CN110992645A - Mobile vendor detection and alarm system in dynamic scene - Google Patents
Mobile vendor detection and alarm system in dynamic scene Download PDFInfo
- Publication number
- CN110992645A CN110992645A CN201911244134.7A CN201911244134A CN110992645A CN 110992645 A CN110992645 A CN 110992645A CN 201911244134 A CN201911244134 A CN 201911244134A CN 110992645 A CN110992645 A CN 110992645A
- Authority
- CN
- China
- Prior art keywords
- mobile vendor
- detection
- module
- mobile
- controller
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/02—Protocols based on web technology, e.g. hypertext transfer protocol [HTTP]
- H04L67/025—Protocols based on web technology, e.g. hypertext transfer protocol [HTTP] for remote control or remote monitoring of applications
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/55—Push-based network services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
- H04L69/16—Implementation or adaptation of Internet protocol [IP], of transmission control protocol [TCP] or of user datagram protocol [UDP]
- H04L69/161—Implementation details of TCP/IP or UDP/IP stack architecture; Specification of modified or new header fields
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Computer Security & Cryptography (AREA)
- Medical Informatics (AREA)
- Image Analysis (AREA)
- Alarm Systems (AREA)
Abstract
A mobile vendor detection and alarm system in a dynamic scene is disclosed, wherein a mobile vendor detection module comprises a mobile vendor detection device, a controller and a power module, the mobile vendor detection device is respectively connected with a video acquisition module, the controller and the power module, the controller is connected with a prompt alarm module and other push platforms, the mobile vendor detection device is used for receiving and processing image information sent by the video acquisition module through a video and transmitting an analysis result to the controller, and the controller starts the prompt alarm module and transmits the analysis result to other push platforms; the invention utilizes the target detection technology based on deep learning, has advanced algorithm, can meet the real-time detection requirement, has stronger generalization performance in different scenes, and can be used for mobile vendor detection early warning tasks in various scenes such as static, dynamic and the like; meanwhile, the result can be analyzed without transmitting a large amount of video data to a background server, so that the cost is saved, the structure is simple, and the installation is easy.
Description
Technical Field
The invention relates to the technical field of image processing and intelligent video monitoring, in particular to a mobile vendor detection and alarm system in a dynamic scene.
Background
With the rapid development of social economy, the continuous acceleration of urban construction, the frequent movement of foreign population, the closing, stopping, merging and turning of enterprises and other factors, the unlicensed mobile vendors are often prohibited, become the focus of complaints of the masses and the social reflection, become the key and difficult points of urban management work, and have become very painful for urban managers in various regions.
At present, some cities mainly adopt a mode of combining fixed-point monitoring and mobile patrol to supervise mobile vendors, but the fixed-point monitoring mode cannot completely cover all corners of the city, the point positions are limited, and the monitoring scenes are limited; the mobile patrol mode can not find problems in time, needs to consume a large amount of manpower, and is easy to conflict with mobile vendors, so that personal safety is endangered.
Disclosure of Invention
The present invention is directed to a system for detecting and alarming a mobile vendor in a dynamic scene, so as to solve the above problems in the prior art.
The technical problem solved by the invention is realized by adopting the following technical scheme:
the utility model provides a mobile vendor detects and alarm system in dynamic scene, including video acquisition module, mobile vendor detection module and suggestion alarm module, wherein, mobile vendor detection module is including mobile vendor detection device, controller and power module, mobile vendor detection device respectively with video acquisition module, controller and power module are connected, controller and suggestion alarm module, other propelling movement platforms are connected, mobile vendor detection device is used for receiving the image information of handling video acquisition module video transmission and conveys the analysis result to the controller, the controller starts suggestion alarm module, transmit the analysis result to other propelling movement platforms simultaneously.
In the invention, the video acquisition module comprises network cameras which are connected with each other.
In the invention, the network camera is connected with the mobile vendor detection device through a TCP/IP protocol.
In the invention, the power supply module is connected with the network camera and the controller, and the output molding interface of the power supply module is a plug.
In the invention, the video acquisition module and the mobile vendor detection module are installed on a bus or a mobile law enforcement vehicle in urban management.
In the invention, a memory is arranged in the mobile vendor detection device, and the memory is a detachable hard disk.
