CN111382707A - Method for simultaneously processing multiple high beam vehicles in same video - Google Patents
Method for simultaneously processing multiple high beam vehicles in same video Download PDFInfo
- Publication number
- CN111382707A CN111382707A CN202010164327.8A CN202010164327A CN111382707A CN 111382707 A CN111382707 A CN 111382707A CN 202010164327 A CN202010164327 A CN 202010164327A CN 111382707 A CN111382707 A CN 111382707A
- Authority
- CN
- China
- Prior art keywords
- high beam
- video
- vehicles
- information
- processing
- 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
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000001514 detection method Methods 0.000 claims abstract description 33
- 238000013135 deep learning Methods 0.000 claims abstract description 10
- 238000004590 computer program Methods 0.000 claims abstract description 8
- 238000000605 extraction Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000009434 installation Methods 0.000 abstract description 5
- 235000017166 Bambusa arundinacea Nutrition 0.000 description 5
- 235000017491 Bambusa tulda Nutrition 0.000 description 5
- 241001330002 Bambuseae Species 0.000 description 5
- 235000015334 Phyllostachys viridis Nutrition 0.000 description 5
- 239000011425 bamboo Substances 0.000 description 5
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 238000003490 calendering Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G06T5/90—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- 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
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06V20/584—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of vehicle lights or traffic lights
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30248—Vehicle exterior or interior
- G06T2207/30252—Vehicle exterior; Vicinity of vehicle
-
- 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
- G06V20/625—License plates
-
- 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
- G06V2201/08—Detecting or categorising vehicles
Abstract
The invention relates to a method for simultaneously processing a plurality of high beam vehicles in the same video, which comprises the following steps: firstly, a high beam hardware detection module detects a high beam and sends a high beam signal to an upper computer program; after receiving a 'module 1 high beam' signal sent by the module, the upper computer program starts to call the camera sdk, downloads the video and stores the video to a local industrial personal computer; a plurality of high beam vehicles in the same video are processed simultaneously through the processing system, license plates are captured and screened, and after the processing is completed, data are written into the database for uploading the data. According to the invention, a software and hardware combination mode is adopted, a high beam detection module is utilized, and a target detection technology based on deep learning is added, so that double guarantee is achieved, the problems of difficult installation and debugging are effectively solved, and the detection rate is high and the adaptability is strong. And by adopting a multi-thread and multi-video simultaneous processing technology, the capture rate is supplemented and improved, and the real-time performance of snapshot is greatly improved.
Description
Technical Field
The invention relates to the field of license plate snapshot of road motor vehicles, in particular to a method for simultaneously processing a plurality of high beam vehicles in the same video.
Background
When a driver of a motor vehicle drives at night, the driver does not use the high beam as required, and the driver of the following or oncoming motor vehicle is likely to be blinded within a few seconds, which is a main cause of traffic accidents at night. The traditional high beam detection technology of finding double lamp tubes by video calendering is high in speed, but low in capture rate, only capable of capturing double lamp tubes of vehicles and not universal. Traditional far-reaching headlamp vehicle detection technique of violating regulations utilizes the hardware device to detect, adopts the video press polish to look for a double-lamp section of thick bamboo technique, has certain limitation, embodies in can catching a double-lamp section of thick bamboo motorcycle type only, can't catch a single-lamp section of thick bamboo, a multiple-lamp section of thick bamboo etc. do not have evidence fairness, the capture rate is only 60%, can't accomplish motorcycle type, a lamp section of thick bamboo all standing. The method for detecting the high beam by adopting the hardware device has the advantages that the detection mode is not visual, the installation and debugging difficulty is higher, the influence of the road flatness is easy to cause, and the adaptability is poorer. In the prior art, a target tracking algorithm based on deep learning is adopted for high beam detection, so that the problems of difficult installation, high false alarm rate and the like of the traditional high beam detection method are effectively solved, but the real-time performance is poorer than that of video processing, so that a method for simultaneously processing a plurality of high beam vehicles in the same video, which can solve the problems of high real-time performance, full coverage, high false alarm rate and the like of high beam snapshot, is urgently needed.
Disclosure of Invention
The invention aims to provide a method for simultaneously processing a plurality of high beam vehicles in the same video, which has the advantages of real-time performance, full coverage and low false alarm rate.
In order to solve the above problems, the present invention provides a method for simultaneously processing a plurality of high beam vehicles in the same video, comprising the following steps:
s1: firstly, a high beam hardware detection module detects a high beam and sends a high beam signal to an upper computer program;
s2: after receiving a 'module 1 high beam' signal sent by the module, the upper computer program starts to call the camera sdk, downloads the video and stores the video to a local industrial personal computer;
s3: a plurality of high beam vehicles in the same video are processed simultaneously through the processing system, license plates are captured and screened, and after the processing is completed, data are written into the database for uploading the data.
