06-21 | 阿波罗智联(北京)科技有限公司 | Signal lamp abnormality identification method and device, electronic equipment and road side equipment |
Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101097657A (en) * | 2006-06-26 | 2008-01-02 | 上海宝信软件股份有限公司 | Traffic signal light condition judgement method based on video frequency image processing |
| CN102129779A (en) * | 2010-01-13 | 2011-07-20 | 中国科学院自动化研究所 | Monitoring system of safety and efficiency of traffic signal of intersection |
| CN103236181A (en) * | 2013-05-10 | 2013-08-07 | 西安电子科技大学 | Traffic signal lamp state monitoring system and method based on machine vision |
| CN105160924A (en) * | 2015-08-25 | 2015-12-16 | 公安部第三研究所 | Video processing-based intelligent signal lamp state detection method and detection system |
| CN105930819A (en) * | 2016-05-06 | 2016-09-07 | 西安交通大学 | System for real-time identifying urban traffic lights based on single eye vision and GPS integrated navigation system |
| CN106709511A (en) * | 2016-12-08 | 2017-05-24 | 华中师范大学 | Urban rail transit panoramic monitoring video fault detection method based on depth learning |
| CN107862883A (en) * | 2017-12-21 | 2018-03-30 | 天津市中环系统工程有限责任公司 | The fault detect and alarm of traffic lights and operation management system and implementation method |
| CN107993470A (en) * | 2016-10-26 | 2018-05-04 | 天津职业技术师范大学 | Count down traffic signal lamp condition detection method and the monitoring system based on the method |
| CN108235816A (en) * | 2018-01-10 | 2018-06-29 | 深圳前海达闼云端智能科技有限公司 | Image recognition method, system, electronic device and computer program product |
| CN108337321A (en) * | 2018-03-14 | 2018-07-27 | 卡斯柯信号有限公司 | CBTC signalling arrangements cruising inspection system and method based on video intelligent identification |
| CN108382396A (en) * | 2018-02-02 | 2018-08-10 | 辽宁友邦网络科技有限公司 | Driver's driving condition identifying system and its application process |
| CN108648495A (en) * | 2018-06-08 | 2018-10-12 | 华南理工大学 | A kind of method and system of the intelligence real-time display bus degree of crowding |
| CN109094580A (en) * | 2018-07-05 | 2018-12-28 | 湖南铁路科技职业技术学院 | Train auxiliary driving method, system and computer readable storage medium based on image recognition |
| CN109636777A (en) * | 2018-11-20 | 2019-04-16 | 广州方纬智慧大脑研究开发有限公司 | A kind of fault detection method of traffic lights, system and storage medium |
| CN109670423A (en) * | 2018-12-05 | 2019-04-23 | 依通（北京）科技有限公司 | A kind of image identification system based on deep learning, method and medium |
| CN109726692A (en) * | 2018-12-29 | 2019-05-07 | 重庆集诚汽车电子有限责任公司 | High-definition camera 3D object detection system based on deep learning |
| CN109803108A (en) * | 2019-01-22 | 2019-05-24 | 国网信息通信产业集团有限公司 | A kind of image-recognizing method and device |
| KR20190065907A (en) * | 2017-12-12 | 2019-06-12 | 유동수 | Traffic Management System using Face Detection |
-
2019
- 2019-10-31 CN CN201911049565.8A patent/CN110782692A/en active Pending
Patent Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101097657A (en) * | 2006-06-26 | 2008-01-02 | 上海宝信软件股份有限公司 | Traffic signal light condition judgement method based on video frequency image processing |
| CN102129779A (en) * | 2010-01-13 | 2011-07-20 | 中国科学院自动化研究所 | Monitoring system of safety and efficiency of traffic signal of intersection |
| CN103236181A (en) * | 2013-05-10 | 2013-08-07 | 西安电子科技大学 | Traffic signal lamp state monitoring system and method based on machine vision |
| CN105160924A (en) * | 2015-08-25 | 2015-12-16 | 公安部第三研究所 | Video processing-based intelligent signal lamp state detection method and detection system |
| CN105930819A (en) * | 2016-05-06 | 2016-09-07 | 西安交通大学 | System for real-time identifying urban traffic lights based on single eye vision and GPS integrated navigation system |
| CN107993470A (en) * | 2016-10-26 | 2018-05-04 | 天津职业技术师范大学 | Count down traffic signal lamp condition detection method and the monitoring system based on the method |
| CN106709511A (en) * | 2016-12-08 | 2017-05-24 | 华中师范大学 | Urban rail transit panoramic monitoring video fault detection method based on depth learning |
| KR20190065907A (en) * | 2017-12-12 | 2019-06-12 | 유동수 | Traffic Management System using Face Detection |
| CN107862883A (en) * | 2017-12-21 | 2018-03-30 | 天津市中环系统工程有限责任公司 | The fault detect and alarm of traffic lights and operation management system and implementation method |
| CN108235816A (en) * | 2018-01-10 | 2018-06-29 | 深圳前海达闼云端智能科技有限公司 | Image recognition method, system, electronic device and computer program product |
| CN108382396A (en) * | 2018-02-02 | 2018-08-10 | 辽宁友邦网络科技有限公司 | Driver's driving condition identifying system and its application process |
| CN108337321A (en) * | 2018-03-14 | 2018-07-27 | 卡斯柯信号有限公司 | CBTC signalling arrangements cruising inspection system and method based on video intelligent identification |
| CN108648495A (en) * | 2018-06-08 | 2018-10-12 | 华南理工大学 | A kind of method and system of the intelligence real-time display bus degree of crowding |
| CN109094580A (en) * | 2018-07-05 | 2018-12-28 | 湖南铁路科技职业技术学院 | Train auxiliary driving method, system and computer readable storage medium based on image recognition |
