pdf-icon

UnitV-OV2640

SKU:U078

Description

UnitV-OV2640 is an AI vision processing camera unit equipped with the Kendryte K210, integrating a dual-core 64-bit RISC-V CPU and an advanced neural network processor edge computing system-on-chip. The UnitV-OV2640 AI camera is very compact, making it suitable for embedding into various devices. It has machine vision processing capabilities, supports multiple image recognition functions (such as real-time acquisition of the size and coordinates of detected objects, and real-time acquisition of the types of detected objects), and can perform convolutional neural network calculations under low power consumption. Therefore, UnitV-OV2640 is an excellent zero-threshold machine vision embedded solution. It supports the MicroPython development environment, making your project development code more concise. Equipped with the OV2640 2-megapixel image sensor, it is an ideal choice for machine vision projects. The body features two programmable buttons and a front-facing RGB LED indicator for easy status display. The bottom provides a HY2.0-4P compatible interface and a TYPE-C interface for connecting to the main control device. It supports TF card expansion for easier access to related materials and model files.

Features

  • Dual-core 64-bit RISC-V RV64IMAFDC (RV64GC) CPU / 400Mhz (Normal)
  • Double-precision FPU
  • 8MiB 64-bit on-chip SRAM
  • Neural Network Processor (KPU) / 0.8Tops
  • Programmable I/O Array (FPIOA)
  • AES, SHA256 Accelerator
  • Direct Memory Access Controller (DMAC)
  • Supports MicroPython
  • Firmware encryption support
  • On-board hardware resources:
    • Flash: 16M
    • Camera: OV2640
    • Buttons: 2
    • Status LED: WS2812 LED
    • Expansion card interface: TF card/Micro SD
    • Interface: HY2.0/compatible GROVE

Includes

  • 1 x UnitV-OV2640
  • 1 x HY2.0-4P Grove cable (20cm)
  • 1 x USB Type-C cable (1m)

Applications

  • Object detection/classification
  • Real-time acquisition of target size and coordinates
  • Real-time acquisition of detected target types
  • Shape recognition
  • Video recording

Specifications

Specification Parameters
Kendryte K210 Dual-core 64-bit RISC-V RV64IMAFDC (RV64GC) CPU / 400Mhz (Normal)
SRAM 8Mbit
Flash 16M
Input voltage 5V @ 500mA
KPU Neural Network Size 5.5MiB-5.9MiB
Interface TypeC x 1, HY2.0-4P (I2C+I/O+UART) x 1
RGB LED WS2812 x 1
Buttons Custom buttons x 2
Camera OV2640
FOV 65°
External storage TF Card/Micro SD
Net weight 8g
Gross weight 45g
Product dimensions 40 x 24 x 13mm
Package dimensions 70 x 50 x 30mm
Casing material Plastic (PC)

Learn

About KENDRYTE K210

Kendryte K210 is a system-on-chip (SoC) with integrated machine vision capabilities. Using TSMC's ultra-low-power 28nm advanced process, it features a dual-core 64-bit processor with excellent power efficiency, stability, and reliability. This solution aims for zero-threshold development, enabling rapid deployment in user products with AI applications.

  • Equipped with machine vision capabilities
  • Better low-power vision processing speed and accuracy
  • Equipped with a convolutional neural network hardware accelerator KPU for high-performance convolutional neural network operations
  • TSMC 28nm advanced process, temperature range -40°C to 125°C, stable and reliable
  • Supports firmware encryption, difficult to crack using ordinary methods
  • Unique programmable IO array, making product design more flexible
  • Low voltage, lower power consumption compared to systems with the same processing power
  • 3.3V/1.8V dual voltage support, no level conversion required, cost-saving

This product features a dual-core 64-bit high-performance low-power CPU based on RISC-V ISA, with the following characteristics:

  • Core count: Dual-core processor
  • Processor width: 64-bit CPU 400MHz
  • Nominal frequency: 400MHz
  • Instruction set extension: IMAFDC
  • Floating-point unit (FPU): Double precision
  • Platform interrupt management: PLIC
  • Local interrupt management: CLINT
  • Instruction cache: 32KiB x 2
  • Data cache: 32KiB x 2
  • On-chip SRAM: 8MiB

About OV2640

  • Supported output formats (8-bit):
    • YUV (422/420) /YCbCr422
    • RGB565/555
    • 8-bit compressed data
    • 8-/10-bit Raw RGB data
  • Scanning mode: Progressive
  • Camera specifications:
    • CCD size: 1/4 inch
    • Field of view: 65°
    • Maximum pixels: 2M
    • Sensor optimal resolution: 1600 x 1200

SD Card Test

Unit V currently cannot recognize all types of MicroSD cards. We have tested some common MicroSD cards, and the results are as follows.


Brand Memory Type Transfer Speed Partition Format Test Result
Kingston 8G HC Class4 FAT32 OK
Kingston 16G HC Class10 FAT32 OK
Kingston 32G HC Class10 FAT32 NO
Kingston 64G XC Class10 exFAT OK
SanDisk 16G HC Class10 FAT32 OK
SanDisk 32G HC Class10 FAT32 OK
SanDisk 64G XC Class10 / NO
SanDisk 128G XC Class10 / NO
XIAKE 16G HC Class10 FAT32 OK (Purple)
XIAKE 32G HC Class10 FAT32 OK
XIAKE 64G XC Class10 / NO
TURYE 32G HC Class10 / NO

PinMap

UnitV-OV2640

UnitV G8 G19 G18 G34, G35
Hardware RGB LED Button A Button B
HY2.0-4P Interface

Model Size

Datasheets

Softwares

Quick Start

Choose the development platform you want to use and check the corresponding tutorial & quick start.

UIFlow 1.0

UIFlow 2.0

  • Coming soon...

USB Driver

FAQ
UnitV may not work without drivers in some systems. Users can manually install the FTDI driver to fix this issue. For example, in a Win10 environment, download the driver file matching the operating system, extract it, and install it via the Device Manager. (Note: In some system environments, the driver may need to be installed twice to take effect. The unrecognized device name is usually M5Stack or USB Serial. Windows recommends using the driver file to install directly in the Device Manager (custom update), as the executable installation method may not work properly).

Other

Video