CN113014876B - Video monitoring method and device, electronic equipment and readable storage medium - Google Patents
Video monitoring method and device, electronic equipment and readable storage medium Download PDFInfo
- Publication number
- CN113014876B CN113014876B CN202110212326.0A CN202110212326A CN113014876B CN 113014876 B CN113014876 B CN 113014876B CN 202110212326 A CN202110212326 A CN 202110212326A CN 113014876 B CN113014876 B CN 113014876B
- Authority
- CN
- China
- Prior art keywords
- preset
- video image
- video
- different scenes
- 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.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N17/00—Diagnosis, testing or measuring for television systems or their details
- H04N17/002—Diagnosis, testing or measuring for television systems or their details for television cameras
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream, rendering scenes according to MPEG-4 scene graphs
- H04N21/44008—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream, rendering scenes according to MPEG-4 scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/442—Monitoring of processes or resources, e.g. detecting the failure of a recording device, monitoring the downstream bandwidth, the number of times a movie has been viewed, the storage space available from the internal hard disk
- H04N21/4425—Monitoring of client processing errors or hardware failure
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
- H04N7/183—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a single remote source
- H04N7/185—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a single remote source from a mobile camera, e.g. for remote control
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The application discloses a video monitoring method, a video monitoring device, electronic equipment and a readable storage medium, wherein the method comprises the following steps: under the condition that the video monitoring terminal is controlled to move according to a preset moving track, acquiring video data acquired by the video monitoring terminal at a current preset position, wherein the preset moving track comprises N preset positions, each preset position in the N preset positions is provided with at least one electronic fence, the electronic fences are in one-to-one correspondence with a detection model, the N preset positions comprise the current preset position, and N is an integer larger than 1; decoding the video data to obtain video images; detecting a video image by using a target detection model to obtain a detection result, wherein the target detection model is a detection model corresponding to the electronic fence of the current preset position; if the detection result is abnormal, generating alarm information. Therefore, the hardware cost of the video monitoring system can be reduced, and the utilization rate of the video monitoring terminal can be improved.
Description
Technical Field
The application belongs to the technical field of video monitoring, and particularly relates to a video monitoring method, a video monitoring device, electronic equipment and a readable storage medium.
Background
With the development of video monitoring technology, video monitoring systems are widely used in various fields, such as industrial manufacturing, medical treatment, transportation, environmental protection, public safety, and the like. Currently, since video monitoring terminals can only acquire video data with a fixed preset bit, each video monitoring terminal can only monitor video data in a single scene. When the video monitoring system needs to monitor more scenes, more video monitoring terminals need to be arranged, so that the hardware cost of the video monitoring system is increased, and the utilization rate of a single video monitoring terminal is low.
Disclosure of Invention
The embodiment of the application aims to provide a video monitoring method, a video monitoring device, electronic equipment and a readable storage medium, which can solve the problems that the hardware cost of a video monitoring system is high and the utilization rate is low due to the existing video monitoring method.
In order to solve the technical problems, the application is realized as follows:
in a first aspect, an embodiment of the present application provides a video monitoring method, where the method includes:
under the condition that a video monitoring terminal is controlled to move according to a preset moving track, video data acquired by the video monitoring terminal at a current preset position are acquired, wherein the preset moving track comprises N preset positions, at least one electronic fence is arranged at each preset position in the N preset positions, the electronic fences correspond to a detection model one by one, the N preset positions comprise the current preset position, and N is an integer larger than 1;
Decoding the video data to obtain video images;
detecting the video image by using a target detection model to obtain a detection result, wherein the target detection is a detection model corresponding to the current preset electronic fence;
and if the detection result is abnormal, generating alarm information.
Further, the detecting the video image by using the target detection model to obtain a detection result includes:
acquiring the position parameters of the electronic fence of the current preset position;
cutting the video image according to the position parameters to obtain a cut video image;
and detecting the cut video image by using the target detection model to obtain a detection result.
Further, the clipping the video image according to the position parameter to obtain a clipped video image includes:
acquiring target parameters in the position parameters, wherein the target parameters comprise a minimum value and a maximum value in the horizontal direction and a minimum value and a maximum value in the vertical direction in a preset coordinate system;
determining a clipping region according to the minimum value and the maximum value in the horizontal direction and the minimum value and the maximum value in the vertical direction;
And cutting the video image according to the cutting area to obtain a cut video image.
Further, the clipping the video image according to the clipping region to obtain a clipped video image, including:
cutting the video image according to the cutting area to obtain a first intermediate video image;
adjusting the first intermediate video image to a preset size to obtain a second intermediate video image;
and filtering the second intermediate video image to obtain a cut video image.
Further, each preset position in the N preset positions is also provided with preset stay time length and available time range; the preset stay time is the stay time of the video monitoring terminal when the video monitoring terminal moves to each preset position, and the available time range is the effective time range of the detection model corresponding to the electronic fence set in each preset position.
Further, before the detecting the video image using the object detection model, the method further includes:
acquiring sample video images corresponding to different scenes;
training a basic model according to the sample video image to obtain detection models corresponding to different scenes;
And respectively associating the detection models corresponding to the different scenes with the electronic fence arranged in each preset position.
Further, the sample video image includes a training data set and a test data set;
training the basic model according to the sample video image to obtain detection models corresponding to different scenes, wherein the training comprises the following steps:
clustering the sample video images according to a preset clustering algorithm to obtain prior frame parameters corresponding to different scenes;
respectively training the basic model according to the prior frame parameters, preset configurable super parameters and the training data set to obtain a plurality of candidate models corresponding to different scenes;
according to the test data set, evaluating a plurality of candidate models corresponding to different scenes to obtain evaluation results corresponding to different scenes;
and determining detection models corresponding to different scenes according to the evaluation results corresponding to the different scenes.
