CN111681442A - Signal lamp fault detection device based on image classification algorithm - Google Patents
Signal lamp fault detection device based on image classification algorithm Download PDFInfo
- Publication number
- CN111681442A CN111681442A CN202010516080.1A CN202010516080A CN111681442A CN 111681442 A CN111681442 A CN 111681442A CN 202010516080 A CN202010516080 A CN 202010516080A CN 111681442 A CN111681442 A CN 111681442A
- Authority
- CN
- China
- Prior art keywords
- signal lamp
- image classification
- classification algorithm
- detection device
- fault detection
- 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
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a signal lamp fault detection device based on an image classification algorithm, which comprises a camera, wherein an image processor and a central processing unit are respectively arranged at the bottom of the camera, a signal transceiver is arranged at the top of the camera, a clamping block is rotated at one side of the camera through a hinged block, the image processor performs image classification through the image classification algorithm, the image classification algorithm is specifically divided into signal lamp normality, signal lamp failure judgment and signal lamp fault, and the signal lamp fault detection device relates to the technical field of signal lamp fault detection. The signal lamp fault detection device based on the image classification algorithm solves the problem that good effect is difficult to obtain due to small target and under the background of complex night illumination factors in signal lamp detection.
Description
Technical Field
The invention relates to the technical field of signal lamp fault detection, in particular to a signal lamp fault detection device based on an image classification algorithm.
Background
Traffic lights, i.e., traffic lights, are very common signal lights, and with the increase in urban road construction and the number of motor vehicles, the problem of road traffic increasingly becomes a cross-point problem of social attention. The traffic signal lamp is usually directly installed at a traffic intersection, and due to long-term work and the influence of natural environment, the fault probability is high, so when the traffic signal lamp breaks down, related personnel need to know fault information in time, and take safety measures to the intersection where the fault signal lamp is located, so that traffic jam and safety accidents at the traffic intersection are avoided.
In the prior art, the fault detection of the traffic signal lamp usually depends on the special inspection of the supervision personnel or artificial feedback, and the signal lamp detection is often difficult to obtain good effect because the target is small and under the background of complex night illumination factors, so that the fault information of the traffic signal lamp and the like can not be obtained in time, so that the fault maintenance of the fault traffic signal lamp can not be carried out in time, and the road condition passing influence on the traffic intersection is greatly influenced.
Disclosure of Invention
Technical problem to be solved
Aiming at the defects of the prior art, the invention provides a signal lamp fault detection device based on an image classification algorithm, which solves the problem that a good effect is difficult to obtain due to small target and under the background of complex night illumination factors in signal lamp detection.
(II) technical scheme
In order to achieve the purpose, the invention is realized by the following technical scheme: the signal lamp fault detection device based on the image classification algorithm comprises a camera, wherein an image processor and a central processing unit are respectively arranged at the bottom of the camera, and a signal transceiver is arranged at the top of the camera.
Preferably, one side of the camera is rotated by a hinge block to form a clamping block.
Preferably, the image processor performs image classification by an image classification algorithm.
Preferably, the image classification algorithm is specifically divided into normal signal lamp, no judgment of the signal lamp and signal lamp fault.
Preferably, the signal lamp fault is specifically divided into:
A. if the timing lamp and the light supplement lamp are not on, recording as a timing off-line fault;
B. if the traffic light is not bright, recording as an offline fault of the traffic light;
C. the condition that two or more colors of the traffic light and the same light panel are simultaneously lightened is marked as a simultaneous lightening fault.
Preferably, the central processing unit performs a traffic light classification model by a convolutional neural network using a four-classification model.
Preferably, the four-classification model specifically includes: the method comprises a normal model, an offline model, a same brightness model and an undeterminable model, wherein a large amount of real field data are respectively collected for the four classification models to train the classification models.
Preferably, the convolution process adopts hole convolution to enlarge the receptive field and reduce the number of convolution layers, and utilizes grouping convolution to reduce the parameter number and the calculation amount.
