CN113096338A - Community safety early warning method based on intelligent lamp pole - Google Patents
Community safety early warning method based on intelligent lamp pole Download PDFInfo
- Publication number
- CN113096338A CN113096338A CN202110387587.6A CN202110387587A CN113096338A CN 113096338 A CN113096338 A CN 113096338A CN 202110387587 A CN202110387587 A CN 202110387587A CN 113096338 A CN113096338 A CN 113096338A
- Authority
- CN
- China
- Prior art keywords
- frame
- human body
- pixel
- early warning
- method based
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 16
- 238000012549 training Methods 0.000 claims abstract description 8
- 238000013527 convolutional neural network Methods 0.000 claims description 10
- 238000000605 extraction Methods 0.000 claims description 6
- 239000013598 vector Substances 0.000 claims description 5
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 238000005457 optimization Methods 0.000 claims description 4
- 238000005070 sampling Methods 0.000 claims description 4
- 238000012706 support-vector machine Methods 0.000 claims description 4
- 239000012465 retentate Substances 0.000 claims description 3
- 230000014759 maintenance of location Effects 0.000 claims description 2
- 230000002159 abnormal effect Effects 0.000 abstract description 7
- 238000012544 monitoring process Methods 0.000 abstract description 4
- 238000010801 machine learning Methods 0.000 abstract description 2
- 238000007781 pre-processing Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 12
- 238000001514 detection method Methods 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 238000005286 illumination Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 238000009792 diffusion process Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000010223 real-time analysis Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B13/00—Burglar, theft or intruder alarms
- G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
- G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
- G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
- G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/04—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons
- G08B21/0407—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B5/00—Visible signalling systems, e.g. personal calling systems, remote indication of seats occupied
- G08B5/22—Visible signalling systems, e.g. personal calling systems, remote indication of seats occupied using electric transmission; using electromagnetic transmission
- G08B5/36—Visible signalling systems, e.g. personal calling systems, remote indication of seats occupied using electric transmission; using electromagnetic transmission using visible light sources
- G08B5/38—Visible signalling systems, e.g. personal calling systems, remote indication of seats occupied using electric transmission; using electromagnetic transmission using visible light sources using flashing light
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Multimedia (AREA)
- Psychiatry (AREA)
- Psychology (AREA)
- Social Psychology (AREA)
- General Health & Medical Sciences (AREA)
- Gerontology & Geriatric Medicine (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Electromagnetism (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a community safety early warning method based on a smart lamp pole, which belongs to the technical field of smart lamp poles and is used for preprocessing an image acquired by a terminal device, analyzing a moving object, identifying a human body, detecting dangerous factors and emergency events and realizing community safety early warning; the intelligent lamp pole is used as a carrier to interact with the server, and the machine learning training classifier is used for quickly analyzing and detecting abnormal events, so that real-time monitoring and management of communities are realized, the living quality of community residents is improved, and accident potential is eliminated in advance.
Description
Technical Field
The invention belongs to the technical field of intelligent lamp poles, and particularly relates to a community safety early warning method based on an intelligent lamp pole.
Background
The community is used as the most main living environment form in the city, and the security and protection are very important. The traditional community security system and the lighting system are often separated, and sudden dangers such as falling down of old people, high-altitude object throwing, vehicle collision and the like cannot be identified in real time, so that the security system faces the problems of scattered distribution, insufficient definition, limited storage space, incapability of real-time early warning and the like.
The development of a new generation of information technology promotes the construction of smart cities to be accelerated continuously, and the smart lamp pole is used as an important infrastructure for the construction of the smart cities, so that urban illumination and security, especially community illumination and security are integrated, and an optimal foothold is provided for intelligent security. Although the wisdom lamp pole can integrate video monitoring and community illumination at present, nevertheless the distance universe covers, whole controllable still has very big difference. In the aspect of community safety, the utilization rate of video resources is low, image data processing is slow, the after-the-fact response is favored, and the capability of preventing in advance is lacked.