In a second aspect, an embodiment of the present application provides a video monitoring apparatus, including:
the first acquisition module is used for acquiring video data acquired by the video monitoring terminal at a current preset position under the condition that the video monitoring terminal is controlled to move according to a preset movement track, wherein the preset movement track comprises N preset positions, at least one electronic fence is arranged at each preset position in the N preset positions, the electronic fences are in one-to-one correspondence with a detection model, the N preset positions comprise the current preset positions, and N is an integer larger than 1;
The decoding module is used for decoding the video data to obtain video images;
the detection module is used for detecting the video image by using a target detection model to obtain a detection result, wherein the target detection model is a detection model corresponding to the current preset electronic fence;
and the generating module is used for generating alarm information if the detection result is abnormal.
Further, the detection module includes:
the acquisition sub-module is used for acquiring the position parameters of the electronic fence of the current preset position;
the cutting sub-module is used for cutting the video image according to the position parameters to obtain a cut video image;
and the detection sub-module is used for detecting the cut video image by using the target detection model to obtain a detection result.
Further, the clipping submodule includes:
an acquisition unit configured to acquire a target parameter among the position parameters, the target parameter including a minimum value and a maximum value in a horizontal direction and a minimum value and a maximum value in a vertical direction in a preset coordinate system;
a determining unit configured to determine a clipping region according to the minimum and maximum values in the horizontal direction and the minimum and maximum values in the vertical direction;
And the clipping unit is used for clipping the video image according to the clipping region to obtain a clipped video image.
Further, the clipping unit is specifically configured to:
cutting the video image according to the cutting area to obtain a first intermediate video image;
adjusting the first intermediate video image to a preset size to obtain a second intermediate video image;
and filtering the second intermediate video image to obtain a cut video image.
Further, each preset position in the N preset positions is also provided with preset stay time length and available time range; the preset stay time is the stay time of the video monitoring terminal when the video monitoring terminal moves to each preset position, and the available time range is the effective time range of the detection model corresponding to the electronic fence set in each preset position.
Further, the apparatus further comprises:
the second acquisition module is used for acquiring sample video images corresponding to different scenes;
the training module is used for training the basic model according to the sample video image to obtain detection models corresponding to different scenes;
and the association module is used for respectively associating the detection models corresponding to the different scenes with the electronic fence arranged in each preset position.
Further, the sample video image includes a training data set and a test data set; the training module comprises:
the clustering sub-module is used for clustering the sample video images according to a preset clustering algorithm to obtain prior frame parameters corresponding to different scenes;
the training sub-module is used for respectively training the basic model according to the prior frame parameters, preset configurable super parameters and the training data set to obtain a plurality of candidate models corresponding to different scenes;
the evaluation sub-module is used for evaluating a plurality of candidate models corresponding to different scenes according to the test data set to obtain evaluation results corresponding to different scenes;
and the determining submodule is used for determining detection models corresponding to different scenes according to the evaluation results corresponding to the different scenes.
In a third aspect, embodiments of the present application provide an electronic device comprising a processor, a memory and a program or instruction stored on the memory and executable on the processor, the program or instruction implementing the steps of the method according to the first aspect when executed by the processor.
In a fourth aspect, embodiments of the present application provide a readable storage medium having stored thereon a program or instructions which when executed by a processor implement the steps of the method according to the first aspect.
In the embodiment of the application, under the condition that a video monitoring terminal is controlled to move according to a preset moving track, video data acquired by the video monitoring terminal at a current preset position are acquired, wherein the preset moving track comprises N preset positions, at least one electronic fence is arranged at each preset position in the N preset positions, the electronic fences are in one-to-one correspondence with a detection model, the N preset positions comprise the current preset positions, and N is an integer larger than 1; decoding the video data to obtain video images; detecting the video image by using a target detection model to obtain a detection result, wherein the target detection model is a detection model corresponding to the current preset electronic fence; and if the detection result is abnormal, generating alarm information. Through the mode, the video monitoring terminal can be controlled to move among N preset positions, video data of the N preset positions are collected, at the same time, at least one electronic fence is arranged in each preset position of the N preset positions, the video image corresponding to each electronic fence is facilitated to be obtained, the detection model corresponding to each electronic fence is utilized to detect the video image, and therefore video monitoring of different scenes is achieved. Therefore, the use amount of the video monitoring terminal can be reduced, the hardware cost of the video monitoring system is reduced, video images of a plurality of different scenes can be monitored by using the same video monitoring terminal, and the utilization rate of the video monitoring terminal is improved.
Drawings
Fig. 1 is one of flowcharts of a video monitoring method provided in an embodiment of the present application;
FIG. 2 is a second flowchart of a video monitoring method according to an embodiment of the present disclosure;
fig. 3 is a flowchart of setting a movement track according to an embodiment of the present application;
FIG. 4 is a third flowchart of a video monitoring method according to an embodiment of the present disclosure;
fig. 5 is a block diagram of a video monitoring device according to an embodiment of the present application;
fig. 6 is a block diagram of an electronic device according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, are intended to be within the scope of the present application.
The terms first, second and the like in the description and in the claims, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged, as appropriate, such that embodiments of the present application may be implemented in sequences other than those illustrated or described herein, and that the objects identified by "first," "second," etc. are generally of a type and not limited to the number of objects, e.g., the first object may be one or more. Furthermore, in the description and claims, "and/or" means at least one of the connected objects, and the character "/", generally means that the associated object is an "or" relationship.
The video monitoring method provided by the embodiment of the application is described in detail below by means of specific embodiments and application scenes thereof with reference to the accompanying drawings.
Referring to fig. 1, fig. 1 is one of flowcharts of a video monitoring method according to an embodiment of the present application. As shown in fig. 1, the video monitoring method specifically includes the following steps:
Specifically, the video monitoring method of the embodiment of the application is applied to a video monitoring system, and the video monitoring system comprises at least one video monitoring terminal and a server connected with the video monitoring terminal. The video monitoring terminal comprises, but is not limited to, a network camera, or terminal equipment with a camera, such as a mobile phone, a computer and the like. The server is used for detecting the collected video data by using the detection model, and sending out alarm information when abnormal conditions are detected.