In the invention, a mobile vendor detection algorithm based on deep learning is loaded in the mobile vendor detection device and is divided into two steps, wherein in the first step, a detection model is trained to obtain detection model parameters; the second step utilizes the detection model that trains well to carry out mobile vendor's detection, and the first step training detection model includes:
step 1), establishing a mobile vendor picture data set, and establishing a picture data set with a mobile vendor picture in a mode of manual shooting, internet resource and image enhancement;
step 2) data annotation of the data set, wherein a marking tool is adopted to carry out mobile vendor annotation on the pictures in the picture data set to generate an annotated file;
step 3) mobile vendor detection model training based on deep learning, which comprises (1) constructing a network model RefineDet; (2) training the network by using a cross entropy loss function; (3) training the trained network in the step (2) by using a focus loss function, and selecting a plurality of balance factors and a plurality of modulation factors of the focus loss function in a form of a controlled variable method so as to obtain a plurality of trained networks; (4) testing the network obtained in the step (3) by using a test set, and selecting the network corresponding to the highest average detection accuracy as a final mobile vendor detection model;
the second step of mobile vendor detection process is as follows: firstly, preprocessing a video acquired by a video acquisition module to obtain a video frame picture, then inputting the video frame picture into a trained mobile vendor detection model for detection, and analyzing whether a mobile vendor exists in the video frame picture.
In the invention, a wireless signal transmitter is arranged in the controller, and the controller is wirelessly connected with other pushing platforms to push information by the other pushing platforms.
In the invention, the prompt alarm module comprises a voice prompter arranged on the network camera and a message pusher arranged on other platforms.
In the invention, when the system works, the system comprises a video acquisition module and a mobile vendor detection module which are arranged on a bus or a mobile law enforcement bus of urban management, after the system is powered on, the video acquisition module acquires video image information in real time, converts the video image information into digital information and transmits the digital information to the mobile vendor detection module through a TCP/IP protocol, the information such as whether mobile vendor exists is obtained through deep learning algorithm analysis, then the information is transmitted to a prompt alarm module through a control module, and a voice prompt or a message pusher pushes messages through the voice prompt.
Has the advantages that: the invention utilizes the target detection technology based on deep learning, is different from the mobile vendor detection method based on static scenes, has advanced algorithm, can meet the real-time detection requirement, has stronger generalization performance in different scenes, and can be used for mobile vendor detection early warning tasks in various scenes such as static scenes, dynamic scenes and the like; meanwhile, the result can be analyzed without transmitting a large amount of video data to a background server, so that the cost is saved, the structure is simple, and the installation is easy.
Drawings
FIG. 1 is a flow chart illustrating a preferred embodiment of the present invention.
FIG. 2 is a flow chart of the algorithm for mobile vendor detection based on deep learning according to the preferred embodiment of the present invention.
FIG. 3 is a diagram illustrating the detection results of a mobile vendor in a real scene according to the preferred embodiment of the present invention.
Detailed Description
In order to make the technical means, the creation characteristics, the achievement purposes and the effects of the invention easy to understand, the invention is further explained below by combining the specific drawings.
Referring to fig. 1, the system for detecting and alarming mobile vendors in a dynamic scene includes a video acquisition module, a mobile vendor detection module and a prompt alarm module, wherein the video acquisition module is electrically connected to the mobile vendor detection module, and the mobile vendor detection module is electrically connected to the prompt alarm module; the mobile vendor detection device is used for receiving and processing video image information sent by the video acquisition module and transmitting an analysis result to the controller, and the controller starts the prompt alarm module and can transmit the analysis result to other pushing platforms;
further, the video acquisition module and the mobile vendor detection module are installed on a bus or a mobile law enforcement bus in urban management;
further, the network camera is connected with a mobile vendor detection device of the mobile vendor detection module through a TCP/IP protocol;
furthermore, a memory is arranged in the mobile vendor detection device, and the memory is a detachable hard disk;
further, a mobile vendor detection algorithm based on deep learning is loaded in the mobile vendor detection device, as shown in fig. 2, the algorithm is divided into two steps, and in the first step, a detection model is trained to obtain detection model parameters; the second step utilizes the detection model that trains well to carry out mobile vendor's detection, and the first step training detection model includes:
step 1), establishing a mobile vendor picture data set, and establishing a picture data set with a mobile vendor picture in a mode of manual shooting, internet resource and image enhancement;
step 2) data annotation of the data set, wherein a marking tool is adopted to carry out mobile vendor annotation on the pictures in the picture data set to generate an annotated file;
step 3) mobile vendor detection model training based on deep learning, which comprises (1) constructing a network model RefineDet; (2) training the network by using a cross entropy loss function; (3) training the trained network in the step (2) by using a focus loss function, and selecting a plurality of balance factors and a plurality of modulation factors of the focus loss function in a form of a controlled variable method so as to obtain a plurality of trained networks; (4) testing the network obtained in the step (3) by using a test set, and selecting the network corresponding to the highest average detection accuracy as a final mobile vendor detection model;
the second step of mobile vendor detection process is as follows: firstly, preprocessing a video acquired by a video acquisition module to obtain a video frame picture, then inputting the video frame picture into a trained mobile vendor detection model for detection, and analyzing whether a mobile vendor exists in the video frame picture, as shown in fig. 3;
further, a wireless signal transmitter is arranged in the controller, the controller is wirelessly connected with other pushing platforms, and information is pushed by the other pushing platforms;
further, the power supply module supplies power for the network camera, the mobile vendor detection device and the controller, and an output forming interface of the power supply module is a plug;
further, the prompt alarm module comprises a voice prompter arranged on the network camera and a message pusher arranged on other platforms;
in this embodiment, during operation, will include that video acquisition module and mobile vendor detection module install on bus or the mobile law enforcement car of urban management, after the switch on, video acquisition module gathers video image information in real time to convert video image information to digital information and transmit to mobile vendor detection module through TCP/IP agreement, obtain through the analysis of deep learning algorithm whether have information such as mobile vendor, then give information transfer to suggestion alarm module through control module, carry out voice prompt or message pusher propelling movement message by voice prompt.