Disclosure of Invention
The invention provides a community safety early warning method based on a smart lamp pole.
In order to achieve the purpose, the invention adopts the following technical scheme:
a community safety early warning method based on a smart lamp pole comprises the following steps:
s1: acquiring real-time video image data (the video data encoding format is H.264/H.265) acquired by a front-end ball camera, and performing frame image extraction, background difference, denoising and defogging on the image data to obtain a frame image set;
s2: using the first video frame as a background video model frame, and detecting a moving object in a subsequent frame image set by using a Vibe algorithm;
s3: acquiring an outer frame information set of each frame based on an OpenCV outline outer frame function to identify whether a moving target is included; if the data contains the moving target, further tracking the moving target until the target leaves the camera area;
s4: delivering the frame image set to a trained classifier for identification, and classifying the output result of the classifier into a human body or a non-human body;
s5: detecting the retentate after receiving the result output by the classifier as a human body;
s6: if the human body falls down accidentally, the human body stays on the ground for a short time and is marked as a detention state; extracting the characteristics of the detention area, and classifying and judging whether the moving target lands on the ground actively or falls over accidentally by using a support vector machine;
s7: if the moving object is judged to fall accidentally, an early warning is given out, the following data of the preset path of the server are sent to related management personnel, meanwhile, the illuminating lamp of the nearby lamp post is controlled to flicker, and the LED screen flickers in red and blue.
In the above steps, the Vibe algorithm in step S2 specifically includes the following steps:
the first step is as follows: extracting a first frame of a video as a background model, establishing the background model and initializing;
the second step is that: comparing the pixel values with samples of a background model frame by frame and pixel by pixel to classify the pixel values, and classifying the pixel values exceeding a threshold value as a foreground;
the third step: and sampling and updating the background model at random time, and randomly replacing a sample value of the pixel sample set by a new pixel value (a memoryless updating strategy) each time the background model of the pixel is determined to be required to be updated, and simultaneously updating the sample value of an adjacent pixel by the newly adopted pixel (a diffusion mechanism).
The training of the classifier in step S4 specifically includes the following steps: the training of the classifier in step S4 specifically includes the following steps: initializing a human body data set, defining the data type of a convolutional neural network and the network depth h and width w of the convolutional neural network, setting a loss function after a CNN network is built, adopting a Sigmod function and an optimizer thereof, further inputting a human body image data set at least comprising 1000 marked heads and trunks, circularly traversing a data iterator, providing the input to the neural network, and updating the weight, thereby achieving the purpose of optimization.
The stay region feature extraction in step S6 further includes: extracting characteristics of the detention area, obtaining human joint length information by using OpenPose based on OpenCV, wherein the extracted characteristics comprise human height and human head height, in addition, the height-width ratio of an OpenCV external frame and the area of the detention area (external frame) need to be extracted, further judging whether an object is close to the ground by using an SVM classification mode, calculating detention time, and if the detention time exceeds a threshold value, further judging whether the moving human body falls accidentally or lands initiatively by calculating whether a feature vector is lower than a set threshold value.
The invention has the beneficial effects that: the invention provides a community safety early warning method based on a smart lamp pole, which integrates security and lighting equipment in a community, improves the utilization rate of community resources and reduces the energy consumption; the method comprises the following steps of collecting video image data by utilizing a hemispherical camera, rapidly analyzing dangerous factors such as falling of old people, falling objects from high altitudes, vehicle collision and the like or group events, and early warning in real time, so that the life happiness index of community residents is improved; according to the intelligent lamp pole mounted video monitoring system, the video monitoring equipment mounted on the intelligent lamp pole is effectively utilized, the image data are transmitted to the server side to be processed, the accuracy and the efficiency of real-time analysis of the video data are improved by adopting the machine learning training classifier, and the living quality and the safety factor of community residents are improved.
