In today’s digital age, Artificial intelligence (AI) has begun to revolutionize how we live, work and play. AI is offering businesses the prospect of increased productivity and innovation, while at the same time also enabling society to help solve some of its most persistent challenges -- like disease, natural disasters, and security.
 |
| Ray Wu, General Manager of Hikvision |
“As AI becomes more common and more powerful, deep learning will soon become the foundational technology for the security industry. AI will strengthen critical security efforts in every sphere, with new machine learning techniques giving CCTV cameras the ability to spot troubling behavior without human supervision,” said Ray Wu, General Manager of Hikvision.
Deep learning is the next level of AI development. It is beyond machine learning where supervised classification of features and patterns are set into algorithms. Rather, deep learning incorporates unsupervised or “self-learning” principles. “Hikvision is building deep learning into its own analytics algorithms. The result of multi-layer learning and extensive data collection is enhanced accuracy and improved analytics. And this applies to vehicle and human recognition,“ explained Mr. Wu.
Based in Hangzhou, China, Hikvision is a pioneer in intelligent video surveillance and has created diversified products and solutions that utilize AI to address today’s modern challenges. In Hikvision’s AI tour, the company introduced several product models equipped with deep learning algorithms:
- EXIR Flexible PanoVu Network Camera - This camera features 1/2.7" progressive scan CMOS sensors and four lenses with each lens supporting 1080p resolution at 30fps. A compression of up to H.265+ saves bandwidth and storage space. The EXIR is capable of 6 behavior analyses as well as 4 exception detections.
- PanoVu PT Series 2MP + 2MP Target Capture Camera - This camera is able to capture distant images as well as close-up images with the integration design of distant view camera and tracking PTZ camera. This series can be widely used in situations where wide monitoring range and high definition are required, such as the entrance or exit of an airport, park, or station.
- 2MP 25× IR Network Speed Dome - The speed dome is a DarkfighterX series camera that adopts the patented dual-sensor design to process color and brightness independently, and outputs one full-color image. It is equipped with deep-learning-based target classification algorithms for auto-tracking and perimeter protection.
- Facial Recognition Server - Powered by deep learning technology, the server incorporates algorithms for identity confirmation and support for up to 1 million blacklisted targets. It is compatible with third-party cameras and meets varied scenario requirements, including shopping malls and public squares.
- LPR Camera - Powered by Hikvision’s deep learning AI algorithm, this camera can identify more than 6 types of Philippine license plate numbers, including both private and public utility vehicles, at 90% accuracy. Hikvision is dedicated to optimizing its LPR algorithm and expects to up its accuracy to 95% by 2019.
Hikvision has introduced a series of products that can accurately detect, recognize, and analyze human, vehicle, and object features and behavior, and can be widely used in indoor and outdoor scenarios. “With AI, CCTV cameras now have the digital brains to match their eyes, letting them analyze live video with no humans necessary. AI is a game-changer, and we’re seeing firsthand how that technology can be used to guard against theft and improve security,” said Evelyn Ativo, Product Group Head of Iontech Inc., one of the key players in the Information Technology distribution industry in the Philippines.
