In late autumn 2017, the world's largest security Expo opened in Shenzhen. What all walks of life did not expect was that artificial intelligence ushered in a "full explosion" in this vertical conference with the theme of "security". One participant said that the AI ​​craze in the whole museum was still fresh in the memory.
Why is artificial intelligence and security so "close"? The reason is very simple: all walks of life generally believe that the security industry will become the scene of artificial intelligence's largest landing, and the prosperity of China's security industry also provides broad opportunities for artificial intelligence. Various artificial intelligence companies such as Nvidia, Bitmain, SenseTime, and Yitu have also entered this battlefield.
Facing the competition of established artificial intelligence companies, how do Chinese artificial intelligence companies enter? Different companies have given various answers. Some companies focus on competing software algorithms, while others start with the underlying chip and expand from upstream to downstream in an attempt to open up the industry chain. What are the future prospects? We will wait and see.
The "intimate relationship" between artificial intelligence and security
The industry research report pointed out: China's security industry began to emerge in coastal cities in the late 1980s. With the economic growth, it began to grow rapidly during the "Tenth Five-Year Plan" period, with a compound annual growth rate approaching 30%. During the "Eleventh Five-Year Plan" period, with the launch of major projects such as "Safe City", "3111 Pilot Project", and "Beijing Olympics", the market scale has been further expanded, and the entire industry has also achieved rapid growth.
Especially since the 2008 Olympics, smart cities have been highly valued by the Chinese government. During the "13th Five-Year Plan" period, China has regarded smart cities as a key direction for the development of new urbanization; in 2015, the nine central ministries and commissions jointly issued the "Opinions on Strengthening the Construction of Public Security Video Surveillance Network Application Work", and its overall goal is to achieve "Full domain coverage, full network sharing, full time availability, and full controllable" my country's public security video surveillance network application.
In October 2017, the Secretary Office of the Central Comprehensive Management Office, the General Office of the National Development and Reform Commission, the General Office of the Ministry of Public Security, and the Office of the National Standards Commission issued the "Public Security Video Image Information Networking and Sharing Application Standard System (2017 Edition)", requiring public safety The network application of video surveillance construction is the implementation of the Xueliang project.
The “Xueliang Project†is a “mass public security system†with comprehensive management centers at the county, township, and village levels as the command platform, supported by comprehensive management informatization, based on grid management, and focused on the application of public security video surveillance networking. Prevention and Control Engineering". Whether it is a smart city or a bright project, the core of it is "video surveillance". In the face of the massive image and video information generated by the security system, how to effectively recognize each face and license plate is whether a safe city can be truly realized key.
In fact, video intelligent analysis is one of the areas where artificial intelligence technology is best. In recent years, driven by the in-depth construction of smart cities and Xueliang projects, the trend of commercial applications of artificial intelligence has become more and more obvious. In terms of AI+ security landing, video structuring, that is, the recognition and extraction of video data is the core technology is particularly critical. Video structuring technology is a combination of cutting-edge artificial intelligence technologies such as machine vision, image processing, pattern recognition, and deep learning, and is the cornerstone of AI+ security.
For example, Tang Weiwei, director of product strategy at Bitmain, told reporters that in order to solve the case, hundreds of police officers once bought out all the eye drops in the city in order to check video recordings. However, after using artificial intelligence video analysis technology, people, cars, and objects in the video screen can be identified, and through feature analysis, the police can quickly help the police catch the suspect. This is also the most direct transformation of artificial intelligence technology to urban safety.
Security + AI starts from the core
In the video surveillance industry chain, the security chip is at the top of the pyramid, and its technical indicators determine the overall technical indicators of the entire security surveillance system. With the rapid development of artificial intelligence, artificial intelligence chips will become the focus of the future development of semiconductor companies. In terms of international manufacturers, Intel, Qualcomm, IBM, Nvidia, Micron, etc. have begun to deploy and develop artificial intelligence chips. Domestic chip companies have also taken steps in the development and application of artificial intelligence chips, including Bitmain and Cambrian. , Horizon, etc.