Drawings
FIG. 1 is a schematic diagram of an intelligent light pole according to an embodiment of the present invention;
FIG. 2 is a flow diagram of an abnormal event analysis and detection function according to an embodiment of the present invention;
in the figure, 1 is the light, 2 is the crossbeam, and 3 is the camera, and 4 is the lamp pole, and 5 is the support frame, and 6 is the LED display screen, and 7 is temperature and humidity sensor, 8 are the RFID inductor, and 9 are the lamp stand.
Detailed Description
The invention is described in further detail below with reference to the following figures and specific examples:
as shown in FIG. 2, a community safety early warning method based on a smart lamp pole comprises the following steps:
s1: acquiring real-time video image data (the video data encoding format is H.264/H.265) acquired by a front-end ball camera, and performing frame image extraction, background difference, denoising and defogging on the image data to obtain a frame image set;
s2: using the first video frame as a background video model frame, and detecting a moving object in a subsequent frame image set by using a Vibe algorithm;
s3: acquiring an outer frame information set of each frame based on an OpenCV outline outer frame function to identify whether a moving target is included; if the data contains the moving target, further tracking the moving target until the target leaves the camera area;
s4: delivering the frame image set to a trained classifier for identification, and classifying the output result of the classifier into a human body or a non-human body;
s5: detecting the retentate after receiving the result output by the classifier as a human body;
s6: if the human body falls down accidentally, the human body stays on the ground for a short time and is marked as a detention state; extracting the characteristics of the detention area, and classifying and judging whether the moving target lands on the ground actively or falls over accidentally by using a support vector machine;
s7: if the moving object is judged to fall accidentally, an early warning is given out, the following data of the preset path of the server are sent to related management personnel, meanwhile, the illuminating lamp of the nearby lamp post is controlled to flicker, and the LED screen flickers in red and blue.
In the above steps, the Vibe algorithm in step S2 specifically includes the following steps:
the first step is as follows: extracting a first frame of a video as a background model, establishing the background model and initializing;
the second step is that: comparing the pixel values with samples of a background model frame by frame and pixel by pixel to classify the pixel values, and classifying the pixel values exceeding a threshold value as a foreground;
the third step: and sampling and updating the background model at random time, and randomly replacing a sample value of the pixel sample set by a new pixel value (a memoryless updating strategy) each time the background model of the pixel is determined to be required to be updated, and simultaneously updating the sample value of an adjacent pixel by the newly adopted pixel (a diffusion mechanism).
The training of the classifier in step S4 specifically includes the following steps: the training of the classifier in step S4 specifically includes the following steps: initializing a human body data set, defining the data type of a convolutional neural network and the network depth h and width w of the convolutional neural network, setting a loss function after a CNN network is built, adopting a Sigmod function and an optimizer thereof, further inputting a human body image data set at least comprising 1000 marked heads and trunks, circularly traversing a data iterator, providing the input to the neural network, and updating the weight, thereby achieving the purpose of optimization.
The stay region feature extraction in step S6 further includes: extracting characteristics of the detention area, obtaining human joint length information by using OpenPose based on OpenCV, wherein the extracted characteristics comprise human height and human head height, in addition, the height-width ratio of an OpenCV external frame and the area of the detention area (external frame) need to be extracted, further judging whether an object is close to the ground by using an SVM classification mode, calculating detention time, and if the detention time exceeds a threshold value, further judging whether the moving human body falls accidentally or lands initiatively by calculating whether a feature vector is lower than a set threshold value.