The invention has the advantages that the high beam detection module is utilized by adopting a software and hardware combination mode, the target detection technology based on deep learning is added at the same time, double guarantee is realized, the problems of difficult installation and debugging are effectively solved, the detection rate is high, and the adaptability is strong. By adopting a multithreading and multi-video simultaneous processing technology, the capture rate is supplemented and improved, the real-time performance of snapshot is greatly improved, and a firm foundation is laid by law enforcement on the site of a high beam rather than according to regulations.
Further, the step S3 further includes the following steps:
a1: firstly, detecting and judging high beam lamps, and according to folders with the same number of high beam lamps, wherein videos in each folder correspond to the coordinate position of each initial high beam lamp vehicle;
a2, tracking and identifying license plates of high beam vehicles, and deleting repeated vehicle information;
a3: and writing the sorted data into a database for uploading the data.
Further, the step a1 further includes the following steps:
b1: firstly, starting a sub-thread in a main thread for detecting the information of the high beam vehicle;
b2: extracting a first video frame through a high beam detection algorithm based on deep learning, and finding out high beam vehicles contained in the video frame;
b3: after the detection is finished, copying different folders according to the detected information of the number of the vehicles using the high beam without regulations, wherein the folders contain initial video information, recording the coordinate positions of the vehicles, sending the coordinates and the folder paths to the main thread, and waiting for the processing of the main thread;
b4: and after all the information is sent to the main thread, destroying the high beam detection thread.
Further, the step a2 further includes the following steps:
c1: after the main thread receives the coordinate and path information sent by the high beam detection thread, different video processing threads are respectively created, and the coordinate information and the corresponding video path information of the high beam vehicle which are not used according to the regulations are transmitted by each video processing thread;
c2: in the video processing process, an original high beam vehicle is tracked by reading a video and utilizing an opencv target tracking algorithm;
c3: when a vehicle arrives at the near end of a video, license plate information is extracted by using a license plate extraction algorithm based on deep learning, and after the vehicle leaves a frame, redundant video frames are deleted;
c4: and sending the processed license plate and video path information to a main thread for deleting repeated license plate information.
Further, the step B2 further includes the following steps:
d1: and further filtering out non-high beam regions by setting different high beam detection confidence threshold values.
Further, the step C4 further includes the following steps:
e1: the main thread receives license plate information sent by the video processing thread and stores the license plate information into a List;
e2: and comparing the last 20 license plate information in the List, if the license plate information is the same, determining that the license plate information contains repeated license plates, and deleting the folder information in the path.
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 flow chart of a method of the present invention for simultaneously processing multiple high-beam vehicles in the same video.
Detailed Description
The present invention will be further described in detail with reference to the following specific examples:
the invention aims to provide a method for simultaneously processing a plurality of high beam vehicles in the same video, which has the advantages of real-time performance, full coverage and low false alarm rate.
As shown in fig. 1, to solve the above problem, the present invention provides a method for simultaneously processing multiple high beam vehicles in the same video, comprising the following steps:
s1: firstly, a high beam hardware detection module detects a high beam and sends a high beam signal to an upper computer program;
s2: after receiving a 'module 1 high beam' signal sent by the module, the upper computer program starts to call the camera sdk, downloads the video and stores the video to a local industrial personal computer;
s3: a plurality of high beam vehicles in the same video are processed simultaneously through the processing system, license plates are captured and screened, and after the processing is completed, data are written into the database for uploading the data.
The invention has the advantages that the high beam detection module is utilized by adopting a software and hardware combination mode, the target detection technology based on deep learning is added at the same time, double guarantee is realized, the problems of difficult installation and debugging are effectively solved, the detection rate is high, and the adaptability is strong. By adopting a multithreading and multi-video simultaneous processing technology, the capture rate is supplemented and improved, the real-time performance of snapshot is greatly improved, and a firm foundation is laid by law enforcement on the site of a high beam rather than according to regulations.
Further, the step S3 further includes the following steps:
a1: firstly, detecting and judging high beam lamps, and according to folders with the same number of high beam lamps, wherein videos in each folder correspond to the coordinate position of each initial high beam lamp vehicle;
a2, tracking and identifying license plates of high beam vehicles, and deleting repeated vehicle information;
a3: and writing the sorted data into a database for uploading the data.
