Video monitoring device and monitoring method
Technical Field
The invention relates to the technical field of road traffic monitoring, in particular to a video monitoring device and a monitoring method.
Background
With the rapid and continuous development of national economy, the number of motor vehicles and drivers is rapidly increased, the traffic flow of roads is also rapidly increased, the road traffic management faces huge traffic safety pressure, and various traffic illegal behaviors are increasingly serious. How to relieve the current traffic pressure, improve the management level and reduce the occurrence of traffic accidents becomes the focus of more and more attention of people. The results are not ideal for the relevant department, although a great deal of effort has been made.
Violations are generally traffic violations, i.e. motor vehicles, notDriver of motor vehicleOr a pedestrian, violationRoad traffic safetyRegulations and traffic management activities. In addition, a violation may also refer to a violation of conventional regulation handling.
In order to urge drivers or pedestrians to avoid traffic violation behaviors, a snapshot camera probe, namely a traffic monitoring device, is usually arranged at a traffic intersection of an existing road, videos snapshot by the traffic monitoring devices need to be sent to a traffic management department through a cloud in real time, workers of the traffic management department need to supervise the snapshot videos in real time, the workload is high, a large number of material resources and economy can be consumed, the traffic violation behaviors are discovered only by means of the vision of the workers, omission easily occurs, and the supervision effect is poor.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a video monitoring device and a video monitoring method, which solve the problems that the existing video picture for traffic violation snapshot needs real-time supervision of a worker at the background, the process has large workload, consumes a large amount of material resources and economy, and the traffic violation behavior is discovered only by the vision of the worker, so that omission easily occurs and the supervision effect is poor.
In order to achieve the purpose, the invention is realized by the following technical scheme: a video surveillance apparatus comprising:
the video acquisition units are provided with a plurality of groups and are distributed at each intersection in a unit area;
the video transmission unit is connected with the data total output serial ports of the plurality of groups of video acquisition units and provides a transmission channel for the video data acquired by the plurality of groups of video acquisition units;
the data processing unit is connected in series with the data output port of the video transmission unit and is responsible for shearing the video data and comparing the characteristics in the video data;
the communication base station and the data sending terminal of the data processing unit form a data transmission channel in a wireless mode;
the remote control center and a signal transmission terminal of the communication base station form a data transmission channel in a wireless mode and are responsible for remotely calling the characteristic video and remotely adjusting the acquisition state of the video acquisition unit;
and the cloud storage is connected with the signal transmission terminal of the communication base station in real time in a wireless mode and is responsible for storing the remotely called characteristic video data.
Preferably, each group of video acquisition units comprises a plurality of high-definition cameras, and data interfaces of the high-definition cameras are connected in series.
Preferably, the data processing unit includes a video data receiving module connected in series with a data output port of the video transmission unit, a data output port of the video data receiving module is connected in series with a data input port of the video clip module, and a data output port of the video clip module is connected in series with a data input port of the video clip repository.
Preferably, the data processing unit further includes a neural network comparison module in which one data input interface is connected in series with the data output interface of the video segment repository, another data input interface of the neural network comparison module is connected in series with the data output interface of the neural network model, and the data input interface of the neural network model is connected in series with the data output port of the neural network training module.
Preferably, a data output port of the neural network comparison module is provided with a data input port of the feature video memory in series, the data output port of the feature video memory is connected with a data input port of the data remote sending module in series, and the data input port of the data remote sending module is connected with a data output port of the state data receiving module in series.
Preferably, the remote control center comprises a data remote receiving module which is connected with the communication base station in real time in a wireless mode, a data output end of the data remote receiving module is respectively connected with a data input end of the feature video extraction module and a data input end of the state console in series, a data output end of the feature video extraction module is connected with a data input end of the feature video player in series, and a data input end of the feature video extraction module is connected with a data output end of the feature input module in series.
Preferably, the system also comprises a state feedback unit and a state adjusting unit which are respectively connected in series between the video acquisition unit and the data processing unit, a horizontal angle sensor, a vertical angle sensor, a distance measuring sensor and a data processor are arranged in the state feedback unit, the horizontal angle sensor and the vertical angle sensor are respectively responsible for constantly monitoring the angles of each high-definition camera in the video acquisition unit in the horizontal direction and the vertical direction, the distance measuring sensor is responsible for monitoring the height of each high-definition camera in the video acquisition unit from the ground, and the horizontal angle sensor, the vertical angle sensor and the data processor transmit data acquired from the video acquisition unit to the data processor and transmit the data to the data processing unit through the data processor.
Preferably, the state adjusting unit comprises a horizontal angle adjusting motor, a vertical angle adjusting motor and a height adjusting motor which are respectively matched with the horizontal angle sensor, the vertical angle sensor and the distance measuring sensor for use, the horizontal angle adjusting motor, the vertical angle adjusting motor and the height adjusting motor are in parallel connection, and corresponding actions are executed under the control of the data processor.
