Six investigation modes based on three video analysis models
I. Introduction
At this stage, the video investigation work mode is mainly based on manual review, supplemented by simple and intelligent video analysis. Professional duty personnel are still required to conduct 24-hour video patrols or browse and query a large number of videos based on case event information. The work mode based on manual intervention has gradually become the bottleneck of video applications. And with the explosive growth of video resources, the low application efficiency of massive video resources began to restrict the development of video business applications. At this time, the construction of security big data applications was put on the agenda. Just this year, the security industry leader Hikvision Also put forward the idea of ​​SDT security big data.
At the two sessions in 2015, Premier Li Keqiang clearly put forward the action plan for big data construction in the government work report, and big data has become a basic strategic resource for the country. The application of big data is an inevitable requirement for realizing the construction of the intelligence-led police model and an inevitable development trend for the construction of public security informatization. In the construction of security big data applications, advanced technologies such as video structured description technology and intelligent image analysis technology are used to extract structured data from people, vehicles, and moving objects in massive videos, and mass unstructured videos and semi-structured data The conversion of pictures into valuable structured information and the realization of "video analysis" will be an important foundation for public security organs to realize the construction of large-scale security applications. In this article, the author will talk about the new changes brought about by "video analysis" for video investigation based on the current market environment and his own work experience in the security industry.
2. What is "video analysis"?
What is "video analysis"? Video parsing is a technology that organizes into textual information that can be recognized, understood, and retrieved by computers and people by using analysis methods such as target detection, feature extraction, object recognition, and deep learning. The video analysis methods can be divided into three types: the first is the traditional "pure manual" analysis mode. The policeman manually marks the suspect target while reviewing the video, which is convenient for later retrieval and calling. This method has low efficiency. , Can no longer meet the needs of massive video analysis at the current stage; the second type is "mainly artificial, supplemented by intelligence", due to the advent of front-end intelligence and back-end intelligent resolution equipment, the original video can be preliminarily analyzed to a certain extent. Free up manpower and improve analysis efficiency; the third is the “intelligence-based, manual-assisted†analysis mode. With the development of intelligent analysis equipment, video images are structured through video analysis and big data, cloud computing and other technologies. Services such as structured analysis of resources, fast retrieval, and big data comparison and collision have gradually been replaced by manpower. However, since the current smart devices do not have strict guidelines on the specific analysis items of the suspect target, and for more complex cases, the content of the analysis cannot meet the needs of the case for clues, so at this stage, manual analysis is still required. "Mainly, supplemented by manpower" analysis method is widely used at this stage.
This article combines video reconnaissance with the third analysis method widely used at this stage, and explores a system construction model of "offline analysis system + centralized analysis system + manual analysis assistance". Among them, the offline analysis system can be passed directly at the crime scene The offline forensics equipment analyzes the video and pictures involved in the case, and the resulting clue information can be directly studied and judged by the user on site, which can improve the cracking efficiency of the case to a certain extent. The centralized analysis system receives the original video and pictures transmitted from the front-end camera, and can perform real-time and historical video analysis of people, vehicles, and moving objects, and performs secondary information recognition on the passing pictures of the bayonet. Finally, the standard external service interface is The public security business provides structured basic data, which is mainly used in the analysis of massive video in major cases. Finally, some large cases, major cases and special cases also need the assistance of manual analysis. When investigators view the video, they directly structure the suspected target to form clue information, and assist the offline analysis system and the centralized analysis system to complete the clue screening. And integration.
Third, the new model of video investigation
"Video parsing" provides powerful data support for video reconnaissance and lays a foundation for the exploration of new modes of video reconnaissance. Based on the "offline analysis system + centralized analysis system + manual analysis assistance", this paper proposes a video detection mode that is different from the traditional manual intervention and supplemented by intelligence. According to the type, size and severity of the case, etc. Differentiate and explore new video reconnaissance modes including: single combat mode, cluster combat mode, mobile reconnaissance mode, case-focused mode, video reconnaissance studio mode, and synthetic combat six modes. Through different video reconnaissance modes, it can meet The video investigation work of different scenes and different needs provides certain ideas for the public security to solve the case.
1. Individual combat mode
When a general case occurs, the investigator can bring the current video survey equipment and offline forensics equipment to the crime scene to understand the situation, collect the video of the case and the surrounding area, and conduct target screening through the "offline analysis system". The structured data formed can be used for on-site research and judgment by using offline forensics equipment, quickly locate the suspected target, and complete the video investigation work. The single combat mode of operation can meet the conventional business needs of general case detection.
2. Cluster combat mode
When an emergency case occurs, multiple investigators can be assigned to carry the multiple offline forensics equipment to the crime scene to understand the situation and form multiple LAN forensics equipment into a local area network to form a video investigation team, multi-person collaboration, and resource sharing , To jointly complete the video collection of the crime scene and surrounding areas, and to conduct target screening through the "offline analysis system" to form structured data, provide data support for on-site judgment and improve the efficiency of the police. The cluster combat work mode is mainly suitable for the case detection needs where the cases are more complicated and time is more precious.