In actual operation, the method for simultaneously processing a plurality of high beams in the same video mainly comprises the following steps: detecting and judging the high beam; according to the folders with the same number of high beam lamps, wherein the video in each folder corresponds to the coordinate position of each initial high beam lamp vehicle; tracking and license plate recognition of high beam vehicles; deleting duplicate vehicle information; the process is complete, etc. In order to process data simultaneously, a large number of thread operations, thread starting, thread processing, thread destroying and the like need to be performed.
Further, the step a1 further includes the following steps:
b1: firstly, starting a sub-thread in a main thread for detecting the information of the high beam vehicle;
b2: extracting a first video frame through a high beam detection algorithm based on deep learning, and finding out high beam vehicles contained in the video frame;
b3: after the detection is finished, copying different folders according to the detected information of the number of the vehicles using the high beam without regulations, wherein the folders contain initial video information, recording the coordinate positions of the vehicles, sending the coordinates and the folder paths to the main thread, and waiting for the processing of the main thread;
b4: and after all the information is sent to the main thread, destroying the high beam detection thread.
Further, the step a2 further includes the following steps:
c1: after the main thread receives the coordinate and path information sent by the high beam detection thread, different video processing threads are respectively created, and the coordinate information and the corresponding video path information of the high beam vehicle which are not used according to the regulations are transmitted by each video processing thread;
c2: in the video processing process, an original high beam vehicle is tracked by reading a video and utilizing an opencv target tracking algorithm;
c3: when a vehicle arrives at the near end of a video, license plate information is extracted by using a license plate extraction algorithm based on deep learning, and after the vehicle leaves a frame, redundant video frames are deleted;
c4: and sending the processed license plate and video path information to a main thread for deleting repeated license plate information.
Further, the step B2 further includes the following steps:
d1: and further filtering out non-high beam regions by setting different high beam detection confidence threshold values.
Further, the step C4 further includes the following steps:
e1: the main thread receives license plate information sent by the video processing thread and stores the license plate information into a List;
e2: and comparing the last 20 license plate information in the List, if the license plate information is the same, determining that the license plate information contains repeated license plates, and deleting the folder information in the path.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (6)
1. A method for simultaneously processing a plurality of high beam vehicles in the same video, which is characterized by comprising the following steps:
s1: firstly, a high beam hardware detection module detects a high beam and sends a high beam signal to an upper computer program;
s2: after receiving a 'module 1 high beam' signal sent by the module, the upper computer program starts to call the camera sdk, downloads the video and stores the video to a local industrial personal computer;
s3: a plurality of high beam vehicles in the same video are processed simultaneously through the processing system, license plates are captured and screened, and after the processing is completed, data are written into the database for uploading the data.
2. The method for processing multiple high-beam vehicles in the same video at the same time according to claim 1, wherein the step S3 further comprises the steps of:
a1: firstly, detecting and judging high beam lamps, and according to folders with the same number of high beam lamps, wherein videos in each folder correspond to the coordinate position of each initial high beam lamp vehicle;
a2, tracking and identifying license plates of high beam vehicles, and deleting repeated vehicle information;
a3: and writing the sorted data into a database for uploading the data.
3. The method for processing multiple high-beam vehicles in the same video at the same time according to claim 2, wherein said step a1 further comprises the following steps:
b1: firstly, starting a sub-thread in a main thread for detecting the information of the high beam vehicle;
b2: extracting a first video frame through a high beam detection algorithm based on deep learning, and finding out high beam vehicles contained in the video frame;
b3: after the detection is finished, copying different folders according to the detected information of the number of the vehicles using the high beam without regulations, wherein the folders contain initial video information, recording the coordinate positions of the vehicles, sending the coordinates and the folder paths to the main thread, and waiting for the processing of the main thread;
b4: and after all the information is sent to the main thread, destroying the high beam detection thread.
4. The method for processing multiple high-beam vehicles in the same video at the same time according to claim 2, wherein said step a2 further comprises the following steps:
c1: after the main thread receives the coordinate and path information sent by the high beam detection thread, different video processing threads are respectively created, and the coordinate information and the corresponding video path information of the high beam vehicle which are not used according to the regulations are transmitted by each video processing thread;
c2: in the video processing process, an original high beam vehicle is tracked by reading a video and utilizing an opencv target tracking algorithm;
c3: when a vehicle arrives at the near end of a video, license plate information is extracted by using a license plate extraction algorithm based on deep learning, and after the vehicle leaves a frame, redundant video frames are deleted;
c4: and sending the processed license plate and video path information to a main thread for deleting repeated license plate information.
