Background of the plan
In order to implement the work requirements of the Notice on Strengthening the Construction of Social Security Prevention and Control System issued by the General Office of the Communist Party of China Central Committee and the General Office of the State Council, as well as the Work Plan for Strengthening the Construction of Online Applications of Public Safety Video Surveillance (2015-2020) issued by the National Development and Reform Commission, the Comprehensive Management Office of the Central Committee, and the Ministry of Public Security, and the Several Opinions on Strengthening the Construction of Online Applications of Public Safety Video Surveillance (Development and Reform High tech [2015] No. 996), To achieve full coverage, full network sharing, all-time availability, full process controllability, and comprehensive application of public safety video surveillance in XX city, solve the shortcomings of existing video surveillance systems, further strengthen public security prevention and control, combat crime, serve urban management, and innovate social governance.
Plan Introduction
Cloud edge fusion architecture, long-term development of the system
The construction of information system with video and the Internet of Things as the core needs not only the support of cloud computing, but also the support of edge computing. Therefore, in order to ensure the long-term development of the system, the design concept of cloud edge integration should be followed.
Edge computing mainly injects AI capabilities into edge nodes to achieve multi-dimensional perception data collection and front-end intelligent processing, so that edge nodes have more accurate perception computing capabilities and more agile response capabilities.
Cloud computing, on the other hand, mainly provides the aggregation, storage, processing, and intelligent application capabilities of perception data and business data, as well as the ability to integrate multi-dimensional data and analyze big data applications. Ultimately, it can support multi-dimensional big data analysis needs such as prediction, warning, and situation analysis.
Integration of IoT data and comprehensive empowerment of applications
Traditional video construction applications are relatively single. With the advent of AI, there is an increasing amount of value data that can be extracted from videos. In addition, more and more IoT sensor data is being collected, leading to a surge in value data related to videos and the Internet of Things. Meanwhile, with the continuous deepening of business applications, the types and scale of business data are also increasing. How to utilize this data, incubate new value applications, and solve practical problems for users is a key issue that needs to be addressed.
To this end, it is necessary to build a data resource pool that integrates multiple perception and business data. Through the collection of multidimensional data, structuring of video image data, data cleaning, labeling, association, and other methods, various types of topic libraries, topic libraries, search libraries, relationship libraries, etc., can be constructed to make corresponding applications according to different needs and empower various industries.
Advantages of the plan
1. To promote the accuracy of dynamic evaluation through source governance, and to promote the scientificity of construction through evaluation, forming an effective closed loop of construction, governance, and evaluation.
2. Data supports cross level and cross network exchange, and flows to the places where the application needs it on demand, solving the problem of data silos.
Strengthen the classification and governance capabilities of IoT data and business data, improve data quality, and effectively integrate various data assets.
4. Deep integration of IoT data, achieving complementary advantages between IoT data and business data, and enhancing practical application value.
Based on the algorithm warehouse system, achieve unified management and scheduling of resources such as algorithms, computing power, data, and services.
Based on an AI algorithm training platform, we aim to create a new AI development model that is simple, user-friendly, and user-friendly. We aim to provide a one-stop algorithm customization service, effectively solving problems such as insufficient algorithm personalization, long development cycle, cumbersome deployment, high cost, and insufficient privatization.
7. Unified operation and maintenance, effectively overcoming difficulties such as complex and diverse equipment types, large number of equipment, scattered construction, and inconsistent protocols.
