Global

Global - English

Asia

Europe

Oceania

ANZ - English

Latin America

North America

Middle East and Africa

DS-IX2003-A3U/X

Intelligent Fusion Server

Integrates analysis, storage and application as one, and applicable to various scenarios
Concurrent analysis of Max. 90 channels of 1080P network cameras for a single device
Face picture analysis and processing of Max. 510 pieces per second for a single device
Provides Max. 50 million pieces of pictures and data storage for a single device
Performs various functions, including list alarm, personnel archive, personnel trace analysis, search by human body picture, and vehicle arming, etc. via web application interface
Equipped with professional GPU accelerator and integrates deep learning-based intelligent algorithms
Supports cluster deployment to deal with balanced load, dynamic adding and failover
Supports adding Max. 5 devices under cluster mode.

View Less

Available models:

DS-IX2003-A3U/X

Product Description
Device VersionV1.4.0
Device TypeAll-in-one

Face Analysis
Face AttributeSmile, as well as behaviors of wearing a mask, or glasses
Analysis TypeSupports analysis, comparison and classification of faces in pictures, real-time videos or recordings.
Face Picture Analysis and Comparison510 pieces/second Format: JPEG, BMP, TIF, PNG
Face Video Analysis and Comparison90 channels (2 MP) 36 channels (2 MP to 5 MP) 18 channels (5 MP to 8 MP)
Concurrent Performance-1 V 1 Comparison30 pairs per second
Max. Number of List Libraries128
Max. Number of Arming Lists3 million
Max. Number of Static Lists10 million
Imported Video AnalysisSingle video size ≤ 10 GB, and total video size ≤ 100 GB Max. 40 times accelerated analysis speed Format: MP4, AVI, TS, PS, MBF, ASF, SVAC
Facial Video FormatResolution: 720 P to 8 MP Real-time video stream: H.264, Hikvision H.265 Local recording: H.264, Hikvision H.265
Video Frame Rate Requirement12 to 30 frames per second (25 frames per second recommended)
Face Picture FormatJPEG, BMP, TIF, PNG
Face Picture RequirementFace analysis target: Resolution: 48 × 48 to 4096 × 4096 (16 MP) Size: ≤ 8 MB Pupil distance: 20 to 300 pixels recommended Face angle deviation: Face analysis: ≤ 35° (vertical); ≤55° (horizontal) Face picture of list library: ≤ 25° (vertical); ≤30° (horizontal)

Video Structuralization
Analysis TypeSupports analysis and modelling for human bodies or vehicles in pictures, real-time videos or recordings.
Human Body AttributeHair style, clothes type, clothes color, as well as the following behaviors: wearing a hat, a mask, or glasses; carrying a bag or luggage; driving a non-motor vehicle (the number of passengers, vehicle type, vehicle size, driving direction, driving speed)
Vehicle AttributeVehicle color, vehicle type, vehicle main brand, license plate number
Analysis of Human Body and Vehicle Videos90 channels (2 MP) 48 channels (2 MP to 5 MP) 24 channels (5 MP to 8 MP)
Analysis of Human Body and Vehicle Pictures180 pieces per second
Imported Video AnalysisSingle video size ≤ 10 GB, and total video size ≤ 100 GB Max. 40 times accelerated analysis speed Format: MP4, AVI, TS, PS, MBF, ASF, SVAC
Video FormatResolution: 720 P to 8 MP Real-time video stream: H.264, Hikvision H.265
Video Frame Rate Requirement12 to 30 frames per second (25 frames per second recommended)
Picture FormatJPEG, BMP, TIF, PNG
Human Body Picture RequirementResolution: 256 × 256 to 4096 × 2160 (8 MP) Human body width: 70 to 300 pixels recommended Completeness: the blocking area should be less than 30 percent of the target Angle deviation: the depression angle should be no greater than 45°
Vehicle Picture RequirementResolution: 256 × 256 to 4096 × 2160 (8 MP) Vehicle width: 100 to 300 pixels recommended License plate width: 80 to 120 pixels recommended Angle deviation: less than 30°(horizontal) and 25°(vertical), modelling for both the front and rear

Data Storage
Picture and Structured Data50 million
Search by Face Picture-Response Time2 s
Search by Human Body Picture-Response Time2 s
Search by Vehicle Picture-Response Time2 s

Web Application
AlarmBlocklist, allowlist, and license plate
1V1 ComparisonSupports rapid and accurate comparison of 2 face pictures
Personnel ArchiveSupport
Smart SearchList library search, capture library search, alarming search
Device Operation and MaintenanceSupport