(III) advantageous effects
The invention provides a signal lamp fault detection device based on an image classification algorithm. The method has the following beneficial effects:
the signal lamp fault detection device based on the image classification algorithm is realized by adopting a method of manually calibrating a signal lamp area, firstly, a rectangular frame of a signal lamp is manually calibrated for all signal lamps needing fault detection, and when whether the signal lamp is in fault needs to be judged, only images of the rectangular frame are input into a judgment model, so that the detection time is saved, misjudgment and missed detection are caused, and the problem that good effect is difficult to obtain due to small target and under the background of complex night illumination factors in signal lamp detection is solved.
Drawings
FIG. 1 is a schematic structural view of the present invention;
FIG. 2 is a diagram of a signal lamp fault detection model architecture in accordance with the present invention;
in the figure, 1 camera, 2 image processor, 3 central processing unit, 4 signal transceiver, 5 articulated block, 6 clamping block.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the 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.
Referring to fig. 1-2, an embodiment of the present invention provides a technical solution: the utility model provides a signal lamp fault detection device based on image classification algorithm, includes camera 1, and the bottom of camera 1 is provided with image processor 2 and central processing unit 3 respectively, and the top of camera 1 is provided with signal transceiver 4.
Further, one side of the camera 1 is rotated with a grip block 6 by a hinge block 5.
Further, the image processor 2 performs image classification by an image classification algorithm.
Further, the image classification algorithm is specifically divided into normal signal lamp, no judgment of the signal lamp and signal lamp fault.
Further, the signal lamp faults are specifically divided into:
A. if the timing lamp and the light supplement lamp are not on, recording as a timing off-line fault;
B. if the traffic light is not bright, recording as an offline fault of the traffic light;
C. the condition that two or more colors of the traffic light and the same light panel are simultaneously lightened is marked as a simultaneous lightening fault.
Further, the central processing unit 3 performs a traffic light classification model through a convolutional neural network using a four-classification model.
Further, the four-classification model specifically includes: the method comprises a normal model, an offline model, a same brightness model and an undeterminable model, wherein a large amount of real field data are respectively collected for the four classification models to train the classification models.
Further, in the convolution process, the cavity convolution is adopted to enlarge the receptive field and reduce the number of convolution layers, and the grouping convolution is adopted to reduce the parameter number and the calculated amount.
The utility model provides a signal lamp fault detection device based on image classification algorithm, includes camera 1, and the bottom of camera 1 is provided with image processor 2 and central processing unit 3 respectively, and the top of camera 1 is provided with signal transceiver 4, and one side of camera 1 rotates through articulated piece 5 has grip block 6.
The image processor 2 of the invention carries out image classification by an image classification algorithm, the image classification algorithm is specifically divided into the following steps that a signal lamp is normal, the signal lamp cannot be judged, and the signal lamp fails are specifically divided into the following steps: A. if the timing lamp and the light supplement lamp are not on, recording as a timing off-line fault; B. if the traffic light is not bright, recording as an offline fault of the traffic light; C. the condition that two or more colors of the traffic light and the same light panel are simultaneously lightened is marked as a simultaneous lightening fault.
The central processing unit 3 of the invention uses four classification models for the traffic light classification model through the convolutional neural network, and the four classification models are specifically as follows: the method comprises a normal model, an offline model, a same-brightness model and an undeterminable model, wherein a large amount of real field data are respectively collected for four classification models to train the classification models, the convolution process adopts cavity convolution to enlarge the receptive field and reduce the number of convolution layers, grouping convolution is utilized to reduce the number of parameters and calculated quantity, overfitting is avoided, and the last layer uses a full-connection layer to output the corresponding classification quantity.
In the detection process, because the alarm is only needed when the signal lamp really has a long-time fault in the actual engineering, each frame of image of the video does not need to be analyzed, only one frame of image of the video needs to be extracted every second for analysis, in the actual process, for example, several frames of short time are in a non-bright state at the switching moment from the red light countdown ending to the green light countdown of the timing light, the last few seconds of flickering also exist when the red light is switched from one color to the other color, and the lights of two colors are in a bright or non-bright state at the color switching moment, aiming at the above situations, the detection result of a single frame of image often has a large number of false alarms, in order to avoid the above phenomena, the alarm is carried out by using a sliding window judgment mode, the judgment result of the latest 10 moments is kept for each signal lamp, wherein 0 is normal, 1, when a fault detection result occurs occasionally, an alarm is not triggered, and the signal lamp fault is reported to the system and the fault type is given when the fault detection result continuously reaches 50 percent and is judged to be a fault.
