Products
TD-N5104
Edge Box
TD-N5104 is an AI edge box developed independently by TVT. With the advanced deep-learning algorithm, it has powerful capability of video content analysis and extraction. Multiple intelligent analytics algorithms are supported, including face detection, face recognition, human body/ motor vehicle/non-motor vehicle detection and classification. It supports RTSP/1400 protocol, compatible with cameras of the major brands. As a result, the product can be widely used in different scenes, like smart construction site, restaurant, smart city, smart school, smart community, mixed lanes with pedestrian and vehicle lanes and so on.
Introduction
Specification
Downloads
Features
Support video access via RTSP protocol
A maximum of 4CH video input
Support 8MP/6MP/5MP/4MP/3MP/1080P video input
Support H.265S/H.265+/H.265/H.264 coding
Support human/motor vehicle/non-motor vehicle detection, classification and feature extraction
Support face capture and face feature extraction (like age, gender, whether to wear glasses/mask)
Support edge computing, easy to use and cost-effective
Support remote upgrade; module and algorithm can be updated
The target attribute information can be displayed and stored in real time
Supports search by attribute
Industrial-grade design, stable and reliable
The complete API business platform can satisfy higher business demands of customers.
Dimension
| System | Operation System | Linux |
| Video | IP Video Input | 4CH |
| Resolution | 8MP/6MP/5MP/4MP/3MP/1080P/1280×1024/960P/720P | |
| Video Output | HDMITM output; 1920×1080 | |
| Video Compression | H.264/H.265/H.265+/H.265S | |
| Audio | Network Input | 4CH IPC audio input |
| Local Input | RCA ×1 (for two-way audio) | |
| Local Output | RCA ×1 | |
| Audio Compression | G.711(U/A) | |
| Live View | Single Channel Display | Main stream display, up to 8MP@30 |
| Face | Performance | Face detection: max.4CH video stream |
| Max. 10 faces can be captured simultaneously in a single window | ||
| Support feature extraction, like gender, age, mask-wearing status, glasses-wearing status | ||
| Search | Support search by time/features | |
| Snapshot Mode | Snapshot when leaving, interval snapshot, comprehensive snapshot | |
| Human/Motor Vehicle/ Non-motor Vehicle Detection | Performance | A maximum of 4 CH detection A maximum of 4 rules can be set up |
| Attribute Display | Backpack, upper clothing type, lower clothing type, hat, mask, vehicle color and brand, non-motor vehicle color and type | |
| Perimeter Protection | Region entrance/exiting/intrusion detection, line crossing detection | |
| People Counting | Yes | |
| Leaving Detection | Yes | |
| Behavior Analysis | People Following | Yes |
| People Intrusion | Yes | |
| Snapshot Storage | Storage Type | EMMC |
| Storage Capacity | 500MB | |
| Network | Protocol | TCP/IP, DHCP, HTTP, DNS, NTP, RTSP, HTTP, HTTPS, GA/T1400, MQTT, etc. |
| Interface | Network interface | 1000M×2 |
| RS485 | RS485×1, full-duplex | |
| Alarm | 8CH alarm input; 4CH alarm output | |
| USB | USB2.0×2; USB3.0×1 | |
| Dimensions（mm） | 300×249×45 | |
| Power | DC12/2A | |
| Certificate | CE、FCC | |
| Environment Protection | RoHS 2.0、REACH、WEEE | |
| Power Consumption | ≤20W | |
| Working Environment | -10~50℃，10%~90%（humidity） |
* The terms HDMI, HDMI High-Definition Multimedia Interface, HDMI Trade dress and the HDMI Logos are trademarks or registered trademarks of HDMI Licensing Administrator, Inc.