5. The method for processing multiple high-beam vehicles in the same video at the same time according to claim 3, wherein said step B2 further comprises the following steps:
d1: and further filtering out non-high beam regions by setting different high beam detection confidence threshold values.
6. The method for processing multiple high-beam vehicles in the same video at the same time as claimed in claim 1, wherein said step C4 further comprises the steps of:
e1: the main thread receives license plate information sent by the video processing thread and stores the license plate information into a List;
e2: and comparing the last 20 license plate information in the List, if the license plate information is the same, determining that the license plate information contains repeated license plates, and deleting the folder information in the path.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010164327.8A CN111382707A (en) | 2020-03-11 | 2020-03-11 | Method for simultaneously processing multiple high beam vehicles in same video |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010164327.8A CN111382707A (en) | 2020-03-11 | 2020-03-11 | Method for simultaneously processing multiple high beam vehicles in same video |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111382707A true CN111382707A (en) | 2020-07-07 |
Family
ID=71217272
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010164327.8A Pending CN111382707A (en) | 2020-03-11 | 2020-03-11 | Method for simultaneously processing multiple high beam vehicles in same video |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111382707A (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102298845A (en) * | 2011-08-29 | 2011-12-28 | 安徽超远信息技术有限公司 | Far-beam light video detection method and system |
CN105184227A (en) * | 2015-08-13 | 2015-12-23 | 安徽超远信息技术有限公司 | Traffic signal control system with automobile high beam light detection function |
CN205028457U (en) * | 2015-08-06 | 2016-02-10 | 杭州智诚惠通科技有限公司 | City far -reaching headlamp detection device |
CN110823533A (en) * | 2019-09-10 | 2020-02-21 | 李东华 | Non-standard light detection system and detection method thereof |
-
2020
- 2020-03-11 CN CN202010164327.8A patent/CN111382707A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102298845A (en) * | 2011-08-29 | 2011-12-28 | 安徽超远信息技术有限公司 | Far-beam light video detection method and system |
CN205028457U (en) * | 2015-08-06 | 2016-02-10 | 杭州智诚惠通科技有限公司 | City far -reaching headlamp detection device |
CN105184227A (en) * | 2015-08-13 | 2015-12-23 | 安徽超远信息技术有限公司 | Traffic signal control system with automobile high beam light detection function |
CN110823533A (en) * | 2019-09-10 | 2020-02-21 | 李东华 | Non-standard light detection system and detection method thereof |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US6442474B1 (en) | Vision-based method and apparatus for monitoring vehicular traffic events | |
CN202077142U (en) | Vehicle-mounted intelligent video detecting and analyzing system | |
US8019122B2 (en) | Vehicle side image recognition method and apparatus | |
CN102568242B (en) | Signal lamp state detection method and system based on video processing | |
CN101630412B (en) | Camera-based lane marker detection | |
CN103279756B (en) | Vehicle detection based on integrated classifier analyzes system and determination method thereof | |
CN110298300B (en) | Method for detecting vehicle illegal line pressing | |
GB2503322A (en) | Real-time video triggering for traffic surveillance and photo enforcement applications using near infrared video acquisition | |
CN104504908A (en) | Monitoring method and system for illegal parking | |
CN110580808B (en) | Information processing method and device, electronic equipment and intelligent traffic system | |
CN110738857A (en) | vehicle violation evidence obtaining method, device and equipment | |
CN105551261A (en) | False-license-plate vehicle detection method and system | |
CN102799862A (en) | System and method for pedestrian rapid positioning and event detection based on high definition video monitor image | |
CN112364742A (en) | Parking space management method and device, computer equipment and storage medium | |
KR102306789B1 (en) | License Plate Recognition Method and Apparatus for roads | |
KR100942409B1 (en) | Method for detecting a moving vehicle at a high speed | |
CN111382707A (en) | Method for simultaneously processing multiple high beam vehicles in same video | |
CN112365719A (en) | Parking space management method and device, computer equipment and storage medium | |
CN111310738A (en) | High beam vehicle snapshot method based on deep learning | |
CN108985197B (en) | Automatic detection method for taxi driver smoking behavior based on multi-algorithm fusion | |
CN105761501A (en) | Intelligent vehicle behavior detecting and snapshotting method | |
KR101294498B1 (en) | Method of objects recognition for tunnel accident monitoring system | |
CN112435479B (en) | Target object violation detection method and device, computer equipment and system | |
CN110766949B (en) | Violation snapshot method and device | |
JP2008287367A (en) | Vehicle detection system |
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 |