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. Without further limitation.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (8)
1. The utility model provides a signal lamp fault detection device based on image classification algorithm, includes camera (1), its characterized in that: the bottom of the camera (1) is provided with an image processor (2) and a central processing unit (3) respectively, and the top of the camera (1) is provided with a signal transceiver (4).
2. The signal lamp fault detection device based on the image classification algorithm according to claim 1, characterized in that: one side of the camera (1) is rotated by a hinged block (5) to form a clamping block (6).
3. The signal lamp fault detection device based on the image classification algorithm according to claim 1, characterized in that: the image processor (2) performs image classification by an image classification algorithm.
4. The signal lamp fault detection device based on the image classification algorithm according to claim 3, characterized in that: the image classification algorithm is specifically divided into normal signal lamp, no judgment of signal lamp and signal lamp fault.
5. The signal lamp fault detection device based on the image classification algorithm according to claim 4, characterized in that: the signal lamp faults are specifically divided into:
A. if the timing lamp and the light supplement lamp are not on, recording as a timing off-line fault;
B. if the traffic light is not bright, recording as an offline fault of the traffic light;
C. the condition that two or more colors of the traffic light and the same light panel are simultaneously lightened is marked as a simultaneous lightening fault.
6. The signal lamp fault detection device based on the image classification algorithm according to claim 1, characterized in that: the central processing unit (3) performs traffic light classification model through a convolutional neural network and uses four classification models.
7. The signal lamp fault detection device based on the image classification algorithm according to claim 6, characterized in that: the four classification models are specifically: the method comprises a normal model, an offline model, a same brightness model and an undeterminable model, wherein a large amount of real field data are respectively collected for the four classification models to train the classification models.
8. The signal lamp fault detection device based on the image classification algorithm according to claim 6, characterized in that: in the convolution process, the cavity convolution is adopted to enlarge the receptive field and reduce the number of convolution layers, and the grouping convolution is adopted to reduce the parameter number and the calculated amount.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010516080.1A CN111681442A (en) | 2020-06-09 | 2020-06-09 | Signal lamp fault detection device based on image classification algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010516080.1A CN111681442A (en) | 2020-06-09 | 2020-06-09 | Signal lamp fault detection device based on image classification algorithm |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111681442A true CN111681442A (en) | 2020-09-18 |
Family
ID=72435606
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010516080.1A Pending CN111681442A (en) | 2020-06-09 | 2020-06-09 | Signal lamp fault detection device based on image classification algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111681442A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113781778A (en) * | 2021-09-03 | 2021-12-10 | 新奇点智能科技集团有限公司 | Data processing method and device, electronic equipment and readable storage medium |
CN114333334A (en) * | 2022-03-11 | 2022-04-12 | 南京企朋软件技术有限公司 | Automatic road monitoring method and system and network side server |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP0251097A1 (en) * | 1986-06-25 | 1988-01-07 | Siemens Aktiengesellschaft | Monitoring device for signal lights of a road traffic signal arrangement |
CN102629419A (en) * | 2012-04-12 | 2012-08-08 | 中国计量学院 | Fault detection device for LED traffic lights |
CN108376253A (en) * | 2018-03-05 | 2018-08-07 | 西南交通大学 | A kind of signal trouble monitoring method based on machine vision |
