EOXYS

Product Details

XENO+ Vision ML SOM – R2L100 is a compact, high-performance System-On-Module designed specifically for advanced video and image processing using Artificial Intelligence and Machine Learning. It acts as the central processing unit for building edge AI devices powered by Linux OS.

With a built-in AI engine delivering 1 TOPS, this SOM is capable of running complex Deep Neural Networks (DNN) and Convolutional Neural Networks (CNN) directly on video and image data, enabling real-time analytics and decision-making at the edge.

The module supports high-resolution video input ranging from 2MP to 5MP at 30 FPS in RGB/RGB-IR format through the MIPI CSI-2 interface, ensuring crystal-clear vision processing. Its solderable design with 176 LGA pins and compact 40mm x 40mm footprint makes it ideal for embedding in space-constrained devices.

The XENO+ R2L100 SOM is the perfect solution for developing smart devices requiring advanced vision capabilities, such as facial recognition, people counting, object classification, and more.

Target Applications

People Counting and Tracking: Accurately count and track people in real-time for smart surveillance and retail analytics.
Face Detection and Recognition: Enhance security and user authentication with precise face detection and recognition.
Human/Pet/Vehicle Classification: Classify and differentiate between humans, pets, and vehicles in various scenarios.
Object Classification and Recognition: Identify and classify objects for inventory management and automation.
License Plate Recognition: Efficiently recognize license plates for smart parking and traffic management.

Target Devices

Smart Video Doorbell/Intercom: Enhance security and visitor interaction with intelligent video processing.
Smart Point of Sales Systems: Integrate facial recognition for personalized customer experiences.
Smart AI Cameras: Deploy intelligent surveillance and monitoring systems with advanced vision capabilities.
Smart HMI Tablet: Create interactive human-machine interfaces with precise object recognition and tracking.

Block Diagram and Video

Specifications

Features:

High-Performance AI Engine with 1 TOPS:
Built-in AI engine delivering 1 TOPS for efficient execution of Deep Neural Networks (DNN) and Convolutional Neural Networks (CNN) on video and image data.

Enables real-time analytics and decision-making at the edge, reducing latency and enhancing performance.
Advanced Video/Image Processing:
Supports high-resolution input ranging from 2MP to 5MP at 30 FPS in RGB/RGB-IR format using the MIPI CSI-2 interface.

Delivers exceptional video quality and accurate image recognition for a wide range of vision-based applications.
Compact and Solderable Design:
Solderable footprint with 176 LGA pins, ensuring secure and reliable integration into custom hardware designs.

Compact form factor of 40mm x 40mm, suitable for space-constrained applications without compromising performance.
Linux OS Compatibility:
Designed to run on Linux OS, providing a flexible and open-source development environment.

Supports various AI frameworks, enabling developers to deploy custom AI models easily.
Edge AI Computing Capability:
Enables on-device AI computing, minimizing the need for cloud processing.

Enhances data privacy and security by processing sensitive information locally.
Versatile Integration Options:
Offers multiple I/O interfaces for seamless integration with cameras, sensors, and other peripherals.

Flexible configuration options for different application needs.
Reliable and Production-Ready:
Designed for long-term reliability and consistent performance in production environments.

Ideal for building scalable AI/ML solutions in mass production.

No	Feature	Specification
1.	Part Number	R2L100
2.	Main CPU	Dual ARM Cortex A55 @ 1.2GHz
3.	MCU	ARM Cortex M33 @ 200MHz
4.	RAM	16Bit 512MB DDR4-1600 (Expandable up to 2GB)
5.	Flash	8GB eMMC Flash (Expandable up to 32GB)
6.	Serial Flash	16MB Serial NOR Boot flash
7.	ISP	Simple ISP
8.	AI Accelerator	DRP-AI Accelerator
9.	Video CODEC	H.264 Enc/Dec 2K/30fps
10.	Graphics Engine	ARM Mali-G31 3D GPU
11.	Camera Interface	1x MIPI CSI-2 (4 lanes)
12.	Display Interface	1x MIPI DSI (4 lanes)
13.	Ethernet	2x RGMII Gigabit Ethernet interface
14.	SD Interface	1x SD interface for WiFi module or SD Card
15.	USB	2x USB2.0
16.	Other Peripherals	4x I2C, 5x UART, 2x SPI, 8x ADC, 2x CAN-FD
17.	OS	Linux OS
18.	Module Size	40mm X 40mm
19.	Footprint	Solderable with 176 pins

Technical Documentation

S.no	Title	Document Link
1.	Product Brochure	Click Here to View
2.	Use Case	Click Here to View

XENO + Vision ML SOM