| CN109636777A (en) * | 2018-11-20 | 2019-04-16 | 广州方纬智慧大脑研究开发有限公司 | A kind of fault detection method of traffic lights, system and storage medium |
| CN109670423A (en) * | 2018-12-05 | 2019-04-23 | 依通（北京）科技有限公司 | A kind of image identification system based on deep learning, method and medium |
| CN109726692A (en) * | 2018-12-29 | 2019-05-07 | 重庆集诚汽车电子有限责任公司 | High-definition camera 3D object detection system based on deep learning |
| CN109803108A (en) * | 2019-01-22 | 2019-05-24 | 国网信息通信产业集团有限公司 | A kind of image-recognizing method and device |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111428647A (en) * | 2020-03-25 | 2020-07-17 | 浙江浙大中控信息技术有限公司 | Traffic signal lamp fault detection method |
| CN111428647B (en) * | 2020-03-25 | 2023-07-07 | 浙江中控信息产业股份有限公司 | Traffic signal lamp fault detection method |
| CN111681442A (en) * | 2020-06-09 | 2020-09-18 | 贾若然 | Signal lamp fault detection device based on image classification algorithm |
| CN111882910A (en) * | 2020-06-15 | 2020-11-03 | 太原市高远时代科技有限公司 | High-accuracy traffic signal lamp fault detection method and system |
| CN111882910B (en) * | 2020-06-15 | 2021-07-20 | 太原市高远时代科技有限公司 | Traffic signal lamp fault detection method and system |
| CN112180285B (en) * | 2020-09-23 | 2024-05-31 | 阿波罗智联(北京)科技有限公司 | Method and device for identifying traffic signal lamp faults, navigation system and road side equipment |
| CN111931724A (en) * | 2020-09-23 | 2020-11-13 | 北京百度网讯科技有限公司 | Signal lamp abnormity identification method and device, electronic equipment and road side equipment |
| CN112180285A (en) * | 2020-09-23 | 2021-01-05 | 北京百度网讯科技有限公司 | Method and device for identifying faults of traffic signal lamp, navigation system and road side equipment |
| CN111931724B (en) * | 2020-09-23 | 2024-06-21 | 阿波罗智联(北京)科技有限公司 | Signal lamp abnormality identification method and device, electronic equipment and road side equipment |
| CN112528787A (en) * | 2020-12-02 | 2021-03-19 | 博云视觉科技(青岛)有限公司 | Signal lamp fault detection method based on deep learning |
| CN112634621A (en) * | 2020-12-25 | 2021-04-09 | 北京百度网讯科技有限公司 | Traffic light data quality detection method, device, electronic equipment and medium |
| CN116704720B (en) * | 2023-08-03 | 2023-11-03 | 四川富基智能科技有限公司 | Intelligent traffic violation early warning system and method |
| CN116704720A (en) * | 2023-08-03 | 2023-09-05 | 四川富基智能科技有限公司 | Intelligent traffic violation early warning system and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110782692A (en) | Signal lamp fault detection method and system | |
| CN102496282B (en) | Traffic intersection signal light state identification method based on RGB color transformation | |
| CN103106766B (en) | Forest fire identification method and forest fire identification system | |
| US6985172B1 (en) | Model-based incident detection system with motion classification | |
| CN101872526B (en) | Smoke and fire intelligent identification method based on programmable photographing technology | |
| CN104766086B (en) | The monitoring and managing method and system of a kind of way mark | |
| CN111339905B (en) | CIM well lid state visual detection system based on deep learning and multiple visual angles | |
| CN111931726B (en) | Traffic light detection method, device, computer storage medium and road side equipment | |
| KR102174556B1 (en) | Apparatus for monitoring image to control traffic information employing Artificial Intelligence and vehicle number | |
| CN112149509B (en) | Traffic signal lamp fault detection method integrating deep learning and image processing | |
| CN111523397B (en) | Intelligent lamp post visual identification device, method and system and electronic equipment thereof | |
| CN111950536A (en) | Signal lamp image processing method and device, computer system and road side equipment | |
| CN205665896U (en) | Intersection signal lamp state recognition device | |
| CN104469309A (en) | Tunnel pedestrian intrusion detection device and method | |
| CN107749055A (en) | A kind of fault detection method, system and the device of LED traffic guidances screen | |
| CN109782364A (en) | Traffic mark board based on machine vision lacks detection method | |
| CN103456024A (en) | Moving object line crossing judgment method | |
| CN111428647A (en) | Traffic signal lamp fault detection method | |
| CN110826456A (en) | Countdown board fault detection method and system | |
| CN109001210B (en) | System and method for detecting aging and cracking of sealing rubber strip of civil air defense door | |
| CN110708830B (en) | Intelligent lamp inspection system | |
| CN112393186A (en) | Intelligent LED street lamp with people flow density monitoring and alarming functions | |
| CN113111771A (en) | Method for identifying unsafe behaviors of power plant workers | |
| CN103489202A (en) | Intrusion detection method based on videos | |
| CN111767775B (en) | Monitoring scene detection method and device and electronic equipment |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200211 |
|
| RJ01 | Rejection of invention patent application after publication |