CN109740412A (en) * | 2018-11-09 | 2019-05-10 | 浙江浩腾电子科技股份有限公司 | A kind of signal lamp failure detection method based on computer vision |
CN110782692A (en) * | 2019-10-31 | 2020-02-11 | 青岛海信网络科技股份有限公司 | Signal lamp fault detection method and system |
CN110992725A (en) * | 2019-10-24 | 2020-04-10 | 合肥讯图信息科技有限公司 | Method, system and storage medium for detecting traffic signal lamp fault |
-
2020
- 2020-06-09 CN CN202010516080.1A patent/CN111681442A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP0251097A1 (en) * | 1986-06-25 | 1988-01-07 | Siemens Aktiengesellschaft | Monitoring device for signal lights of a road traffic signal arrangement |
CN102629419A (en) * | 2012-04-12 | 2012-08-08 | 中国计量学院 | Fault detection device for LED traffic lights |
CN108376253A (en) * | 2018-03-05 | 2018-08-07 | 西南交通大学 | A kind of signal trouble monitoring method based on machine vision |
CN109740412A (en) * | 2018-11-09 | 2019-05-10 | 浙江浩腾电子科技股份有限公司 | A kind of signal lamp failure detection method based on computer vision |
CN110992725A (en) * | 2019-10-24 | 2020-04-10 | 合肥讯图信息科技有限公司 | Method, system and storage medium for detecting traffic signal lamp fault |
CN110782692A (en) * | 2019-10-31 | 2020-02-11 | 青岛海信网络科技股份有限公司 | Signal lamp fault detection method and system |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113781778A (en) * | 2021-09-03 | 2021-12-10 | 新奇点智能科技集团有限公司 | Data processing method and device, electronic equipment and readable storage medium |
CN113781778B (en) * | 2021-09-03 | 2022-09-06 | 新奇点智能科技集团有限公司 | Data processing method and device, electronic equipment and readable storage medium |
CN114333334A (en) * | 2022-03-11 | 2022-04-12 | 南京企朋软件技术有限公司 | Automatic road monitoring method and system and network side server |
CN114333334B (en) * | 2022-03-11 | 2022-05-17 | 南京企朋软件技术有限公司 | Automatic road monitoring method and system and network side server |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109636777A (en) | A kind of fault detection method of traffic lights, system and storage medium | |
US7786877B2 (en) | Multi-wavelength video image fire detecting system | |
CN101458865B (en) | Fire disaster probe system and method | |
CN101334924B (en) | Fire hazard probe system and its fire hazard detection method | |
CN111681442A (en) | Signal lamp fault detection device based on image classification algorithm | |
CN113129591B (en) | Traffic signal lamp fault detection method based on deep learning target detection | |
CN103260049A (en) | Intelligent skynet video quality diagnostic system | |
CN111428647B (en) | Traffic signal lamp fault detection method | |
CN106534700B (en) | The quick method of soft light is realized based on automatic exposure and automatic white balance algorithm | |
CN102176758A (en) | Video quality diagnosis system and realization method thereof | |
CN112149509B (en) | Traffic signal lamp fault detection method integrating deep learning and image processing | |
CN107103330A (en) | A kind of LED status recognition methods and device | |
CN106530772A (en) | Intelligent traffic signal lamp, control system of the same and emergency control method of the same | |
CN107123274A (en) | Double parking stall video detecting devices and method | |
CN116091901A (en) | Operation and maintenance management platform based on fault diagnosis and analysis of monitoring equipment | |
CN110135343A (en) | A kind of street lamp intelligent measurement and night status judgment method | |
CN112560816A (en) | Equipment indicator lamp identification method and system based on YOLOv4 | |
CN201091014Y (en) | Fire detecting device | |
CN104267719A (en) | Bus-system LED display system point-by-point fault detection method and application thereof | |
CN106710253A (en) | High-reliability intelligent intersection traffic control system and control method | |
CN107450439A (en) | A kind of street lamp intelligent failure diagnosis method | |
CN110708830B (en) | Intelligent lamp inspection system | |
KR20180036039A (en) | Control system for street light by closely weather | |
CN104091402B (en) | Distinguishing system and distinguishing method of multi-state alarm color-changing lamp of 24V power supply cabinet | |
CN116721095A (en) | Aerial photographing road illumination fault detection method and device |
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: 20200918 |
|
RJ01 | Rejection of invention patent application after publication |