The foregoing shows and describes the general principles and broad features of the present invention and advantages thereof. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are described in the specification and illustrated only to illustrate the principle of the present invention, but that various changes and modifications may be made therein without departing from the spirit and scope of the present invention, which fall within the scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (10)
1. The utility model provides a mobile vendor detection and alarm system in dynamic scene, including video acquisition module, mobile vendor detection module and suggestion alarm module, a serial communication port, mobile vendor detection module includes mobile vendor detection device, controller and power module, mobile vendor detection device respectively with video acquisition module, controller and power module are connected, controller and suggestion alarm module, other propelling movement platforms are connected, mobile vendor detection device that is loaded with mobile vendor detection algorithm is used for receiving the image information of handling video acquisition module video transmission and conveys the analysis result to the controller, the controller starts suggestion alarm module, transmit the analysis result to other propelling movement platforms simultaneously.
2. The system of claim 1, wherein the video capture module comprises a network camera coupled to one another.
3. The system of claim 2, wherein the power module is connected to the webcam and the controller, and the output form interface of the power module is a plug.
4. The system of claim 1, wherein the video capture module and the mobile vendor detection module are installed on a bus or a mobile law enforcement vehicle in urban management.
5. The system of claim 1, wherein a memory is disposed in the mobile vendor detection and alarm system, and the memory is a removable hard disk.
6. The system of claim 1, wherein the controller is configured with a wireless transmitter, and the controller is wirelessly connected to other push platforms.
7. The system of claim 1, wherein the alert module comprises a voice prompt on a webcam and a message pusher on another platform.
8. The system of claim 1, wherein the mobile vendor detection and alarm system in a dynamic scene is loaded with a mobile vendor detection algorithm based on deep learning, the algorithm is divided into two steps, the first step trains a detection model to obtain detection model parameters; and the second step of using the trained detection model to carry out mobile vendor detection.
9. The system of claim 8, wherein the first training of the detection model comprises:
step 1), establishing a mobile vendor picture data set, and establishing a picture data set with a mobile vendor picture in a mode of manual shooting, internet resource and image enhancement;
step 2) data annotation of the data set, wherein a marking tool is adopted to carry out mobile vendor annotation on the pictures in the picture data set to generate an annotated file;
step 3) mobile vendor detection model training based on deep learning, which comprises (1) constructing a network model RefineDet; (2) training the network by using a cross entropy loss function; (3) training the trained network in the step (2) by using a focus loss function, and selecting a plurality of balance factors and a plurality of modulation factors of the focus loss function in a form of a controlled variable method so as to obtain a plurality of trained networks; (4) and (4) testing the network obtained in the step (3) by using a test set, and selecting the network corresponding to the highest average detection accuracy as a final mobile vendor detection model.