Product Performance
Authorization TypeSoft Encryption
Video Accelerating Capacity40

Hardware Specifications
CPUIntel Xeon Gold 6230 × 2
Memory256 GB
Network InterfaceGigabit Ethernet self-adaptive interface × 4
Max. Power Consumption750 W
Rated Power Consumption1600 W
HDD Specification 240 GB M.2 × 1 + 480 GB SSD × 3 + 8 TB SATA × 4
GPU CardTesla T4 × 3
PSU Hot-plugged efficient 1+1 redundant power module
Device Dimensions715.5 mm × 482 mm × 86.8 mm (28.2 inch × 19.0 inch × 3.4 inch)
Package Dimensions965 mm × 670 mm × 320 mm (38.0 inch × 29.4 inch × 12.6 inch)

Physical and Environmental Parameters
Storage and Transportation Temperature -40 °C to 70 °C (-40 °F to 158 °F)
Working Temperature5 °C to 35 °C (41 °F to 95 °F)
Working Humidity8% to 90% (without condensation)

Specification

Product Description

Device Version

V1.4.0

Device Type

All-in-one

Face Analysis

Face Attribute

Smile, as well as behaviors of wearing a mask, or glasses

Analysis Type

Supports analysis, comparison and classification of faces in pictures, real-time videos or recordings.

Face Picture Analysis and Comparison

510 pieces/second Format: JPEG, BMP, TIF, PNG

Face Video Analysis and Comparison

90 channels (2 MP) 36 channels (2 MP to 5 MP) 18 channels (5 MP to 8 MP)

Concurrent Performance-1 V 1 Comparison

30 pairs per second

Max. Number of List Libraries

128

Max. Number of Arming Lists

3 million

Max. Number of Static Lists

10 million

Imported Video Analysis

Single video size ≤ 10 GB, and total video size ≤ 100 GB Max. 40 times accelerated analysis speed Format: MP4, AVI, TS, PS, MBF, ASF, SVAC

Facial Video Format

Resolution: 720 P to 8 MP Real-time video stream: H.264, Hikvision H.265 Local recording: H.264, Hikvision H.265

Video Frame Rate Requirement

12 to 30 frames per second (25 frames per second recommended)

Face Picture Format

JPEG, BMP, TIF, PNG

Face Picture Requirement

Face analysis target: Resolution: 48 × 48 to 4096 × 4096 (16 MP) Size: ≤ 8 MB Pupil distance: 20 to 300 pixels recommended Face angle deviation: Face analysis: ≤ 35° (vertical); ≤55° (horizontal) Face picture of list library: ≤ 25° (vertical); ≤30° (horizontal)

Video Structuralization

Analysis Type

Supports analysis and modelling for human bodies or vehicles in pictures, real-time videos or recordings.

Human Body Attribute

Hair style, clothes type, clothes color, as well as the following behaviors: wearing a hat, a mask, or glasses; carrying a bag or luggage; driving a non-motor vehicle (the number of passengers, vehicle type, vehicle size, driving direction, driving speed)

Vehicle Attribute

Vehicle color, vehicle type, vehicle main brand, license plate number

Analysis of Human Body and Vehicle Videos

90 channels (2 MP) 48 channels (2 MP to 5 MP) 24 channels (5 MP to 8 MP)

Analysis of Human Body and Vehicle Pictures

180 pieces per second

Imported Video Analysis

Single video size ≤ 10 GB, and total video size ≤ 100 GB Max. 40 times accelerated analysis speed Format: MP4, AVI, TS, PS, MBF, ASF, SVAC

Video Format

Resolution: 720 P to 8 MP Real-time video stream: H.264, Hikvision H.265

Video Frame Rate Requirement

12 to 30 frames per second (25 frames per second recommended)

Picture Format

JPEG, BMP, TIF, PNG

Human Body Picture Requirement

Resolution: 256 × 256 to 4096 × 2160 (8 MP) Human body width: 70 to 300 pixels recommended Completeness: the blocking area should be less than 30 percent of the target Angle deviation: the depression angle should be no greater than 45°

Vehicle Picture Requirement

Resolution: 256 × 256 to 4096 × 2160 (8 MP) Vehicle width: 100 to 300 pixels recommended License plate width: 80 to 120 pixels recommended Angle deviation: less than 30°(horizontal) and 25°(vertical), modelling for both the front and rear

