0%

0%

AMB21/22/23 TensorFlow Lite - Person Detection

This project demonstrated how to apply a simple machine learning model trained via Google Tensor Flow, and transfer it to AMB21/22/23 board

splendidemendax

Splendide_Mendax

Following Follow project

Liked Like project

Become a Hackaday.io member

OR

Just one more thing

To make the experience fit your profile, pick a username and tell us what interests you.

OK, I'm done! Skip

Public Chat

Similar projects worth following

Description

This project demos how TFL Hello World Example works with Realtek AMB21/22/23 EVB.
This Hello World Example is meant to train a simple light-weighed Neural Network for microcontrollers, using the Google TensorFlow training platform.

Details

Please check the detailed information in the "INSTRUCTIONS" section below.

Components

1 × AmebaD [ AMB23 / AMB21 / AMB22 ]
1 × Arducam Mini 2MP Plus OV2640 SPI Camera Module
3 × LED Fiber Optics / Emitters

Build Instructions

Collapse

1

Wiring Diagram

AMB21 / AMB22 Wiring Diagram:

Connect the camera and LEDs to the RTL8722 board following the diagram.

AMB23 Wiring Diagram:
2

Firmware & Libraries Preparation

Download the Ameba customized version of TensorFlow Lite for Microcontrollers library at https://github.com/ambiot/ambd_arduino/tree/master/Arduino_zip_libraries.

Follow the instructions at https://www.arduino.cc/en/guide/libraries to install it.

Ensure that the patch files found at https://github.com/ambiot/ambd_arduino/tree/master/Ameba_misc/ are also installed.
You will also need to install the Ameba_ArduCAM library, found together with the TensorFlow Lite library.

In the Arduino IDE library manager, install the JPEGDecoder library. This example has been tested with version 1.8.0 of the JPEGDecoder library.
3

Code Modifiaction

Once the library has been installed, you will need to configure it to disable some optional components that are not compatible with the RTL8722DM. Open the following file:
Arduino/libraries/JPEGDecoder/src/User_Config.h

Make sure that both #define LOAD_SD_LIBRARY and #define LOAD_SDFAT_LIBRARY are commented out, as shown in this excerpt from the file:
//#define LOAD_SD_LIBRARY // Default SD Card library
//#define LOAD_SDFAT_LIBRARY // Use SdFat library instead, so SD Card SPI can be bit bashed

View all 8 instructions

Discussions