Preferably, each node of the neural network model is inserted into a video segment clipped by one of the video clipping modules.
The monitoring method of the video monitoring device comprises the following steps:
s101, collecting traffic video pictures of corresponding intersections in real time by using a plurality of groups of video collecting units arranged at different intersections;
s102, transmitting a video picture acquired in real time to a video data receiving module through a video transmission unit in real time, transmitting the video picture to a video clipping module through the video data receiving module, and clipping a whole video content into a plurality of different small segments;
s103, storing the small fragment in a video fragment storage library;
s104, introducing a neural network model into a neural network comparison module, training the neural network model by using a neural network training module, and enabling the neural network model to have self-adaptive capacity, wherein the neural network comparison module extracts all lower segments in a video segment storage library to each node, sets a characteristic threshold value, judges whether a traffic violation phenomenon exists in each lower segment, correspondingly stores the small segment in a characteristic video storage if the traffic violation phenomenon exists, and automatically deletes the small segment if the traffic violation phenomenon does not exist;
s105, inputting a traffic violation feature word into the system by a worker on a remote control center by using a feature input module;
s106, the characteristic video extraction module sends a demand signal to a communication base station by means of a data remote receiving module according to the characteristic words, the communication base station transmits the signal to a data remote sending module, and the data remote sending module extracts the video small segments with the corresponding characteristics from a characteristic video memory into the characteristic video extraction module;
and S107, putting the extracted small video segments into a characteristic video player by a worker, and providing the worker with secondary confirmation of whether a traffic violation exists.
Advantageous effects
The invention provides a video monitoring device and a monitoring method. Compared with the prior art, the method has the following beneficial effects:
1. the video monitoring device and the monitoring method are provided with a data processing unit which is arranged between a remote control center and a video acquisition unit, when the video monitoring device works normally, the video acquisition unit is responsible for acquiring video pictures of each traffic crossing and transmitting the video pictures to the data processing unit, the data processing unit clips a whole video picture into a plurality of small segments through a built-in video clipping module, the small video segments are respectively led into each node of a neural network model to be compared, if traffic violation behaviors exist in the videos of the small segments, the small video segments are stored in a memory, when a worker of a subsequent traffic control department transfers the picture, the worker only needs to input the traffic violation characteristic words such as 'running red light and occupying the road', and can transfer the video picture of corresponding content, whether traffic violation exists is confirmed manually, the mode greatly simplifies daily work of workers, improves the efficiency of the workers and is suitable for popularization.
2. According to the video monitoring device and the monitoring method, the state feedback unit and the state adjusting unit which are provided with the video acquisition unit are arranged, the specific states comprise the horizontal and vertical angles of each high-definition camera and the height from the ground, and workers can remotely adjust the states so as to achieve the optimal video shooting effect.
Drawings
FIG. 1 is a block diagram schematically illustrating the structure of the present invention;
FIG. 2 is a block diagram schematically illustrating the structure of a data processing unit according to the present invention;
FIG. 3 is a block diagram schematically illustrating the structure of a remote control center according to the present invention;
FIG. 4 is a block diagram schematically illustrating the structure of a state feedback unit according to the present invention;
FIG. 5 is a block diagram schematically illustrating the structure of a state adjustment unit according to the present invention;
FIG. 6 is a block diagram of the workflow of the data processing unit of the present invention;
fig. 7 is a block diagram of the work flow of the remote control center of the present invention.
In the figure: 1. a video acquisition unit; 2. a video transmission unit; 3. a state feedback unit; 31. a horizontal angle sensor; 32. a vertical angle sensor; 33. a ranging sensor; 34. a data processor; 4. a state adjustment unit; 41. a horizontal angle adjustment motor; 42. a vertical angle adjusting motor; 43. a height adjustment motor; 44. a data processor; 5. a data processing unit; 51. a video data receiving module; 52. a video editing module; 53. a video clip repository; 54. a neural network comparison module; 55. a neural network model; 56. a neural network training module; 57. a feature video memory; 58. a data remote transmission module; 59. a status data receiving module; 6. a communication base station; 7. a remote control center; 71. a data remote receiving module; 72. a feature video extraction module; 73. a feature input module; 74. a feature video player; 75. a status console; 8. and the cloud storage.
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, the present invention provides a technical solution: a video surveillance apparatus comprising:
the system comprises video acquisition units 1, a plurality of groups of video acquisition units 1 and a plurality of groups of video acquisition units, wherein the video acquisition units 1 are distributed at each intersection in a unit area, each group of video acquisition units 1 comprises a plurality of high-definition cameras, and data interfaces of the high-definition cameras are mutually connected in series;
the video transmission unit 2 is connected with the data total output serial ports of the plurality of groups of video acquisition units 1 and provides a transmission channel for the video data acquired by the plurality of groups of video acquisition units 1;
the data processing unit 5 is connected in series with the data output port of the video transmission unit 2 and is responsible for cutting the video data and comparing the characteristics in the video data;
a communication base station 6 which forms a data transmission channel with the data transmission terminal of the data processing unit 5 in a wireless manner;
the remote control center 7 forms a data transmission channel with a signal transmission terminal of the communication base station 6 in a wireless mode and is responsible for remotely calling the characteristic video and remotely adjusting the acquisition state of the video acquisition unit 1;
and the cloud storage 8 is wirelessly connected with the signal transmission terminal of the communication base station 6 in real time and is responsible for storing remotely-called characteristic video data.