Video surveillance is mainly divided into three parts: front-end, transmission, and back-end/cloud. In the front-end "AI+camera" link, manufacturers can directly solidify AI companies’ facial recognition and other algorithms into IP modules and embed them in video surveillance SoC chips; they can also use more general vision processors in the cameras, such as Intel Movidius VPU, NVIDIA’s Jetson TX series, etc. "AI + back-end" is to send the video information collected by the camera to the back-end central control system or the cloud, and then use the powerful computing power of the server to analyze and process it.
In the field of AI chips, there is still a big gap between China and the United States, and China does not yet have a strong voice. However, in the past two years, many AI companies have emerged, attracting the world's top AI talents and chip design talents. AI chips will be a very important direction and opportunity for China's rise in the chip field. With the help of the rise in the AI ​​chip market, It can quickly get closer and even surpass the United States in the field of security chips.
For AI chips, the country has also placed a very important position. In December 2017, the Ministry of Industry and Information Technology of China issued the "Three-year Action Plan for Promoting the Development of the New Generation of Artificial Intelligence Industry (2018-2020)" (hereinafter referred to as the "Plan"), which focused on machine learning training applications. Develop high-performance, high-scalability, and low-power cloud neural network chips, develop low-power, high-performance terminal neural network chips suitable for machine learning calculations for terminal applications, and develop compilers that sit on neural network chips. Industrialization support tools such as driver software and development environment; realize large-scale commercial use of neural network chips in key areas such as smart terminals, autonomous driving, smart security, and smart home.
Case study: AI chip company Bitmain
At the end of 2017, Bitmain officially released its Sophon TPU chip BM1680, which is a special customized chip for tensor calculation acceleration processing for deep learning applications, suitable for inference and prediction of deep neural networks such as CNN, RNN, DNN, etc. (Inference) And training. BM1680 single chip can provide 2TFlops single-precision accelerated computing capability.
At the same time, Bitmain released the intelligent video analysis server Sophon SS1, which is a new deep learning server. This deep learning server based on Bitmain's newly developed SC1/SC1+ deep learning accelerator card and deep understanding of image recognition algorithms is designed to provide powerful deep learning for various application scenarios such as video surveillance and Internet image processing. Acceleration capability, with standard functions such as face detection, face recognition, human detection, and machine-non-human detection classification.
Tang Weiwei, Director of Product Strategy of Bitmain, told reporters that in the field of AI+security, Bitmain’s advantage lies in its ability to provide integrated solutions from the underlying chip to the upper-layer application technology, which can provide the security industry with the ultimate cost-effective and performance-to-power ratio. Program.
Tang Weiwei believes that the security industry is a traditional industry with a long history, with a long industrial chain and a complex value chain. In the past ten years of industry development, digitization, networking, and high-definition technology are the technology trends, and many worlds have emerged in China. The leading sex industry has become the trend of the times. The in-depth integration of artificial intelligence and the security industry in this round requires not only the application of the upper layer, but also the technological breakthrough on the underlying chip.
The utility model discloses an Electronic Cigarette with atomizer oil core separation structure, which comprises a atomizer assembly and a battery assembly; The utility model has the advantages of reasonable structure design and high practicability. During operation, because the oil storage tank of the atomizer is completely sealed and the oil guide material is completely isolated from the oil guide material, the oil guide material will never contact with the oil guide material in the storage process
Advantages:
Oil core separation, Mesh removable heating core, Ultra quiet design,100% oil leakage free, Excellent taste, Cost far less than the industry price.
A variety of colors and finishing are available, can be adjusted according to your market needs.
Oil Coil Separation Pod Patent,Oil Coil Separetion Vape Pod Oem,Innovated Vape Products Oem,Oil Coil Separation Vape Products Oem
Shenzhen MASON VAP Technology Co., Ltd. , https://www.disposablevapepenfactory.com