The method is completed based on the device shown in fig. 1, and the device comprises an intelligent lamp pole and a server-side abnormal event analysis and detection function module;
the intelligent lamp post comprises a lamp post 4 arranged in a community, an illuminating lamp 1 mounted on the lamp post, an LED display screen 6, a camera 3, an RFID sensor 8 and a temperature and humidity sensor 7; the lamp post 4 is arranged on the lamp holder 9, and a cross beam 2 required for hanging two illuminating lamps and a support frame 5 required for supporting the LED display screen are arranged above the lamp post 4;
the lamp is provided with two lamps, one lamp is positioned on the left side of the top end of the lamp post, and the other lamp is positioned on the right side; the illuminating lamp is provided with a fan-shaped or other-shaped lampshade; the camera is a dome camera provided with a holder capable of rotating the camera, and the dome camera can adjust the shooting direction through remote control; the LED screen is a single-side normally-bright rectangular screen, is fixed on the lamp post, and is convenient to watch with the light emitting surface inclined downwards; the RFID sensor is rectangular and is fixed on the lamp post; the temperature and humidity sensor is arranged above the RFID sensor;
the server-side abnormal event analysis and detection functional module is communicated with a front-end camera on the basis of wireless communication equipment and acquires video data; transmitting the video data to a server end through a twisted pair, and detecting abnormal events by adopting a community safety early warning method; if the abnormality is detected, the module sends alarm information, triggers the server to control the LED screen to display early warning information, and simultaneously sends preset text information to the mobile phone of the attendant.
Taking human body falling identification as an example, the abnormal event analysis and detection function is explained, a video camera and an illuminating lamp mounted on a smart lamp pole are used for collecting high-quality video image data, an abnormal event analysis and detection function module configured by a server is used for preprocessing the video data collected by the camera of the ball machine, and a frame image set is output;
further carrying out moving object detection on a subsequent frame image collection by using a Vibe algorithm, using a first video frame as a background video model frame, comparing pixel values with samples of a background model frame by frame and pixel by pixel to classify the pixel values, and classifying the pixel values exceeding a threshold value as a foreground; at the same time, sampling and updating the background model at random time, and randomly replacing a sample value of the pixel point sample set by a new pixel value each time the background model of the pixel point is determined to be updated, and simultaneously updating the sample value of the adjacent pixel point by the newly adopted pixel point;
and acquiring an external frame information set of each frame based on an OpenCV outline external frame function, and if a moving target is detected, delivering the frame image set to a trained classifier for human body identification. And inputting at least 1000 human body pictures during training of the classifier, marking the head and the trunk of the human body, circularly traversing the data iterator after defining the convolutional neural network and the loss function, and providing the input to the neural network for optimization. If the classifier outputs that the human body is not a human body, returning to continuously detect the moving target; if the output of the classifier is a human body, carrying out retention detection on the moving target;
the method comprises the steps of marking that a human body stays on the ground for a short time to be in a detention state, using a support vector machine to extract features of a detention area, wherein the extracted feature vectors comprise the height of the human body, the height-width ratio of an external frame and the like, and judging whether the human body lands accidentally or not by calculating whether the feature vectors are lower than a threshold value. If the extracted human body is detected to touch the ground accidentally, an early warning is given out, and if the extracted human body is detected to touch the ground intentionally, the detention detection is carried out again.
Fig. 1 is a structural diagram of a smart lamp post according to an embodiment of the invention, wherein the smart lamp post is mounted with two illuminating lamps 1, a camera 3, an LED display screen 6, a temperature and humidity sensor 7 and an RFID identification module 8. The lamp post 4 is arranged on the lamp holder 9, and a cross beam 2 for hanging two illuminating lamps and a support frame 5 for supporting the LED display screen are arranged above the lamp post.
While the invention has been described in terms of its preferred embodiments, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention.