Referring to fig. 2, further, the data processing unit 5 includes a video data receiving module 51 connected in series with a data output port of the video transmitting unit 2, a data output port of the video data receiving module 51 is connected in series with a data input port of the video clip module 52, a data output port of the video clip module 52 is connected in series with a data input port of the video clip repository 53, the data processing unit 5 further includes a neural network comparing module 54 having one data input interface connected in series with a data output interface of the video clip repository 53, another data input interface of the neural network comparing module 54 is connected in series with a data output interface of the neural network model 55, a data input interface of the neural network model 55 is connected in series with a data output port of the neural network training module 56, a data output port of the neural network comparing module 54 is connected in series with a data input port of the feature video memory 57, the data output port of the feature video memory 57 is connected in series with the data input port of the data remote sending module 58, the data input port of the data remote sending module 58 is connected in series with the data output port of the status data receiving module 59, and each node of the neural network model 55 is inserted into a video segment clipped by the video clipping module 52.
Referring to fig. 3, the remote control center 7 includes a data remote receiving module 71 connected to the communication base station 6 in real time in a wireless manner, a data output end of the data remote receiving module 71 is connected in series to a data input end of the feature video extracting module 72 and a data input end of the status console 75, a data output end of the feature video extracting module 72 is connected in series to a data input end of the feature video player 74, and a data input end of the feature video extracting module 72 is connected in series to a data output end of the feature input module 73.
Referring to fig. 4-5, the video monitoring apparatus further includes a state feedback unit 3 and a state adjusting unit 4 respectively connected in series between the video capturing unit 1 and the data processing unit 5, a horizontal angle sensor 31 and a vertical angle sensor 32 are built in the state feedback unit 3, the distance measuring sensor 33 and the data processor 34, the horizontal angle sensor 31 and the vertical angle sensor 32 are respectively responsible for constantly monitoring the angle of each high-definition camera in the video acquisition unit 1 in the horizontal direction and the vertical direction, the distance measuring sensor 33 is responsible for monitoring the height of each high-definition camera in the video acquisition unit 1 from the ground, the horizontal angle sensor 31, the vertical angle sensor 32 and the data processor 34 transmit the data acquired from the video acquisition unit 1 to the data processor 34, and the data are transmitted to the data processing unit 5 through the data processor 34.
Referring to fig. 5, the state adjustment unit 4 includes a horizontal angle adjustment motor 41, a vertical angle adjustment motor 42, and a height adjustment motor 43 respectively corresponding to the horizontal angle sensor 31, the vertical angle sensor 32, and the distance measurement sensor 33, wherein the horizontal angle adjustment motor 41, the vertical angle adjustment motor 42, and the height adjustment motor 43 are connected in parallel, and are controlled by the data processor 44 to perform corresponding operations.
Further, the monitoring method of the video monitoring device comprises the following steps:
s101, collecting traffic video pictures of corresponding intersections in real time by using a plurality of groups of video collecting units 1 arranged at different intersections;
s102, transmitting a video picture acquired in real time to a video data receiving module 51 in real time through a video transmission unit 2, transmitting the video picture to a video clipping module 52 through the video data receiving module 51, and clipping a whole section of video content into a plurality of different small segments;
s103, storing the small fragments in the video fragment storage 53;
s104, introducing a neural network model 55 into the neural network comparison module 54, training the neural network model 55 by using a neural network training module 56, enabling the neural network model 55 to have self-adaptive capacity, extracting each lower segment from a video segment storage 53 to each node by using the neural network comparison module 54, setting a characteristic threshold value, judging whether a traffic violation phenomenon exists in each lower segment, if the traffic violation phenomenon exists, correspondingly storing the small segment in a characteristic video storage 57, and if the traffic violation phenomenon does not exist, automatically deleting the small segment;
s105, inputting a traffic violation feature word into the system by a worker on the remote control center 7 by using the feature input module 73;
s106, the feature video extraction module 72 sends a demand signal to the communication base station 6 by means of the data remote receiving module 71 according to the feature words, the communication base station 6 transmits the signal to the data remote sending module 58, and the data remote sending module 58 extracts the video small segments with corresponding features from the feature video memory 57 into the feature video extraction module 72;
and S107, the staff puts the extracted small video segments into the characteristic video player 74 and provides the staff with secondary confirmation of whether traffic violation behaviors exist or not.
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.
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.