The preset bit refers to the physical position where the video monitoring terminal collects video data. The electronic fence is used for indicating the areas needing to be monitored in a key mode on each preset position, and the server can conveniently cut video images of the preset positions based on the electronic fence. For example, assuming that the preset position a needs to monitor the condition of the parking violations on both sides of the road, two electronic fences can be set at the preset position a and respectively set in the areas on both sides of the road, so that when the video data of the preset position a is analyzed, only the video images in the two cut electronic fences need to be analyzed, and the monitoring of the parking violations on both sides of the road is realized. And if the exhaust emission and the waste water emission of a certain factory are required to be monitored at the preset position B, two electronic fences can be arranged at the preset position B and respectively arranged above the smoke outlet and the sewage outlet of the factory, so that when the video data of the preset position B are analyzed, only video images in the two cut electronic fences are analyzed, and the monitoring of the exhaust emission and the waste water emission of the factory is realized.
It should be noted that, the monitoring scenes corresponding to the electronic fences may be the same or different, so that the detection models corresponding to the electronic fences may be the same or different according to the types of the monitoring scenes. Each electronic fence corresponds to one detection model one by one, and therefore, the server can detect a monitoring scene corresponding to the electronic fence according to the detection model corresponding to the electronic fence.
In the embodiment of the present application, the server is preset with a preset movement track corresponding to the video monitoring terminal. The preset moving track comprises N preset bits, and each preset bit is provided with at least one electronic fence, so that the server can control the video monitoring terminal to move according to the preset moving track, and video data acquired by each preset bit are acquired in the moving process.
After the server acquires the video data, the video data can be decoded to obtain a corresponding video image. Specifically, since the video data acquired by the video monitoring terminal is a video stream, such as an RTSP (or Real Time Streaming Protocol for short) stream, when the server receives the video stream transmitted by the video monitoring terminal, the video stream needs to be transcoded into a video image, such as image data in a base64 format, according to a preset frame rate. As an implementation manner, decoding of the video stream may be implemented by using a video capture-like feature in OpenCV (a cross-platform computer vision and machine learning software library issued for BSD-based license) for processing pictures and videos, and of course, this embodiment may also be implemented by other video decoding software, which is not specifically limited in this application.
And 103, detecting the video image by using a target detection model to obtain a detection result, wherein the target detection model is a detection model corresponding to the electronic fence which is currently preset.
After the server obtains the video image, the electronic fence corresponding to the current preset position can be obtained according to the current preset position, and then the target detection model corresponding to the electronic fence is obtained. Thus, the server can detect the video image according to the obtained target detection model to obtain a detection result.
It should be noted that the number of the object detection models matches the number of the electronic fences currently preset. For example, assuming that the current preset bit is set with 2 electronic fences, the number of target detection models is 2. The target detection model herein may be any target detection model, for example, an R-CNN (abbreviated as Region-based Convolutional Neural Network) model, a Fast R-CNN (abbreviated as Fast Region-based Convolutional Neural Network) model, an SSD (abbreviated as Single Shot Multibox Detector) model, a Yolo model, and the like.
And 104, if the detection result is abnormal, generating alarm information.
When the server detects that the detection result is abnormal, the server can generate alarm information to prompt a user. Specifically, the detection result is abnormal, which can be understood that the target object exists in the video image or the number of the target objects reaches a preset threshold. For example, assuming that the target detection model is to monitor the illegal parking vehicles on two sides of the road, when the illegal parking vehicles exist on two sides of the road in the video image, determining that the detection result is abnormal; and if the number of vehicles appearing in the video image reaches a preset threshold value, judging that the detection result is abnormal and the like. The alarm information includes but is not limited to alarm information in the forms of text, pictures, sound, photoelectricity and the like.
In this embodiment, the server may control the video monitoring terminal to move between N preset bits, collect video data of the N preset bits, and set at least one electronic fence in each preset bit of the N preset bits, so as to facilitate obtaining video images corresponding to each electronic fence, and detect the video images by using a detection model corresponding to each electronic fence, thereby implementing video monitoring of different scenes. Therefore, the use amount of the video monitoring terminal can be reduced, the hardware cost of the video monitoring system is reduced, video images of a plurality of different scenes can be monitored by using the same video monitoring terminal, and the utilization rate of the video monitoring terminal is improved.
Further, referring to fig. 2, fig. 2 is a second flowchart of a video monitoring method according to an embodiment of the present application. Based on the embodiment shown in fig. 1, the step 103 of detecting the video image by using the target detection model to obtain a detection result specifically includes the following steps:
After the server acquires the video image, the position parameters of the electronic fence which are preset currently can be obtained. The location parameter herein may be understood as the location of the area enclosed by the electronic fence in the preset coordinate system.
And 202, clipping the video image according to the position parameters to obtain a clipped video image.
After the server obtains the position parameters of the electronic fence which is preset currently, the server can cut the video image according to the position parameters. Specifically, as an implementation manner, according to the position parameters of the electronic fence set currently in advance, the edge information of each electronic fence can be determined, and cutting is performed along the edge of each electronic fence. As another embodiment, the maximum value and the minimum value in the horizontal direction and the maximum value and the minimum value in the vertical direction in the preset coordinate system in the position parameters of each electronic fence can be determined according to the position parameters of the electronic fence which are currently preset, and then the clipping region of each electronic fence is determined according to the maximum value and the minimum value in the horizontal direction and the maximum value and the minimum value in the vertical direction, and clipping is performed according to the clipping region of each electronic fence. Therefore, by removing unimportant areas in the video image, the server only needs to identify and detect the video image obtained after cutting, so that the identification and detection of the whole video image are avoided, and the identification and detection efficiency is improved.
And 203, detecting the cut video image by using the target detection model to obtain a detection result.
And inputting the video image obtained after cutting based on each electronic fence into a target detection model corresponding to the electronic fence, and detecting the video image obtained after cutting by using the target detection model to obtain a detection result corresponding to each scene. The video data collected by each preset bit of the N preset bits can be detected by adopting the mode, so that the video image corresponding to each electronic fence of each preset bit can be detected.
In this embodiment, by clipping the video image, only the clipped video image needs to be identified and detected, so that the identification and detection of the whole video image are avoided, and the efficiency of identification and detection is improved. And moreover, the cut video images are input to the corresponding target detection models for detection, so that detection results corresponding to all scenes can be obtained, the video monitoring function of a plurality of different scenes can be realized through one video monitoring terminal, and the flexibility and the utilization rate of the video monitoring terminal are effectively improved.