Data Storage

Picture and Structured Data

50 million

Search by Face Picture-Response Time

2 s

Search by Human Body Picture-Response Time

2 s

Search by Vehicle Picture-Response Time

2 s

Web Application

Alarm

Blocklist, allowlist, and license plate

1V1 Comparison

Supports rapid and accurate comparison of 2 face pictures

Personnel Archive

Support

Smart Search

List library search, capture library search, alarming search

Device Operation and Maintenance

Support

Product Performance

Authorization Type

Soft Encryption

Video Accelerating Capacity

Hardware Specifications

CPU

Intel Xeon Gold 6230 × 2

Memory

256 GB

Network Interface

Gigabit Ethernet self-adaptive interface × 4

Max. Power Consumption

750 W

Rated Power Consumption

1600 W

HDD Specification

240 GB M.2 × 1 + 480 GB SSD × 3 + 8 TB SATA × 4

GPU Card

Tesla T4 × 3

PSU

Hot-plugged efficient 1+1 redundant power module

Device Dimensions

715.5 mm × 482 mm × 86.8 mm (28.2 inch × 19.0 inch × 3.4 inch)

Package Dimensions

965 mm × 670 mm × 320 mm (38.0 inch × 29.4 inch × 12.6 inch)

Physical and Environmental Parameters

Storage and Transportation Temperature

-40 °C to 70 °C (-40 °F to 158 °F)

Working Temperature

5 °C to 35 °C (41 °F to 95 °F)

Working Humidity

8% to 90% (without condensation)

Product Description

Device VersionV1.4.0
Device TypeAll-in-one

Face Analysis

Face AttributeSmile, as well as behaviors of wearing a mask, or glasses
Analysis TypeSupports analysis, comparison and classification of faces in pictures, real-time videos or recordings.
Face Picture Analysis and Comparison510 pieces/second Format: JPEG, BMP, TIF, PNG
Face Video Analysis and Comparison90 channels (2 MP) 36 channels (2 MP to 5 MP) 18 channels (5 MP to 8 MP)
Concurrent Performance-1 V 1 Comparison30 pairs per second
Max. Number of List Libraries128
Max. Number of Arming Lists3 million
Max. Number of Static Lists10 million
Imported Video AnalysisSingle video size ≤ 10 GB, and total video size ≤ 100 GB Max. 40 times accelerated analysis speed Format: MP4, AVI, TS, PS, MBF, ASF, SVAC
Facial Video FormatResolution: 720 P to 8 MP Real-time video stream: H.264, Hikvision H.265 Local recording: H.264, Hikvision H.265
Video Frame Rate Requirement12 to 30 frames per second (25 frames per second recommended)
Face Picture FormatJPEG, BMP, TIF, PNG
Face Picture RequirementFace analysis target: Resolution: 48 × 48 to 4096 × 4096 (16 MP) Size: ≤ 8 MB Pupil distance: 20 to 300 pixels recommended Face angle deviation: Face analysis: ≤ 35° (vertical); ≤55° (horizontal) Face picture of list library: ≤ 25° (vertical); ≤30° (horizontal)

Video Structuralization

Analysis TypeSupports analysis and modelling for human bodies or vehicles in pictures, real-time videos or recordings.
Human Body AttributeHair style, clothes type, clothes color, as well as the following behaviors: wearing a hat, a mask, or glasses; carrying a bag or luggage; driving a non-motor vehicle (the number of passengers, vehicle type, vehicle size, driving direction, driving speed)
Vehicle AttributeVehicle color, vehicle type, vehicle main brand, license plate number
Analysis of Human Body and Vehicle Videos90 channels (2 MP) 48 channels (2 MP to 5 MP) 24 channels (5 MP to 8 MP)
Analysis of Human Body and Vehicle Pictures180 pieces per second
Imported Video AnalysisSingle video size ≤ 10 GB, and total video size ≤ 100 GB Max. 40 times accelerated analysis speed Format: MP4, AVI, TS, PS, MBF, ASF, SVAC
Video FormatResolution: 720 P to 8 MP Real-time video stream: H.264, Hikvision H.265
Video Frame Rate Requirement12 to 30 frames per second (25 frames per second recommended)
Picture FormatJPEG, BMP, TIF, PNG
Human Body Picture RequirementResolution: 256 × 256 to 4096 × 2160 (8 MP) Human body width: 70 to 300 pixels recommended Completeness: the blocking area should be less than 30 percent of the target Angle deviation: the depression angle should be no greater than 45°
Vehicle Picture RequirementResolution: 256 × 256 to 4096 × 2160 (8 MP) Vehicle width: 100 to 300 pixels recommended License plate width: 80 to 120 pixels recommended Angle deviation: less than 30°(horizontal) and 25°(vertical), modelling for both the front and rear