3. Mobile detection mode
When a case with complex scenes occurs, such as when it is not convenient for the investigator to arrive at the scene, or when a secret investigation is required, the investigator can carry mobile investigation equipment for temporary deployment control, and the video in question is transmitted back to the "centralized analysis in real time through the operator VPN private network" "System" for storage and analysis, and remote research and judgment through the actual combat application platform to complete the video reconnaissance work. Mobile detection equipment is an important supplement to conventional fixed video surveillance.
4. Concentration mode
When a major case or a major case occurs, the investigating police can rely on the social security monitoring network sharing platform, video on-site survey and acquisition equipment, and video mobile emergency deployment control equipment to download and upload online videos and off-line uploads of the videos involved in the crime scene. "Centralized analysis system" and "manual analysis assistance" method to conduct preliminary screening of the video involved in the case, forming a case clue, using the case comprehensive combat platform to integrate the clues of the analyzed structured data with the original video, and linking the elements of "people, vehicles and objects" , Time and space trajectory research and judgment, case concatenation, comprehensive collision and other methods of multi-person collaboration, comprehensive research and judgment to determine the next direction of investigation, or generate a report after the case is closed. The case-focused mode relies on the video analysis center to improve the efficiency of clue conversion and improve the rate of solving cases. Finally, the case-focused model should also focus on on-site video case information management, query, sharing, and storage. It provides strong data support and guarantee for the follow-up case, the comprehensive utilization of the video information of the case and the case-by-case.
5. Video Investigation Studio Mode
The so-called video investigation studio is to recruit experienced investigators and imaging technicians to form a professional investigation team according to the requirements of "detection, evidence collection, image processing, analysis and judgment", and according to the video detection business process and application system Deployment requirements, configure the corresponding video reconnaissance workplace and workspace for the reconnaissance team, through the planning of the workspace to enable the reconnaissance personnel to work independently or in joint operations. The overall work flow of the video investigation studio is similar to the centralized mode of the case. From video collection à video review à video analysis à video research and judgment, the video analysis here is also similar to the centralized mode of the case, also through the "centralized analysis system" plus "manual analysis assistance The main difference is that the analysis of the video investigation studio model emphasizes the "special zone special person" and "special person only", establish a professional team of video investigation, and enhance the special skills of video investigation personnel. The layout of the video investigation studio is as follows:
6. Synthetic combat mode
Establish a "synthetic operations" modern police mechanism for criminal investigation, technical investigation, economic investigation, and net investigation to synchronize cases. When a criminal case occurs, the various police types quickly gather and organize the information involved in the case. Simultaneously file cases, and carry out video analysis and comprehensive research and judgment around the clues. This synthetic combat mode can maximize the integration of professional technical resources of various police types, and rely on video surveillance network sharing system, social resource management platform, bayonet subsystem, and actual combat application platform to gather these professional technical resources to the greatest extent to detect and solve cases. To achieve coordinated sharing of resources and precise strikes, thereby improving the comprehensive strike efficiency and comprehensively enhancing the core fighting power of public security organs.
4. Conclusion
The SDT era has arrived, and it has become an inevitable trend for security big data to be applied to video detection and detection. It is necessary to build an analytical system model of "offline analysis system + centralized analysis system + manual analysis assistance", and "process and produce" structured data. People, vehicles, and objects are the focus of the target's video image information clues, suspense database, case events, item library and other special business databases, providing video image information, feature information and other data support for public security business big data applications, so that video Under the background of big data, investigation will fully play the role of demarcating the direction of investigation, narrowing the scope of investigation, locking suspected objects, and even identifying crimes.
It is the thinest phone case made from bullet-proof materials. It is the world's first modular magnetic wallet made from carbon fiber.
It is all about your carrying experience. It is a magic world of seamless carrying experience. And that makes you being different from anyone else.
Exclusive Material. Slim Fit and Sensuous With "Body Armor Grade" material, made of 100% Aramid Fiber. Now, your phone is protected with the same material used in body armor, spacecraft, jet engines, and Formula 1 Race cars. Yet, it retains the thinness of your iPhone the day you purchased it -- before it was covered up with an thick bulky case.
Wireless charging friendly. Bring you the best carrying experience and wireless charging experience by using with PITAKA magnetic car mount(Magmount Series), the unique transit free design with metal plates inside makes you can use PITAKA car mount freely without sticking adhasive metal plate on your device. Yet,no interference to your wireless charging,WIFI,GPS,Apple Pay or your signals.
A Soft 3D-Grip touch. Texture you need to feel to believe. It almost clings to your hand without feeling sticky. A protective three-layer coating and manual polishing is applied to each phone case.
Enjoy your "almost naked" iPhone with a case that is thinner than a dime. Thickness 0.03in. Min. Weight 0.49oz. So light that you even will not feel the weight increase to your phone. Well protect phones from most of the daily drops.
iPhone Case,Waterproof iPhone Case,iPhone 7 Case,iPhone 8 Case
Shenzhen Aokal Technology Co., Ltd. , https://www.aokals.com