Further, the step 202 of cropping the video image according to the position parameter to obtain a cropped video image includes:
acquiring target parameters in the position parameters, wherein the target parameters comprise a minimum value and a maximum value in the horizontal direction and a minimum value and a maximum value in the vertical direction in a preset coordinate system;
determining a clipping region according to the minimum value and the maximum value in the horizontal direction and the minimum value and the maximum value in the vertical direction;
and cutting the video image according to the cutting area to obtain a cut video image.
In an embodiment, the server may obtain the position parameters of each electronic fence, determine the maximum value xmax and the minimum value xmin in the horizontal direction and the maximum value ymax and the minimum value ymin in the vertical direction in the position parameters of each electronic fence, and determine the clipping area of each electronic fence according to the maximum value xmax and the minimum value xmin in the horizontal direction and the maximum value ymax and the minimum value ymin the vertical direction. The clipping area of each electronic fence is a rectangular area surrounded by four points with coordinates (xmin, ymin), (xmin, ymax), (xmax, ymin) and (xmax, ymax). And finally, cutting according to the cutting areas of the electronic fences to obtain cut video images. Therefore, the server can remove video images outside the electronic fence, and only performs identification detection on the video images obtained after cutting, so that the identification detection of the whole video image is avoided, and the identification detection efficiency is improved.
Further, the step of clipping the video image according to the clipping region to obtain a clipped video image specifically includes the following steps:
cutting the video image according to the cutting area to obtain a first intermediate video image;
adjusting the first intermediate video image to a preset size to obtain a second intermediate video image;
and filtering the second intermediate video image to obtain a cut video image.
In an embodiment, the server may crop the video image based on the cropping range corresponding to each electronic fence, to obtain a first intermediate video image corresponding to each electronic fence. And after the first intermediate video images are obtained, the first intermediate video images are subjected to size adjustment, and are adjusted to a preset size, so that second intermediate video images corresponding to the electronic fences are obtained. And then, filtering the second intermediate video images to remove noise interference of the second intermediate video images and obtain cut video images. Therefore, the video image is subjected to size adjustment and filtering processing, so that the image quality can be effectively improved, the accuracy of image identification is improved, and meanwhile, the accuracy of a detection result can be improved when the cut video image is detected by using the target detection model.
In addition, as another embodiment, the step of resizing the video image and the step of filtering the video image may be performed simultaneously, or the video image may be filtered first and then resized, which is not particularly limited in this application.
Of course, as another implementation mode, after the video image is resized and filtered, algorithms such as defogging, raindrop removal and the like can be selected for the video image according to different scenes, so that the influence of the environment on the quality of the video image under different scenes is eliminated, and the quality of the video image is effectively improved.
Further, each preset position in the N preset positions is also provided with preset stay time length and available time range; the preset stay time length refers to stay time length of the video monitoring terminal when the video monitoring terminal moves to each preset position, and the available time range refers to the effective time range of the detection model corresponding to the electronic fence set in each preset position.
Specifically, the preset residence time is residence time of the video monitoring terminal when the video monitoring terminal moves to each preset position, and the residence time of each preset position can be the same or different and can be any time of 5 seconds, 10 seconds, 5 minutes, 1 hour and the like. The usable time ranges refer to effective time ranges of detection models corresponding to the electronic fences set in the preset positions, and the usable time ranges corresponding to the electronic fences can be the same or different. The preset residence time and the usable time range can be set according to actual needs, and the application is not particularly limited.
Specifically, the correspondence among each video monitoring terminal, the preset bit, the electronic fence, the detection model, the preset stay time length and the available time range may be stored in the server in advance. When the server collects video data and invokes the target detection model to detect the video image, the server can collect and detect the video image based on the corresponding relation. For example, assume that the correspondence between each video monitor terminal, preset bit, electronic fence, detection model, preset stay length, and available time range is as shown in the following table:
it can be seen that for the video monitor terminal 001, 3 preset bits may be corresponding to 001001, 001002 and 001003, respectively, and each preset bit includes at least one electronic fence. Each electronic fence is used for carrying out video monitoring on different scenes for one detection model respectively. And each preset bit corresponds to a preset stay time respectively, so that the server can control the video monitoring terminal 001 to move among the 3 preset bits and stay for the corresponding preset stay time when moving to the corresponding preset bit. For example, when the server controls the video monitoring terminal 001 to move to the preset position 001001 at a time point of 9:00, the video monitoring terminal 001 may stay at the preset position 001001 for 2 hours, and the server may determine whether to collect the video data of the preset stay time according to the available time range (i.e., the validation start time and the validation end time) and the validation flag of the detection model. Because the time of the video monitoring terminal 001 staying at the preset position 001001 accords with the available time range of the electronic fence 001001001 and does not accord with the available time range of the electronic fence 001001002, the server only needs to monitor the video image scene corresponding to the electronic fence 001001001. At this time, the server may request the video monitoring terminal 001 to upload video data collected at the preset bit 001001, decode the video data after obtaining the video data at the preset bit 001001, obtain a video image, then crop the video image according to the position parameter of the electronic fence 001001001, obtain a trimmed video image, and finally detect the video image corresponding to the electronic fence 001001001 by using the detection model 03. When the server controls the video monitoring terminal 001 to move to the preset position 001002, the time point is 11:00, the server stays at the preset position 001002 for 1 hour, and the time of the video monitoring terminal 001 at the preset position 001002 accords with the available time range of the electronic fences 001002001 and 001002002, so that the server needs to monitor video image scenes corresponding to the electronic fences 001002001 and 001002002. At this time, the server may request the video monitoring terminal 001 to upload video data collected at the preset position 001002, decode the video data after obtaining the video data at the preset position 001002, obtain a video image, then clip the video image according to the position parameters of the electronic fences 001002001 and 001002002, obtain a clipped video image, and finally detect the video image corresponding to the electronic fence 001002001 by using the detection model 13, and detect the video image corresponding to the electronic fence 001002002 by using the detection model 06. When the server controls the video monitoring terminal 001 to move to the preset position 001003, the time point is 12:00, the server stays at the preset position 001003 for 0.5 hour, and the time of the video monitoring terminal 001 at the preset position 001003 accords with the available time range of the electronic fence 001003002, so that the server needs to monitor the video image scene corresponding to the electronic fence 001003002. At this time, the server may request the video monitoring terminal 001 to upload the video data collected at the preset position 001003, decode the video data after obtaining the video data at the preset position 001003, obtain a video image, then crop the video image according to the position parameter of the electronic fence 001003002, obtain a trimmed video image, and finally detect the video image corresponding to the electronic fence 001003002 by using the detection model 11. By analogy, the server, after having completed monitoring the preset bit 001003, may return to the preset bit 001001 again, and re-perform the above-described loop operation. Of course, in practical application, parameters such as the sequence of movement of each preset position in the preset movement track, the residence time of each preset position, the number of electronic fences of each preset position, the detection modules corresponding to each electronic fence, the available time range of the detection model corresponding to each electronic fence, whether the detection model is effective or not and the like can be flexibly set, and the application is not limited specifically.
