Enabling a safer society and smarter living through artificial intelligence innovation
Dahua Technology, a leading solutions provider in the global video surveillance industry, had a strong presence at ISC West this year. A highlight of their participation in this year’s show was their keynote speech titled “AI Creates Value” during the afternoon of April 11, the first day of the exhibition, where the company introduced and showcased their innovations and progress in the field of artificial intelligence.
The program kicked off with Dahua President Ke Li delivering opening remarks to a sold-out audience packing Ballroom J of Sands Expo, the venue of ISC West. Li expressed gratitude to all guests as well as to Dahua’s global network of partners and customers for another year of success that the company achieved in 2017. Dahua, currently ranking No. 3 on a&s’ Security 50, achieved US$2.88 billion in total revenue last year, a year-on-year growth of 41.4 percent compared to 2016.
“Innovation has been the persistent force behind Dahua’s technology development,” said Li. “Dahua will continue to lead the application of AI in the security industry, becoming a world-leading video-centric smart IoT solution and service provider that enables a safer society and smarter living.”
Li’s remarks were followed by keynote speeches given by Adam Burns, GM of IoT Vision Solutions at Intel, and James Wang, Deputy GM of Dahua Overseas Business at Dahua.
In his keynote, Burns spoke to the trends in AI technology. “It’s been years since Dahua and Intel started their cooperation,” said Burns. “The evolution from video to vision makes surveillance more connected, intelligent and autonomous. Intel AI solutions, including Movidius, Intel FPGA and nervana, drive intelligence in the camera, at the edge and in the cloud, enabling various vision-based intelligent analytics and the expansion of vision to all IoT markets.”
In his keynote, Wang defined the challenges of big data, the value of cognitive analytics, the increasing business intelligence opportunities, the expanding possibilities for improved business operations, and the system architecture components that make all of this possible.
According to Wang, AI is the next “physical security industry transformation” following analog-based CCTV, digital signal, and IP-based, high-definition video. He further cited three key enablers of AI, namely big data generated in the public and private sectors; higher computing power, as demonstrated by chips launched by Intel, NVIDIA and others; as well as advanced deep learning algorithms as those developed by Google, Microsoft, Face++ and Sensetime. Together, AI addresses some of the biggest challenges facing end users, including the need to analyze huge amounts of data generated by countless cameras and other IP devices.
“There have been several noteworthy advancements in AI in intelligence, computing, ecosystem and sensing,” said James Wang. “Dahua AI-powered security technologies mitigate the data challenges in physical security solutions like low processing efficiency, isolated data and massive data.”
According to him, AI applications can be roughly divided into two categories: those based on persons and crowds as well as those based on vehicles/traffic flow. AI applications based on persons/crowds include face recognition, behavioral analysis and crowd density. AI applications based on vehicles/traffic flow include automatic numeric plate recognition, vehicle flow analysis and vehicle behavior analysis which detects and recognizes abnormalities or traffic law violations, for example when someone hasn’t buckled up, runs a red light or swerves between lanes.
The power of AI has made it suitable for security as well as non-security purposes, according to Wang. In terms of security, AI can be used for criminal prediction, where the end user can predict a potential crime according to behavior and characteristic analysis based on real-time and historic data; crowd analysis where the user can monitor human traffic flows and respond accordingly when the flow exceeds the set threshold value; and traffic flow analysis where the user can monitor and predict traffic flows to help traffic police control road traffic.
When applied to non-security, AI can help create value by enhancing the user’s business intelligence, Wang said. In retail, for example, customer flow analysis and customer demographics analysis can help the retailer better staff their stores and design better sales and marketing strategies. In logistics and warehousing, AI can help achieve automated warehousing by way of smart algorithm for goods recognition; as well as enable delivery route planning through real-time traffic management.
“Dahua AI security solutions, when applied to applications like retail, traffic management, logistics and warehousing, can boost business intelligence and performance, creating true value for physical security industry,” Wang said.
Wang also introduced the Dahua deepsense series, a suite of products that adopt state-of-the-art AI-deep learning technology, which has been released during 2018 H1. Dahua deepsense series, including network (PTZ) cameras, network video recorders and servers, supports various kinds of security solutions with an array of advanced capabilities including human characteristics analysis, face recognition, ANPR, metadata collection and analysis, people counting, image search, facial flow, traffic incident detection and traffic data statistics.
Once thought of as science fiction material, AI has become more and more realistic and practical, thanks to more data, better computing power and more advanced algorithms. Beyond security, the power of AI-deep learning allows devices to understand the world in a better way, offering higher flexibility and accuracy for end users and more business opportunities for Dahua customers and partners. Indeed, Dahua’s exhibition at ISC West were only a beginning; we can expect more advanced and cutting-edge AI products and solutions to come in the near future.