Claims (4)
1. The utility model provides a community's safety precaution method based on wisdom lamp pole which characterized in that includes following step:
s1: acquiring real-time video image data acquired by a camera of a front-end ball machine, and performing frame image extraction, background difference, denoising and defogging on the image data to obtain a frame image set;
s2: using the first video frame as a background video model frame, and detecting a moving object in a subsequent frame image set by using a Vibe algorithm;
s3: acquiring an outer frame information set of each frame based on an OpenCV outline outer frame function to identify whether a moving target is included; if the data contains the moving target, further tracking the moving target until the target leaves the camera area;
s4: delivering the frame image set to a trained classifier for identification, and classifying the output result of the classifier into a human body or a non-human body;
s5: detecting the retentate after receiving the result output by the classifier as a human body;
s6: if the human body falls down accidentally, the human body stays on the ground for a short time and is marked as a detention state; extracting the characteristics of the detention area, and classifying and judging whether the moving target lands on the ground actively or falls over accidentally by using a support vector machine;
s7: if the moving object is judged to fall accidentally, an early warning is given out, the following data of the preset path of the server are sent to related management personnel, meanwhile, the illuminating lamp of the nearby lamp post is controlled to flicker, and the LED screen flickers in red and blue.
2. The community safety early warning method based on the intelligent lamp pole as claimed in claim 1, wherein the Vibe algorithm in the step S2 specifically comprises the following steps:
the first step is as follows: extracting a first frame of a video as a background model, establishing the background model and initializing;
the second step is that: comparing the pixel values with samples of a background model frame by frame and pixel by pixel to classify the pixel values, and classifying the pixel values exceeding a threshold value as a foreground;
the third step: and sampling and updating the background model at random time, and randomly replacing a sample value of the pixel point sample set by a new pixel value when the background model of the pixel point is determined to be updated each time, and simultaneously updating the sample value of the adjacent pixel point by the newly adopted pixel point.
3. The community safety precaution method based on intelligent lamp posts as claimed in claim 1, wherein the step S4 of training the classifier specifically comprises the following steps: initializing a human body data set, defining the data type of a convolutional neural network and the network depth h and width w of the convolutional neural network, setting a loss function after a CNN network is built, further inputting a human body image data set at least containing 1000 marked heads and trunks by adopting a Sigmod function and an optimizer thereof, circularly traversing a data iterator, providing the input to the neural network, and updating the weight, thereby achieving the purpose of optimization.
4. The community safety early warning method based on intelligent lamp pole according to claim 1, wherein the extraction of the retention area features in the step S6 specifically comprises the following steps: extracting characteristics of the detention area, obtaining human joint length information by using OpenPose based on OpenCV, wherein the extracted characteristics comprise human height and human head height, in addition, the height-width ratio of an OpenCV external frame and the area of the detention area (external frame) need to be extracted, further judging whether an object is close to the ground by using an SVM classification mode, calculating detention time, and if the detention time exceeds a threshold value, further judging whether the moving human body falls accidentally or lands initiatively by calculating whether a feature vector is lower than a set threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110387587.6A CN113096338A (en) | 2021-04-09 | 2021-04-09 | Community safety early warning method based on intelligent lamp pole |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110387587.6A CN113096338A (en) | 2021-04-09 | 2021-04-09 | Community safety early warning method based on intelligent lamp pole |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113096338A true CN113096338A (en) | 2021-07-09 |
Family
ID=76677310