In this embodiment, each preset position is provided with a corresponding electronic fence, and a corresponding preset stay time length and an available time range, so that the server can control the stay time of the video monitoring terminal at each preset position to reach the preset stay time length according to the preset stay time length corresponding to each preset position, so as to ensure that the server can acquire video data required by each preset position. The server can control the effective time of each detection model according to the available time range, namely the effective time range of the detection model corresponding to the electronic fence set in each preset position, and can detect the video data acquired in the effective time range of each detection model without detecting the video data acquired outside the effective time range, so that the load of the server can be effectively reduced, and the running efficiency of the server is improved.
Further, based on the embodiment shown in fig. 1, before the detecting the video image by using the object detection model in step 103, the method further includes:
step 301, obtaining sample video images corresponding to different scenes.
The sample video image refers to a huge amount of video images obtained for training a model and evaluating the model. When the sample video images are acquired, the monitoring videos in different scenes can be acquired according to the scenes corresponding to the detection models, the monitoring videos are subjected to frame extraction to obtain a large number of video images, and then the large number of video images are sorted, screened and counted to ensure sample equalization. And then, carrying out data annotation on the video image by image annotation software such as LabelImg and the like to obtain an XML-format annotation file. In order to enhance the robustness and generalization performance of the detection model, data enhancement techniques such as random inversion, random clipping, noise addition, random brightness variation, random channel variation, random contrast variation, random saturation and chromaticity variation are introduced to amplify the sample video image. And finally, dividing the amplified video image and the labeling data into a training data set and a testing data set according to a preset proportion. The preset ratio may be 8:2, 9:1, etc., which is not specifically limited in this application.
And 302, training the basic model according to the sample video image to obtain detection models corresponding to different scenes.
The basic model is a deep learning model, and the detection model is a target detection model based on the deep learning model, such as an R-CNN (short for Region-based Convolutional Neural Network) model, a Fast R-CNN (short for Fast Region-based Convolutional Neural Network) model, an SSD (short for Single Shot Multibox Detector) model, a YOLO model and the like.
Specifically, after obtaining a sample video image, the server may input a training data set into a basic model to obtain a plurality of candidate models in different scenes, and evaluate the plurality of candidate models based on a test data set, and select a model with optimal performance as a detection model.
Step 303, associating detection models corresponding to different scenes with the electronic fences set in each preset position respectively.
After the server obtains the detection models corresponding to different scenes, the detection models corresponding to different scenes can be established to be respectively associated with the electronic fences arranged in the preset positions, so that when the server obtains the video data of the preset positions, the corresponding target detection models can be determined according to the electronic fences, and the video monitoring function of different scenes is realized.
In this embodiment, it is necessary to establish detection models corresponding to different scenes, and establish association relationships between each detection model and each electronic fence, so that the server can detect video images of different scenes based on different detection models.
Further, the sample video image includes a training data set and a test data set;
step 302, training the basic model according to the sample video image to obtain detection models corresponding to different scenes, including:
clustering the sample video images according to a preset clustering algorithm to obtain prior frame parameters corresponding to different scenes;
training the basic model according to the prior frame parameters, preset configurable super parameters and a training data set to obtain a plurality of candidate models corresponding to different scenes;
according to the test data set, evaluating a plurality of candidate models corresponding to different scenes to obtain evaluation results corresponding to different scenes;
and determining detection models corresponding to different scenes according to the evaluation results corresponding to the different scenes.
The preset clustering algorithm may be any one of a k-means clustering algorithm (k-means clustering algorithm), a mean shift clustering algorithm, and the like. The prior frame parameters refer to anchors values for test model training.
In one embodiment, a k-means algorithm may be used to cluster training data sets in the sample video image, obtain the size and aspect ratio of box labels in the training data sets, obtain the anchors value, and replace the initial values of anchors in the model configuration file according to the obtained anchors value. Initializing configurable superparameters of a basic model, such as learning_rate and burn_in, respectively training the basic model according to the obtained anchors value, preset configurable superparameters and training data sets to obtain a plurality of candidate models corresponding to different scenes, using a test data set to evaluate the candidate models generated in each round, and recording evaluation indexes of each candidate model, such as data of accuracy (precision), recall (recovery), merging ratio (intersection over union, IOU), average precision (average precision, AP), average precision average (mean average precision, mAP) and the like. And comprehensively comparing the evaluation index data of each candidate model in different scenes, and taking the model with the optimal performance in different scenes as a detection model in the scene.
In this embodiment, a plurality of candidate models may be obtained through test training, and then the detection model may be determined from the candidate models based on the test data set, so as to improve accuracy of a detection result of the detection model.
In an application example, referring to fig. 3, fig. 3 is a flowchart of setting a movement track. As shown in fig. 3, the setting process of the movement track includes the following steps:
the parameters of the preset bit i include, but are not limited to, horizontal angle, inclination angle of the pan-tilt, focal length of the camera, and the like.
the detection rules include, but are not limited to, rules of preset stay time, available time range and the like.
if the preset bit is continuously set, i+1, and repeating steps 310 to 350; if the preset bit is not to be set, step 370 is performed.
