Deepinmind Traffic Incident Detection Server
SKU: FX-iDS-TP40-16B.
Categories: , .

Deepinmind Traffic Incident Detection Server

  • Incorporates up to 16 channels of 3 MP IP cameras for traffic incident detection and traffic data collection.
  • Detects congestion, lane line driving, parking violations, wrong-way driving, falling objects, illegal lane changes, roadblocks, driving or occupying the emergency lane, and pedestrians.
  • Collects data on a wide range of traffic situations/conditions such as lane flow, space headway, time headway, space occupancy, and more…
Professional and Reliable:

  • Professional embedded hardware design, based on deep-learning face analysis algorithm
  • Support face recognition modelling, feature extract and other intelligent analysis
  • Support clustered deployment, to meet the demand of load balancing, dynamic expansion and fail-over
High Performance:

  • Up to 80 face pictures per second for analysis process on one device
  • Support most 300 thousand blacklist library
  • Up to 32 pairs of pictures for 1V1 comparison
Intelligent Functions:

  • Search dace by face through modelling of face pictures
  • Support feature recognition by gender, age, whether wear glasses and etc.
  • Support blacklist alarm
  • Support 1V1 comparison
Specifications
Model
DEEPINMIND SERIES PURE ANALYSIS SERVER
Speed of Facial modeling
and analysis
80 pics/sec
Access Face Recognition
Camera
16-ch
Active/Stand by ModeSupport (Overlay IP is not supported)
Quantity of clustered device1~30(Without external VMS)
30~200(With external VMS)
ResolutionUp-to 16MP, at least 102*126
FormatJPG, JPEG e
SizeUp-to 8MB
Face PixelsPupillary distance:40-300 pixels (60 pixels is recommended)
Deflection AngleUp and down deflection maximum 25 degrees, recommend 15 degrees,
left and right deflection maximum 45 degrees, recommend less than 25
PropertiesGender, age, whether wear glasses
CapacityUp to 300K targets, Up to 16 libraries
Comparison SpeedWithin 3 seconds (300K targets in blacklist)
Response Time1v1 comparison takes up to 1 second with full system load
Pure 1V1 Comparison
Performance (Only jpg)
32 pairs/sec
Retrieval CapabilityWithin 5s (1 Million Pictures )
Picture Storage Performance1 Million face pictures
CPU1*Intel E3
Memory64GB
HDDsEmbedded 1*120 / 150GB SSD
GPU Memory64GB
USB4*USB3.0 interface
2*USB2.0 interface
VGA1
Network4*1000M adaptive interface
PowerHot swappable 1+1 redundant power module
Rated Power400W
Chassis9 in-ch rack-mounted 1U chassis;
Dimension: 439mm(Width)× 591mm(Depth)× 44mm(Height)