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110387587.6A Pending CN113096338A (en) | 2021-04-09 | 2021-04-09 | Community safety early warning method based on intelligent lamp pole |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113096338A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116303867A (en) * | 2023-05-19 | 2023-06-23 | 江西省气象灾害应急预警中心(江西省突发事件预警信息发布中心) | Early warning area identification method, system, computer and readable storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108198207A (en) * | 2017-12-22 | 2018-06-22 | 湖南源信光电科技股份有限公司 | Multiple mobile object tracking based on improved Vibe models and BP neural network |
| CN110321780A (en) * | 2019-04-30 | 2019-10-11 | 苏州大学 | Exception based on spatiotemporal motion characteristic falls down behavioral value method |
| CN111724566A (en) * | 2020-05-20 | 2020-09-29 | 同济大学 | Pedestrian falling detection method and device based on intelligent lamp pole video monitoring system |
| CN112578726A (en) * | 2021-01-06 | 2021-03-30 | 常州百芝龙智慧科技有限公司 | Human body abnormity monitoring equipment for independently learning human body behavior habits |
-
2021
- 2021-04-09 CN CN202110387587.6A patent/CN113096338A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108198207A (en) * | 2017-12-22 | 2018-06-22 | 湖南源信光电科技股份有限公司 | Multiple mobile object tracking based on improved Vibe models and BP neural network |
| CN110321780A (en) * | 2019-04-30 | 2019-10-11 | 苏州大学 | Exception based on spatiotemporal motion characteristic falls down behavioral value method |
| CN111724566A (en) * | 2020-05-20 | 2020-09-29 | 同济大学 | Pedestrian falling detection method and device based on intelligent lamp pole video monitoring system |
| CN112578726A (en) * | 2021-01-06 | 2021-03-30 | 常州百芝龙智慧科技有限公司 | Human body abnormity monitoring equipment for independently learning human body behavior habits |
Non-Patent Citations (1)
| Title |
|---|
| 吴亚会: "基于视频的人体异常行为检测与分析研究", 《中国优秀博硕士学位论文全文数据库(硕士) 信息科技辑》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116303867A (en) * | 2023-05-19 | 2023-06-23 | 江西省气象灾害应急预警中心(江西省突发事件预警信息发布中心) | Early warning area identification method, system, computer and readable storage medium |
| CN116303867B (en) * | 2023-05-19 | 2023-08-22 | 江西省气象灾害应急预警中心(江西省突发事件预警信息发布中心) | Early warning area identification method, system, computer and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110070033B (en) | Method for detecting wearing state of safety helmet in dangerous working area in power field | |
| CN110745704B (en) | Tower crane early warning method and device | |
| CN108109385B (en) | System and method for identifying and judging dangerous behaviors of power transmission line anti-external damage vehicle | |
| CN109377703A (en) | A kind of forest fireproofing early warning system and its method based on machine vision | |
| CN106128022B (en) | A kind of wisdom gold eyeball identification violent action alarm method | |
| CN108307146A (en) | A kind of ultra-high-tension power transmission line Security Vulnerability Detecting System and method | |
| CN109672863A (en) | A kind of construction personnel's safety equipment intelligent monitoring method based on image recognition | |
| CN106446926A (en) | Transformer station worker helmet wear detection method based on video analysis | |
| CN109255298A (en) | Safety helmet detection method and system in dynamic background | |
| CN102542246A (en) | Abnormal face detection method for ATM (Automatic Teller Machine) | |
| CN111523397B (en) | Intelligent lamp post visual identification device, method and system and electronic equipment thereof | |
| CN112396658A (en) | Indoor personnel positioning method and positioning system based on video | |
| CN103069434A (en) | Multi-mode video event indexing | |
| CN102201146A (en) | Active infrared video based fire smoke detection method in zero-illumination environment | |
| CN113269091A (en) | Personnel trajectory analysis method, equipment and medium for intelligent park | |
| CN108345842A (en) | A kind of processing method based on big data | |
| CN109344842A (en) | A kind of pedestrian's recognition methods again based on semantic region expression | |
| CN110570461A (en) | method for realizing automatic person-searching tracking by combining community lighting network and cloud platform community population information base | |
| CN111401310B (en) | Kitchen sanitation safety supervision and management method based on artificial intelligence | |
| CN111046728A (en) | Straw combustion detection method based on characteristic pyramid network | |
| CN104134067A (en) | Road vehicle monitoring system based on intelligent visual Internet of Things | |
| CN108965804A (en) | A kind of video structural technology for city security protection | |
| CN113191273A (en) | Oil field well site video target detection and identification method and system based on neural network | |
| CN113096338A (en) | Community safety early warning method based on intelligent lamp pole | |
| CN115169673A (en) | Intelligent campus epidemic risk monitoring and early warning system and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210709 |
|
| RJ01 | Rejection of invention patent application after publication |