Edge Computing Box

8-core A53 | 17.6 TOPS Computing Power

● Provides 17.6 TOPS (INT8) peak computing power and 2.2 TFLOPS (FP32) high-precision computing power, with a single chip supporting up to 32 channels of H.264 & H.265 real-time decoding.

● Supports mainstream deep learning frameworks such as Caffe / TensorFlow / MxNet / PyTorch / ONNX / PaddlePaddle, making it one of the few edge computing devices in the industry that is compatible with both domestic and international deep learning frameworks.

● Supports Docker containerized deployment and Python development environment.

● Deep learning development toolkit, including underlying libraries (ffmpeg, OpenCV, etc.), inference deployment tools, and a series of other software tools, covering capabilities such as model optimization and efficient runtime support required during the neural network inference stage.

● Provides open-source pipeline development tool examples, integrating full data flow modules such as stream pulling, decoding, inference analysis, and stream pushing.

● Supports Gigabit Ethernet to provide high-speed network connectivity, matching its ultra-high computing power.

● Expandable with LTE wireless transmission, providing convenience for edge business deployment.

This intelligent workstation, featuring ultra-powerful computing performance and high integration, has a built-in third-generation TPU, providing 17.6 TOPS (INT8) peak computing power. The CPU processor is an 8-core ARM Cortex-A53, with a main frequency up to 2.3GHz. The IVP03C intelligent workstation supports operation in wide temperature environments and is easy to maintain and manage. It can be flexibly deployed in various AI scenarios such as OCR recognition, object detection, and content moderation, and has wide applications in smart factories, smart campuses, smart urban management, smart parks, and other fields.

Category

Type

Model Parameters

Description

Core

Con

fig

Processor

CPU

8-core ARM Cortex-A53 @ 2.3GHz

INT8

17.6 TOPS

FP32

2.2 TFLOPS

Memory

LPDDR4

Default 12 GByte

Flash Storage

eMMC

Default 32 GByte

Hard

ware

Fea

tures

Codec Performance

Video Decoding

32 channels 1080P @ 30fps

Video Encoding

2 channels 1080P @ 25fps

Image Decoding

1080P 480 frames/sec

Ethernet Port (Ethernet)

Supports 10/100/1000M network access

RS-232

Can connect to smoke detectors, infrared detectors, access control systems, and other alarm input devices, or alarm output devices such as alarm bells (Note: RS-232 cable length is recommended not to exceed 10m)

RS-485

Can connect to smoke detectors, infrared detectors, access control systems, and other alarm input devices, or alarm output devices such as alarm bells

HDMI_IN

1080P@60fps HDMI input, supports audio and video input

HDMI_OUT

1080P@60fps HDMI output, supports audio and video output

LINE_IN

One dual-channel LINE_IN audio input

LINE_OUT

One dual-channel LINE_OUT audio output

Relay

Can be used for current loads, voltage below 36V

RST

General GPIO button, can be used for application software reset

USB 3.0

Can connect to USB drives, USB mice, USB keyboards, and other devices

DEBUG

System debug port (Type-C)

MicroSD Card Slot

Can connect MicroSD card to expand storage space, Class 10 or higher recommended

GPIO

Can be used to output control signals to external devices

Power Interface

Supports DC 12V

Expandable

Features

mSATA

Supports expansion of SSD solid-state drive to increase storage space

SATA

Expand solid-state drive (shared with mSATA)

WIFI+BT

Expandable WIFI+BT function, WIFI supports IEEE 802.11 a/b/g/n/ac/ax standard protocols, requires external antenna

M.2

Expandable 4G module, supports wireless transmission of edge data

SIM Card Slot

Used with 4G wireless communication module

Linux

File System

ext4

Ubuntu 20.04 LTS

Soft

ware

Devel

opment

Media Processing

BMCV, OPENCV, FFMPEG, BMLIB

AI Development

Quantization to offline tools such as TensorFlow, Caffe, PyTorch, MxNet, and Paddle Lite

Basic

Modules

Network Settings

Command Execution

Supports static and DHCP network parameter settings

Running Status

CPU, Memory, Disk

Device Information

Device serial number, software version number

Log Management

Running status, runtime errors, etc.

Time

NTP, manual time calibration

Upgrade Management

Firmware Upgrade

Supports TF card upgrade, supports TFTP upgrade

Application Areas

Widely used in various fields such as smart factories, smart campuses, smart urban management, and smart parks.