OpenAMV is an AI visual and voice development board designed based on RV1103G. The onboard rich GPIO can be used. RV1103G is a highly integrated IPC SoC suitable for AI vision related applications. Based on a single core ARM Cortex-A7 32-bit kernel, it integrates NEON and FPU. There is a 32KB I-cache and a 32KB D-cache And 128KB L2 cache. The built-in NPU supports mixed operations of INT4/INT8/INT16, with a computing power of up to 0.5TOPs. Additionally, due to its powerful compatibility The network models of frameworks such as TensorFlow/MXNet/PyTorch/Caffe can be easily converted. RV1103G introduces a new generation of fully hardware based image signal processors. The chip directly inherits numerous algorithm accelerators, Such as HDR, 3A, LSC, 3DNR, 2DNR, sharpening, defogging, gamma correction, etc. Paired with two MIPI CSIs (or LVDS RX), use Users can support multi camera video data
sdk link：https://drive.google.com/file/d/1gOVX4RHDv9fHbQt9-Iw3N4xYa2zB6Ncr/view?usp=sharing
OPENAMV AI Artificial Intelligence Development Board Linux Machine Vision Built in NPU Supports Speech Recognition
Product Introduction:
OPEN AMV is a micro Linux development board based on the Ruixin Micro RV1103 chip. RV1103 is a highly integrated IPC visual processor SoC specifically designed for artificial intelligence related applications.
It is based on a single core ARM Cortex-A7 32-bit kernel, integrating NEON and FPU, and built-in NPU support for INT4/INT8/INT16 mixed operations, with a computing power of up to 0.5TOPs.
In addition, it adopts a brand new hardware based ISP that supports multiple algorithm accelerators such as HDR, 3A LSC、3DNR、2DNR、 Sharpening, defogging, gamma correction, etc.
The built-in 16 bit DRAM DDR2 can maintain demanding memory bandwidth, as well as built-in POR, audio codec, and MAC PHY (OPEN AMV PHY has not been introduced yet).
This development board is suitable for various scenarios, including but not limited to:
- Smart home devices
- Industrial automation equipment
- Robots and Drones
- Intelligent monitoring equipment
- Intelligent transportation equipment
- Intelligent medical devices
- The development board supports multiple interfaces, including GPIO, UART, SPI, I2C, USB, etc., making it convenient for developers to quickly develop and debug. The development board supports both buildroot and Ubuntu 22.04 systems, allowing developers to choose the appropriate system based on project requirements, making it more flexible in meeting specific application requirements.
OPEN AMV parameters:
- Chip: Ruixin Micro RV1103
- Processor: Cortex A7 1.2GHz
- Neural network processor (NPU): 0.5TOPS, supports int4, int8, int16
- Graphics processor (ISP): maximum input 4M @ 30fps
- Memory: 64MB DDR2
- Camera interface: MIPI CSI – lane
- USB interface: USB2.0 Host/Device
- GPIO: 17 GPIO pins
- Network port: None
- Denne er alternativ til Luckfox Pico Pro, begge bedre enn Raspberry pi pico
- Anbefalt kamera: AI Camera Module 3 Million Pixels Compatible with Luckfox Development Board Support RV1103 Chip
- Samme kamera også solgt som; MOEBIUS AI Camera Module 3 Million Pixels Compatible with Luckfox Development Board Support RV1103 Chip
AI Camera Module 3 Million Pixels Compatible with Luckfox RV1103 Development Board Support RV1103 Chip
Product Introduction:
- Supports 3 million pixels; Featuring high sensitivity, high signal-to-noise ratio, and low illumination, it can present a more delicate and realistic night vision imaging effect, better adapting to changes in ambient light.
Application field:
- Security monitoring system
- Mobile device camera
- IoT camera
- Network camera
Product parameters:
- CMOS size: 1/2.8 inches
- Camera type: Color
- Pixel: 3 million (2304 x 1296)
- Shutter mode: roller shutter shutter
- Aperture: F2.0
- Distortion:<1%
- Field of view angle: 100 °
- Focal length: 3.95mm
- Focusing method: fixed focus
- Working temperature range: -30 ° C~+85 ° C
- Best operating temperature range: -20 ° C~+60 ° C
- Maximum image transmission rate: 2304H x 1296V@30fps 10 bit
- Output format: RAW RGB/yuv/jpeg
- Infrared night vision: supported
- Product number: MBS-CS300MP module