Therefore, the setting process of the moving track can be completed, and when the camera is started to acquire video data, the camera can cruise according to the moving track.
In the application example, the utilization rate of the camera can be improved by arranging a plurality of preset positions on the camera and arranging at least one electronic fence on each preset position.
In yet another application, referring to fig. 4, fig. 4 is a specific flowchart of a video monitoring method. As shown in fig. 4, the video monitoring method includes the following steps:
at this time, the server may also load information such as the camera angle and the lens focal length of the preset bit i+1, so as to control the camera to quickly reach the preset bit i+1.
the parameters in the detection rule and the preset table are the same as those in step 320 and step 350 in the above application example, and are not described herein again.
the above steps 404 to 407 are described in detail in the above embodiments, and are not described herein.
if the preset result is abnormal, executing step 409; if the preset result is normal, step 410 is performed.
if the cruise is continued, i+1, repeating the steps 401 to 409; if the cruising is not continued, the cruising is ended.
In the application example, the video images of the scenes are detected through the detection models corresponding to the electronic fences, so that video monitoring of different scenes can be realized. Therefore, the use amount of the video monitoring terminal can be reduced, the hardware cost of the video monitoring system is reduced, video images of a plurality of different scenes can be monitored by using the same video monitoring terminal, and the utilization rate of the video monitoring terminal is improved.
It should be noted that, in the video monitoring method provided in the embodiment of the present application, the execution body may be a video monitoring device, or a control module for the video monitoring method in the video monitoring device. In the embodiment of the application, a video monitoring device executes a video monitoring method as an example, and the video monitoring device provided in the embodiment of the application is described.
Referring to fig. 5, fig. 5 is a block diagram of a video monitoring device according to an embodiment of the present application. As shown in fig. 5, the video monitoring apparatus 500 includes:
The first obtaining module 501 is configured to obtain video data collected by the video monitoring terminal at a current preset position under the condition that the video monitoring terminal is controlled to move according to a preset movement track, where the preset movement track includes N preset positions, each preset position in the N preset positions is provided with at least one electronic fence, the electronic fences are in one-to-one correspondence with the detection model, the N preset positions include the current preset position, and N is an integer greater than 1;
the decoding module 502 is configured to decode the video data to obtain a video image;
the detection module 503 is configured to detect a video image by using a target detection model, to obtain a detection result, where the target detection model is a detection model corresponding to a current preset electronic fence;
and the generating module 504 is configured to generate alarm information if the detection result is abnormal.
Further, the detection module 503 includes:
the acquisition sub-module is used for acquiring the position parameters of the electronic fence of the current preset position;
the cutting sub-module is used for cutting the video image according to the position parameters to obtain a cut video image;
and the detection sub-module is used for detecting the cut video image by using the target detection model to obtain a detection result.
Further, the clipping submodule includes:
an acquisition unit configured to acquire a target parameter among the position parameters, the target parameter including a minimum value and a maximum value in a horizontal direction and a minimum value and a maximum value in a vertical direction in a preset coordinate system;
a determining unit for determining a clipping region according to the minimum and maximum values in the horizontal direction and the minimum and maximum values in the vertical direction;
and the clipping unit is used for clipping the video image according to the clipping region to obtain a clipped video image.
Further, the clipping unit is specifically configured to:
cutting the video image according to the cutting area to obtain a first intermediate video image;
adjusting the first intermediate video image to a preset size to obtain a second intermediate video image;
and filtering the second intermediate video image to obtain a cut video image.
Further, each preset position in the N preset positions is also provided with preset stay time length and available time range; the preset stay time length refers to stay time length of the video monitoring terminal when the video monitoring terminal moves to each preset position, and the available time range refers to the effective time range of the detection model corresponding to the electronic fence set in each preset position.
Further, the video monitoring apparatus 500 further includes:
the second acquisition module is used for acquiring sample video images corresponding to different scenes;
the training module is used for training the basic model according to the sample video image to obtain detection models corresponding to different scenes;
and the association module is used for respectively associating detection models corresponding to different scenes with the electronic fence arranged in each preset position.
Further, the sample video image includes a training data set and a test data set; the training module comprises:
the clustering sub-module is used for clustering the sample video images according to a preset clustering algorithm to obtain prior frame parameters corresponding to different scenes;
the training sub-module is used for respectively training the basic model according to the prior frame parameters, the preset configurable super parameters and the training data set to obtain a plurality of candidate models corresponding to different scenes;
the evaluation sub-module is used for evaluating a plurality of candidate models corresponding to different scenes according to the test data set to obtain evaluation results corresponding to the different scenes;
and the determining submodule is used for determining detection models corresponding to different scenes according to the evaluation results corresponding to the different scenes.
The video monitoring device 500 in the embodiment of the present application may be a device, or may be a component, an integrated circuit, or a chip in a terminal. The device may be a mobile electronic device or a non-mobile electronic device. By way of example, the mobile electronic device may be a cell phone, tablet computer, notebook computer, palm computer, vehicle-mounted electronic device, wearable device, ultra-mobile personal computer (ultra-mobile personal computer, UMPC), netbook or personal digital assistant (personal digital assistant, PDA), etc., and the non-mobile electronic device may be a server, network attached storage (Network Attached Storage, NAS), personal computer (personal computer, PC), television (TV), teller machine or self-service machine, etc., and the embodiments of the present application are not limited in particular.
The video monitoring device 500 in the embodiment of the present application may be a device with an operating system. The operating system may be an Android operating system, an ios operating system, or other possible operating systems, which are not specifically limited in the embodiments of the present application.
The video monitoring apparatus 500 provided in this embodiment of the present application can implement each process implemented by the method embodiments of fig. 1 to 4, and in order to avoid repetition, a description is omitted here.
Optionally, as shown in fig. 6, the embodiment of the present application further provides an electronic device 600, including a processor 601, a memory 602, and a program or an instruction stored in the memory 602 and capable of running on the processor 601, where the program or the instruction implements each process of the embodiment of the video monitoring method when executed by the processor 601, and the process can achieve the same technical effect, so that repetition is avoided, and no further description is given here.
