Hikvision, an IoT solution provider with video as its core competency, today announced a brand-new addition to its DeepinView camera line: the Dedicated Subseries. This unprecedented new addition loads a batch of AI-powered deep learning algorithms into each unit, boasting stunning performance and cost-effective pricing.
Over the last few years, artificial intelligence (AI) has been applied in many ways in security markets. As technology advances, AI chipset performance has improved to enable massive computing power using various algorithms and contributing to multi-intelligence functionality and higher accuracy. The new Dedicated DeepinView Cameras are an example of these advances, incorporating several AI-powered deep learning algorithms in one unit. What’s more, these algorithms can be switched essentially putting 5 or 6 unique cameras in one housing.
“Embedding switchable algorithms is a significant step for Hikvision to take in its AI product development. In a world of ever-changing technologies and functionalities, this approach creates great value for end users to try new technologies to ensure security, as well as to implement business intelligence and other applications,” says Frank Zhang, President of the International Product and Solution Center at Hikvision. “The benefits of our new offerings are numerous including reduced costs, improved efficiency, and speedy and effective incident response.”
Switchable algorithms
The Dedicated DeepinView cameras combine two product categories – the first is vehicle analysis where cameras combine automatic number plate recognition (ANPR) with vehicle attribute recognition. Attributes include the vehicle’s make, color, and direction of movement. Typical uses include installation at checkpoints of city streets and at entrances & exits of buildings or industrial parks.
Models in the second category boast six switchable deep learning algorithms in one camera housing, including facial recognition, face counting, hard hat detection, perimeter protection, queue management, and multiple-target-type detection (detecting multiple targets and multiple types of targets at once). Accordingly, users can simply enable an algorithm manually for dedicated use, then later switch the algorithm as needed.
Here is one example: hard hat detection. This algorithm can be used on construction sites to ensure safety and compliance. Specially-equipped DeepinView cameras can precisely distinguish a worker on the site wearing a hard hat from those without, and automatically deliver alerts when the hard hat violation is detected.
Another example: in a retail setting, a face-counting function can be enabled to precisely count customers entering and leaving the store. Repeat customers and store staff can be automatically excluded in the process, helping store managers count new customers with precision.
Flexibility among algorithms enables users to also switch among:
Perimeter protection – to monitor outdoor areas needing security and deliver accurate alarms upon intrusions.
Facial recognition – to grant authorized access to restricted areas in various organizations, such as school laboratories, archive rooms, and hospital pharmacies.
Queue management – to better understand customer wait times, optimize staff levels, and enhance customer experience.
HD clarity, day and night
Equipped with Hikvision’s DarkFighter and LightFighter technologies, these cameras capture vivid and color images in extremely low-light environments or in scenes with strong backlighting where color and brightness balance is extremely difficult. Smooth Streaming mode further ensures a high-quality live feed.
The Dedicated DeepinView Cameras are available in 2, 4, 8, and 12 MP resolutions for customers to choose from.
More practical and deployable features
Furthermore, metadata is supported to allow third-party platforms to receive data from Hikvision cameras for real-time video analysis or recorded into footage archives to enable rapid searching forensic evidence.
Finally, these camera models also offer Vibration Detection for outdoor use, which detects and notifies users of vandalism.
