| Thương hiệu |
|---|
Overview
1.RV1103 is a highly integrated IPC vision processor SoC specialized for AI related applications.
2. It is based on a single-core ARM Cortex-A7 32-bit core with integrated NEON and FPU, and a built-in NPU supporting INT4 / INT8 / INT16 mixed operations with up to 0.5TOPs of computational power.
3. It adopts a new hardware-based ISP and supports multiple algorithm gas pedals, such as HDR, 3A, LSC, 3DNR, 2DNR, sharpening, de-fogging, gamma correction, etc.
4. It also features built-in 16-bit DRAM DDR2 to sustain demanding memory bandwidth, as well as built-in POR, audio codec and MAC PHY.
5. The RV1103 is a powerful processor for a variety of AI application scenarios.
6. The development board supports multiple interfaces, including GPIO, UART, SPI, I2C, USB, etc., which is convenient for developers to carry out rapid development and debugging.
Chip: RV1103
Processor: Cortex A7 1.2GHz
Neural network processor (NPU): 0.5TOPS, support int4, in8, int16
Image Processor (ISP): Maximum input 4M @30fps
Memory.
Model A: 64MB RAM without FlASH.
Model B: 64MB memory, with 128MB FlASH
Camera Interface: MIPI CSI 2-lane
USB: USB 2.0 Host/Device
GPIO: 17 GPIO pins
Network Port: None
Default Storage Media:
Type A: Micro SD card (sold separately)
Type B:SPI NAND FLASH(128MB)
Single-core ARM Cortex-A7 32-bit core with integrated NEON and FPU
Built-in Rayson Microelectronics self-developed 4th generation NPU with high computing precision, supporting int4, int8, int16 mixed quantization, of which the int8 computing power is 0.5TOPS, and the int4 computing power is up to 1.0TOPS.
Built-in self-developed 3rd generation ISP3.2, supports 4 megapixels, supports HDR, WDR, multi-level noise reduction and other image enhancement and correction algorithms.
Powerful encoding performance, support intelligent encoding, adaptive code stream saving according to the scene, save more than 50% than the conventional CBR mode code rate, so that the shooting picture is both high-definition, smaller, and double the storage space.
Built-in RISC-V MCU design supports low-power fast startup, 250ms fast image capture, and loading AI model library at the same time, which can realize face recognition in “1 second”.
Built-in 16-bit DRAM DDR2 to maintain demanding memory bandwidth.
Integrated with built-in POR, audio codec and MAC PHY.
Size: 28*21mm
1 x Development Board (with or without Pin)

