Dahua DH-IPC-HDBW8242E-Z4FR: DeepSense

DeepSense series products adopt the most advanced AI technologies, including deep learning algorithms that primarily target people and vehicles, which provides higher flexibility and accuracy for end-users. This enables the Dahua DeepSense series to offer various advanced applications such as Face Recognition, ANPR, Metadata, People Counting, traffic data statistics, etc.

The complete lineup of Dahua DeepSense includes network (PTZ) cameras, network video recorders, servers, and platform management products. Beyond seeing the world, the power of AI allows devices to perceive the environment and understand the world in a better way.

System overview
Face recognition AI network camera is designed to focus on the recognition of frontal human faces, it adopts high performance smart chip with deep learning technology, which can provide high precision face image snapshot and comparison. This product supports real-time face detection, face attributes analysis, face images comparison and face image database management.

Functions

Face capture
Face capture is a software application that automatically captures faces from within a digital image or a video frame from a video source. Dahua cameras use advanced deep learning algorithms and are trained by a large number of face data sources, enabling the camera to locate faces quickly and accurately from the video source and capture facial images.

Face attributes
Through the use of deep learning algorithms to analyze facial images, Dahua cameras can extract six facial attributes from each frame, including gender, age, expression (happy, calm, surprised, sad, and angry), glasses, masks, and moustaches. Customers can acquire data of interest via face attributes statistics.

Metadata
Metadata is feature attribute information extracted from a target object which can be used for data retrieval. Dahua face recognition camera can extract six facial attributes and output the metadata for analysis.

Face recognition
Once facial features are extracted from captured faces, they are stored in a database where they can be easily searched and compared against other images. Once a successful match is found, the system outputs the result. The Dahua Face Recognition camera supports a built-in database that stores up to 10,000 facial images, helping the camera achieve real-time face capture and comparison.

Starlight Technology
Featuring Dahua’s Starlight Technology, this camera is ideal for applications with challenging lighting conditions. Its low-light performance delivers usable video with minimal ambient light. Even in extreme low-light conditions, Starlight Technology is capable of delivering colour images in near complete darkness.

Wide dynamic range (WDR)
Embeded with industry leading wide dynamic range (WDR) technology, vivid picutures are achieved even in the most intense contrast lighting conditions. For applications with both bright and low lighting conditions that change quickly, True WDR (120dB) optimises both the bright and dark areas of a scene at the same time to provide usable video.

Smart Codec (H.265+ & H.264+)
Smart Codec is the optimised implementation of standard video compression (H.265 & H.264) that uses a scene-adaptive encoding strategy, dynamic GOP, dynamic ROI, flexible multi-frame reference structure and intelligent noise reduction to deliver high-quality video without straining the network. Smart Codec technology reduces bit rate and storage requirements by up to 70% when compared to the standard video compression.