The application of edge computing technology in the security industry should be earlier than cloud computing. In the security industry, especially in the field of video surveillance, the biggest feature is to handle massive amounts of video data. The transmission and storage of video data has always affected the development of the security industry. The development of the security industry is basically accompanied by the development of edge computing technology. According to the development of the security industry, the application development of edge technology is also roughly divided into the following three stages: 1. Early video coding and encryption Video surveillance from the analog era to the digital age transition, the face of a big problem is the huge amount of video data. For a D1 video with an SD resolution of 704 X 576, if the data is not compressed, it will take about 250 Mbps of data bandwidth to transmit a full 25 frames of video, which was impossible at that time. Even now, network bandwidth has been qualitatively increased, but the demand for high-definition video is such that if data is not compressed, video transmission cannot be performed. Therefore, to transmit digital video, data processing must be performed at the source of the data to compress the video data. This kind of video compression technology is actually an edge computing technology. And with the development of the times, people's demand for high-definition video development is much higher than the development of network bandwidth, so the edge compression technology of video compression has also been progressing. From the early MJPEG and MPEG4 algorithms based on DSP to the H264, H265 and SVAC algorithms based on the special encoding chip algorithm, the video compression rate has been improved. In addition, some requirements require the encrypted transmission of compressed video. Therefore, some digital surveillance devices have the function of encrypting video on the device side. This technology is actually an edge computing technology. Therefore, the application of the early edge computing technology in the security industry is mainly to consider the two characteristics of relief flow pressure and higher security. 2. Mid-term industry-specific analysis algorithms With the development of the security industry, user needs are no longer confined to simple video preview, storage, and playback functions. Security products are no longer just simple generic monitoring products, but have become professional video equipment in various fields. The most well-known among them are the electric police and bayonet devices in the transportation industry because they need the ability to identify license plate numbers, detect red lights, and not wear safety belts. These applications often require extremely high real-time data processing. Sexuality, but also in the data center is only interested in illegal acts and license plate numbers, so the device needs to immediately perform data processing after the data is collected, capture the violation, and then upload the illegal photos and license plate information to the cloud. This is a typical edge computing technology. There are many similar edge computing application cases. For example, Tianweiweiye’s alert series can be used to trigger the speaker when someone enters the alert area, telling people not to enter the prohibited area and reminding pedestrians to pay attention to safety. In addition, the latest alerting products even support the detection of helmets. When workers in construction sites are detected forgetting to bring their helmets, they will be actively reminded through the loudspeakers. This kind of technology is also a typical edge computing technology. If cloud computing technology is used, data is transmitted to the data center for processing and the prompt information is returned. It is difficult to implement real-time alerting. Therefore, the application of the medium-term edge computing technology in the security industry mainly considers the characteristics of low latency and higher efficiency. 3. Current artificial intelligence algorithms such as face recognition based on deep learning Artificial intelligence began in the middle of the 1950s, and it took decades to develop. In the late 1980s, with the rise of research on artificial neural networks, artificial intelligence entered a new phase. In particular, in recent years, deep learning has achieved breakthroughs in the optimization of artificial neural networks, making it possible for machine aids and expanding the field of application of artificial intelligence. Since the artificial intelligence algorithm consumes a great deal of system performance, if only the CPU is used, the promotion at the edge cannot be realized. So previous artificial intelligence algorithms are more focused on the cloud. However, with the gradual maturity of technology, major chip manufacturers have begun to introduce artificial intelligence algorithm chips, making it possible to implement artificial intelligence at the edge. Major security vendors have also introduced artificial intelligence devices based on edge computing technology. Among them, the world's face capture products are typical examples. Based on the edge computing technology, the face capture device makes it possible to parse the face data at the first time when the pedestrian passes through, and sends the face data to the data center for matching processing. Compared to the simple cloud computing solution, Without uploading all video data, data traffic can be greatly reduced, real-time performance can be improved, and data integrity can be guaranteed even on a 4G network. Gear Pump,Gear Oil Pump,Oil Gear Pump,Hydraulic Gear Pump Dongguan Jinzhuang Hydraulic Technology Co., Ltd. , https://www.jinzhuanghydraulic.com