Dahua WizSense is a series of AI products and complete end-to-end solution based on deep-learning. focuses on human and vehicle, benefiting everyone with intelligent functions, simple configuration and inclusive products.
System Overview
The product is equipped with neural network processor; it supports WizSense intelligent functions to effectively improve the detection accuracy, supports people, motor vehicle and non-motor vehicle image capture, and supports perimeter protection.
Functions
Video Metadata
Supports detection of motor vehicles, non-motor vehicles, human faces and human bodies. Supports tracking, capturing pictures, and uploading high-quality snapshots.
Perimeter Protection
Automatically filters out false alarms caused by animals, rustling leaves, bright lights, etc. Enables system to act secondary recognition for the targets to improve alarm accuracy.
Face Detection
Supports 6 attributes and 8 expressions, face enhancement, and attributes extraction.
Starlight Technology
For challenging low-light applications, Dahua's Starlight Ultra-low Light Technology offers best-in-class light sensitivity, capturing color details in low light down to 0.005 Lux. The camera uses a set of optical features to balance light throughout the scene, resulting in clear images in dark environments.
PFA Technology
PFA technology has innovatively introduced new methods of judgment to ensure the accuracy and predictability of the direction of subject distance adjustment. The result is a set of advanced focusing algorithms. PFA ensures clarity of the image throughout the process of zooming and shortens focus time. The realization of PFA technology substantially improves user experience and increases product value.
Auto-tracking
This feature controls the pan/tilt/zoom actions of the camera to automatically track an object in motion and to keep it in the scene. The tracking action can be triggered manually or automatically by defined rules. Once a rule is triggered, the camera can zoom in and track the defined target automatically.