10. The system of claim 8, wherein the second flow detection and alarm process comprises: firstly, preprocessing a video acquired by a video acquisition module to obtain a video frame picture, then inputting the video frame picture into a trained mobile vendor detection model for detection, and analyzing whether a mobile vendor exists in the video frame picture.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911244134.7A CN110992645A (en) | 2019-12-06 | 2019-12-06 | Mobile vendor detection and alarm system in dynamic scene |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911244134.7A CN110992645A (en) | 2019-12-06 | 2019-12-06 | Mobile vendor detection and alarm system in dynamic scene |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110992645A true CN110992645A (en) | 2020-04-10 |
Family
ID=70091017
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911244134.7A Pending CN110992645A (en) | 2019-12-06 | 2019-12-06 | Mobile vendor detection and alarm system in dynamic scene |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110992645A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112508916A (en) * | 2020-12-12 | 2021-03-16 | 江西洪都航空工业集团有限责任公司 | Road well lid damage and loss detection method based on mobile video analysis |
| CN112766055A (en) * | 2020-12-30 | 2021-05-07 | 浙江大华技术股份有限公司 | Stall management method, stall management device, stall management system, storage medium and electronic device |
| CN114022772A (en) * | 2021-11-16 | 2022-02-08 | 华南农业大学 | Method, system, device and storage medium for predicting spatial distribution of mobile vendor |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107733845A (en) * | 2017-06-13 | 2018-02-23 | 银江股份有限公司 | A kind of panorama whole process law-enforcing recorder and acquisition management system |
| CN108647665A (en) * | 2018-05-18 | 2018-10-12 | 西安电子科技大学 | Vehicle real-time detection method of taking photo by plane based on deep learning |
| CN108921083A (en) * | 2018-06-28 | 2018-11-30 | 浙江工业大学 | Illegal flowing street pedlar recognition methods based on deep learning target detection |
| CN109686103A (en) * | 2018-12-29 | 2019-04-26 | 中南大学 | A kind of intelligent transportation based on image recognition is separated to stop monitoring management system and method |
-
2019
- 2019-12-06 CN CN201911244134.7A patent/CN110992645A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107733845A (en) * | 2017-06-13 | 2018-02-23 | 银江股份有限公司 | A kind of panorama whole process law-enforcing recorder and acquisition management system |
| CN108647665A (en) * | 2018-05-18 | 2018-10-12 | 西安电子科技大学 | Vehicle real-time detection method of taking photo by plane based on deep learning |
| CN108921083A (en) * | 2018-06-28 | 2018-11-30 | 浙江工业大学 | Illegal flowing street pedlar recognition methods based on deep learning target detection |
| CN109686103A (en) * | 2018-12-29 | 2019-04-26 | 中南大学 | A kind of intelligent transportation based on image recognition is separated to stop monitoring management system and method |
Non-Patent Citations (1)
| Title |
|---|
| 陈晋音 等: "基于深度学习模型的非法流动摊贩检测方法研究", 《小型微型计算机***》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112508916A (en) * | 2020-12-12 | 2021-03-16 | 江西洪都航空工业集团有限责任公司 | Road well lid damage and loss detection method based on mobile video analysis |
| CN112766055A (en) * | 2020-12-30 | 2021-05-07 | 浙江大华技术股份有限公司 | Stall management method, stall management device, stall management system, storage medium and electronic device |
| CN112766055B (en) * | 2020-12-30 | 2024-04-16 | 浙江大华技术股份有限公司 | Stall management method, device, system, storage medium and electronic device |
| CN114022772A (en) * | 2021-11-16 | 2022-02-08 | 华南农业大学 | Method, system, device and storage medium for predicting spatial distribution of mobile vendor |
| CN114022772B (en) * | 2021-11-16 | 2024-05-03 | 华南农业大学 | Method, system, device and storage medium for predicting space distribution of mobile vendor |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210389293A1 (en) | Methods and Systems for Water Area Pollution Intelligent Monitoring and Analysis | |
| CN110992645A (en) | Mobile vendor detection and alarm system in dynamic scene | |
| CN110189524B (en) | Active violation identification method and system and electronic equipment | |
| CN108046073A (en) | A kind of elevator network monitoring system for things | |
| CN110225143B (en) | Intelligent security data system | |
| CN106781519A (en) | One kind can positioning intelligent preventing road monitoring system | |
| CN109326125B (en) | Picture quality diagnosis system and method based on embedded system | |
| CN105654032A (en) | Statistical system of number of people on bus based on face detection and statistical method thereof | |
| Fawzi et al. | Embedded real-time video surveillance system based on multi-sensor and visual tracking | |
| CN112671104A (en) | Transformer substation multidimensional scene control platform facing complex scene | |
| CN110796580B (en) | Intelligent traffic system management method and related products | |
| CN111027397B (en) | Comprehensive feature target detection method, system, medium and equipment suitable for intelligent monitoring network | |
| CN112016380B (en) | Wild animal monitoring method and system | |
| CN112529466A (en) | Smart city management method and system based on big data | |
| CN116452925A (en) | Monitoring video identification system based on intelligent algorithm | |
| CN116682223A (en) | Fire inspection system and method based on Internet of things and AI under domestic environment | |
| CN113688385B (en) | Lightweight distributed intrusion detection method | |
| WO2022042045A1 (en) | Meter recognition apparatus, meter monitoring system, and monitoring method therefor | |
| CN104023205A (en) | Intelligent security monitoring system | |
| CN107545211A (en) | High in the clouds data handling system | |
| CN113963310A (en) | People flow detection method and device for bus station and electronic equipment | |
| CN113452961A (en) | Water surface monitoring alarm system, method and medium based on edge calculation | |
| CN113065416A (en) | Leakage monitoring device integrated with transformer substation video monitoring device, method and medium | |
| CN112217892A (en) | Smart city monitoring information transmission method and system based on block chain system | |
| CN112291527A (en) | Method and system for providing intelligent security information based on block chain system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200410 |