The electronic device in the embodiment of the application includes the mobile electronic device and the non-mobile electronic device.
The embodiment of the application further provides a readable storage medium, on which a program or an instruction is stored, which when executed by a processor, implements each process of the embodiment of the video monitoring method, and can achieve the same technical effects, so that repetition is avoided, and no further description is given here.
The processor is a processor in the electronic device in the above embodiment. Readable storage media include computer readable storage media such as Read-Only Memory (ROM), random access Memory (Random Access Memory, RAM), magnetic or optical disks, and the like.
It should be noted that, in this document, 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, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element. Furthermore, it should be noted that the scope of the methods and apparatus in the embodiments of the present application is not limited to performing the functions in the order shown or discussed, but may also include performing the functions in a substantially simultaneous manner or in an opposite order depending on the functions involved, e.g., the described methods may be performed in an order different from that described, and various steps may also be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk), including several instructions for causing a terminal (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method of the embodiments of the present application.
The embodiments of the present application have been described above with reference to the accompanying drawings, but the present application is not limited to the above-described embodiments, which are merely illustrative and not restrictive, and many forms may be made by those of ordinary skill in the art without departing from the spirit of the present application and the scope of the claims, which are also within the protection of the present application.
Claims (8)
1. A video monitoring method, the method comprising:
under the condition that a video monitoring terminal is controlled to move according to a preset moving track, video data acquired by the video monitoring terminal at a current preset position are acquired, wherein the preset moving track comprises N preset positions, at least one electronic fence is arranged at each preset position in the N preset positions, the electronic fences correspond to a detection model one by one, the N preset positions comprise the current preset position, and N is an integer larger than 1;
decoding the video data to obtain video images;
acquiring sample video images corresponding to different scenes; training a basic model according to the sample video image to obtain detection models corresponding to different scenes; respectively associating detection models corresponding to different scenes with the electronic fence arranged in each preset position; clustering a training data set in a sample video image by using a k-means algorithm, obtaining the size and the length-width ratio of a box mark in the training data set, obtaining an anchors value, and replacing the anchors value with an initial value of anchors in a model configuration file; initializing configurable hyper-parameters of a basic model, and respectively training the basic model according to the anchors value, the preset configurable hyper-parameters and a training data set to obtain a plurality of candidate models corresponding to different scenes; using a test data set to evaluate the candidate models generated in each round, and recording the evaluation index of each candidate model; comprehensively comparing the evaluation index data of each candidate model in different scenes, and taking the model with the optimal performance in different scenes as a detection model in the scene;
Detecting the video image by using a target detection model to obtain a detection result, wherein the method comprises the following steps: acquiring the position parameters of the electronic fence of the current preset position; cutting the video image according to the position parameters to obtain a cut video image; detecting the cut video image by using the target detection models to obtain a detection result, wherein the number of the target detection models is matched with the number of the electronic fences in the current preset position, and the target detection models are detection models corresponding to the electronic fences in the current preset position;
and if the detection result is abnormal, generating alarm information.
2. The method of claim 1, wherein cropping the video image according to the location parameter to obtain a cropped video image comprises:
acquiring target parameters in the position parameters, wherein the target parameters comprise a minimum value and a maximum value in the horizontal direction and a minimum value and a maximum value in the vertical direction in a preset coordinate system;
determining a clipping region according to the minimum value and the maximum value in the horizontal direction and the minimum value and the maximum value in the vertical direction;
And cutting the video image according to the cutting area to obtain a cut video image.
3. The method according to claim 2, wherein cropping the video image according to the cropping zone to obtain a cropped video image comprises:
cutting the video image according to the cutting area to obtain a first intermediate video image;
adjusting the first intermediate video image to a preset size to obtain a second intermediate video image;
and filtering the second intermediate video image to obtain a cut video image.
4. The method of claim 1, wherein each of the N preset bits is further provided with a preset dwell time and an available time range; the preset stay time is the stay time of the video monitoring terminal when the video monitoring terminal moves to each preset position, and the available time range is the effective time range of the detection model corresponding to the electronic fence set in each preset position.
5. The method of claim 1, wherein the sample video image comprises a training dataset and a test dataset;
Training the basic model according to the sample video image to obtain detection models corresponding to different scenes, wherein the training comprises the following steps:
clustering the sample video images according to a preset clustering algorithm to obtain prior frame parameters corresponding to different scenes;
respectively training the basic model according to the prior frame parameters, preset configurable super parameters and the training data set to obtain a plurality of candidate models corresponding to different scenes;
according to the test data set, evaluating a plurality of candidate models corresponding to different scenes to obtain evaluation results corresponding to different scenes;
and determining detection models corresponding to different scenes according to the evaluation results corresponding to the different scenes.
6. A video surveillance device, the device comprising:
the first acquisition module is used for acquiring video data acquired by the video monitoring terminal at a current preset position under the condition that the video monitoring terminal is controlled to move according to a preset movement track, wherein the preset movement track comprises N preset positions, at least one electronic fence is arranged at each preset position in the N preset positions, the electronic fences are in one-to-one correspondence with a detection model, the N preset positions comprise the current preset positions, and N is an integer larger than 1;
The decoding module is used for decoding the video data to obtain video images;
the second acquisition module is used for acquiring sample video images corresponding to different scenes;
the training module is used for training the basic model according to the sample video image to obtain detection models corresponding to different scenes;
the association module is used for associating the detection models corresponding to the different scenes with the electronic fence arranged in each preset position respectively; clustering a training data set in a sample video image by using a k-means algorithm, obtaining the size and the length-width ratio of a box mark in the training data set, obtaining an anchors value, and replacing the anchors value with an initial value of anchors in a model configuration file; initializing configurable hyper-parameters of a basic model, and respectively training the basic model according to the anchors value, the preset configurable hyper-parameters and a training data set to obtain a plurality of candidate models corresponding to different scenes; using a test data set to evaluate the candidate models generated in each round, and recording the evaluation index of each candidate model; comprehensively comparing the evaluation index data of each candidate model in different scenes, and taking the model with the optimal performance in different scenes as a detection model in the scene;
The detection module is used for detecting the video image by utilizing target detection models to obtain a detection result, wherein the number of the target detection models is matched with the number of the electronic fences in the current preset position, and the target detection models are detection models corresponding to the electronic fences in the current preset position;
the generation module is used for generating alarm information if the detection result is abnormal;
the detection module comprises:
the acquisition sub-module is used for acquiring the position parameters of the electronic fence of the current preset position;
the cutting sub-module is used for cutting the video image according to the position parameters to obtain a cut video image;
and the detection sub-module is used for detecting the cut video image by using the target detection model to obtain a detection result.
7. An electronic device comprising a processor, a memory and a program or instruction stored on the memory and executable on the processor, which when executed by the processor, implements the steps of the video surveillance method of any of claims 1-5.
8. A readable storage medium, characterized in that the readable storage medium has stored thereon a program or instructions which, when executed by a processor, implement the steps of the video surveillance method according to any of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110212326.0A CN113014876B (en) | 2021-02-25 | 2021-02-25 | Video monitoring method and device, electronic equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110212326.0A CN113014876B (en) | 2021-02-25 | 2021-02-25 | Video monitoring method and device, electronic equipment and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113014876A CN113014876A (en) | 2021-06-22 |
| CN113014876B true CN113014876B (en) | 2023-06-02 |
Family
ID=76386804
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110212326.0A Active CN113014876B (en) | 2021-02-25 | 2021-02-25 | Video monitoring method and device, electronic equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113014876B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113808376B (en) * | 2021-09-18 | 2023-06-16 | 中国工商银行股份有限公司 | Self-service teller machine equipment safety monitoring method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103280041A (en) * | 2013-05-08 | 2013-09-04 | 广东电网公司珠海供电局 | Monitoring method and monitoring system for automatic deploying virtual electronic fence |
| CN108900812A (en) * | 2018-07-20 | 2018-11-27 | 合肥云联电子科技有限公司 | It is a kind of based on the indoor video monitoring method remotely controlled |
| CN111131783A (en) * | 2019-12-27 | 2020-05-08 | 泰斗微电子科技有限公司 | Monitoring method and device based on electronic fence, terminal equipment and storage medium |
| CN111163294A (en) * | 2020-01-03 | 2020-05-15 | 重庆特斯联智慧科技股份有限公司 | Building safety channel monitoring system and method for artificial intelligence target recognition |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100139576A1 (en) * | 2008-11-04 | 2010-06-10 | Dt Systems, Inc. | Electronic fence system |
| CN106600951B (en) * | 2017-01-19 | 2019-01-29 | 西安电子科技大学 | It is a kind of to be equipped with vehicle monitor terminal vehicle monitoring system and its method |
-
2021
- 2021-02-25 CN CN202110212326.0A patent/CN113014876B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103280041A (en) * | 2013-05-08 | 2013-09-04 | 广东电网公司珠海供电局 | Monitoring method and monitoring system for automatic deploying virtual electronic fence |
| CN108900812A (en) * | 2018-07-20 | 2018-11-27 | 合肥云联电子科技有限公司 | It is a kind of based on the indoor video monitoring method remotely controlled |
| CN111131783A (en) * | 2019-12-27 | 2020-05-08 | 泰斗微电子科技有限公司 | Monitoring method and device based on electronic fence, terminal equipment and storage medium |
| CN111163294A (en) * | 2020-01-03 | 2020-05-15 | 重庆特斯联智慧科技股份有限公司 | Building safety channel monitoring system and method for artificial intelligence target recognition |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113014876A (en) | 2021-06-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108268867B (en) | License plate positioning method and device | |
| US9189867B2 (en) | Adaptive image processing apparatus and method based in image pyramid | |
| CN113052029A (en) | Abnormal behavior supervision method and device based on action recognition and storage medium | |
| Pujol et al. | A soft computing approach to violence detection in social media for smart cities | |
| CN110956081B (en) | Method and device for identifying position relationship between vehicle and traffic marking and storage medium | |
| CN111814776B (en) | Image processing method, device, server and storage medium | |
| CN112434566A (en) | Passenger flow statistical method and device, electronic equipment and storage medium | |
| CN104573680A (en) | Image detection method, image detection device and traffic violation detection system | |
| CN114445768A (en) | Target identification method and device, electronic equipment and storage medium | |
| CN113014876B (en) | Video monitoring method and device, electronic equipment and readable storage medium | |
| CN112884147B (en) | Neural network training method, image processing method, device and electronic equipment | |
| CN114494990A (en) | Target detection method, system, terminal equipment and storage medium | |
| CN111444788B (en) | Behavior recognition method, apparatus and computer storage medium | |
| CN116824135A (en) | Atmospheric natural environment test industrial product identification and segmentation method based on machine vision | |
| CN111862023A (en) | Pantograph structure abnormity detection method based on computer vision | |
| CN114913470B (en) | Event detection method and device | |
| CN115147756A (en) | Video stream processing method and device, electronic equipment and storage medium | |
| CN112348011B (en) | Vehicle damage assessment method and device and storage medium | |
| CN113947744A (en) | Fire image detection method, system, equipment and storage medium based on video | |
| CN110674780B (en) | Scene change detection method, device, equipment and readable storage medium | |
| CN111476132A (en) | Video scene recognition method and device, electronic equipment and storage medium | |
| CN111950501A (en) | Obstacle detection method and device and electronic equipment | |
| CN111275045A (en) | Method and device for identifying image subject, electronic equipment and medium | |
| CN112084881B (en) | Cross-domain pedestrian re-identification method and device and storage medium | |
| KR101951900B1 (en) | Method and Apparatus for Detecting Object in an Image |
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 | ||
| CB02 | Change of applicant information |
Address after: Room 101, floors 1-3, building 14, North District, yard 9, dongran North Street, Haidian District, Beijing 100029 Applicant after: CHINA TOWER Co.,Ltd. Address before: 100142 19th floor, 73 Fucheng Road, Haidian District, Beijing Applicant before: CHINA TOWER Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |