How to add AI model to OpenMV ecosystem

(Redirected from How to add AI model to OpenMV ecosystem)

1. STM32Cube.AI enabled OpenMV firmware

This tutorial walks you through the process of integrating your own neural network into the OpenMV environment.

The OpenMV open-source project provides the source code for compiling the OpenMV H7 firmware with STM32Cube.AI enabled.

The process for using STM32Cube.AI with OpenMV is described in the following figure.

Process to use STM32Cube.AI with OpenMV
  1. Train your neural network using your favorite deep learning framework.
  2. Convert your trained network to optimized C code using STM32Cube.AI tool
  3. Download the OpenMV firmware source code, and
  4. Add the generated files to the firmware source code
  5. Compile with GCC toolchain
  6. Flash the board using OpenMV IDE
  7. Program the board with microPython and perform inference
Info white.png Information
Licence information:

X-CUBE-AI is delivered under the Mix Ultimate Liberty+OSS+3rd-party V1 software license agreement SLA0048

1.1. Prerequisites

To follow this article it is assumed that a Linux environment is used (tested with Ubuntu 18.04).

Info white.png Information
For Windows users, it is strongly recommended to install the Windows Subsystem for Linux (WSL) Ubuntu 18.04 that provides a Ubuntu Linux environment. Please note: this tutorial has only been tested with WSL1

Once the installation is done, access to the WSL Ubuntu file system can be done from the Windows File explorer at the following location:


All the commands starting with this syntax should be executed in a Linux console:


1.2. Requirements

1.2.1. Check that your environment is up-to-date

 sudo apt update
 sudo apt upgrade
 sudo apt install git zip make build-essential tree

1.2.2. Create your workspace directory

 mkdir $HOME/openmv_workspace
Info white.png Information
This is just a suggestion of directory organization.
All following command lines will refer to this directory

1.2.3. Install the stm32ai command line to generate the optimized code

  • Download the latest version of the X-CUBE-AI-Linux from ST website into your openmv_workspace directory.
Info white.png Information
  • Please make sure you download the X-CUBE-AI for Linux.
  • You need a account to download. If you don't have one, please follow the instruction to register.
  • For WSL users, you can copy the downloaded zip file into the following location in Windows:
Then you need to close and reopen your WSL Ubuntu console.
  • For WSL users, you can access your Windows filesystem in /mnt/c/ from your WSL Ubuntu console.
  • Extract the archive
 cd $HOME/openmv_workspace
 chmod 644
 mv STMicroelectronics.X-CUBE-AI.6.0.0.pack
 unzip -d X-CUBE-AI.6.0.0
 unzip -d X-CUBE-AI.6.0.0/Utilities
  • Add the stm32ai command line to your PATH.
 export PATH=$HOME/openmv_workspace/X-CUBE-AI.6.0.0/Utilities/linux:$PATH
You can verify that the stm32ai command line is properly installed:
 stm32ai --version
stm32ai - Neural Network Tools for STM32AI v1.4.1 ( v6.0.0-RC6)

1.2.4. Install the GNU Arm toolchain version 7-2018-q2 to compile the firmware

 sudo apt remove gcc-arm-none-eabi
 sudo apt autoremove
 sudo -E add-apt-repository ppa:team-gcc-arm-embedded/ppa
 sudo apt update
 sudo -E apt install gcc-arm-embedded

Alternatively, you can download the toolchain directly from ARM with

 tar xf gcc-arm-none-eabi-7-2018-q2-update-linux.tar.bz2 

and add the path to gcc-arm-none-eabi-7-2018-q2-update/bin/ to your PATH environment variable

You can verify that the GNU Arm toolchain is properly installed:

 arm-none-eabi-gcc --version
arm-none-eabi-gcc (GNU Tools for Arm Embedded Processors 7-2018-q2-update) 7.3.1 20180622 (release) [ARM/embedded-7-branch revision 261907]
Copyright (C) 2017 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
Warning white.png Warning
If you don't use this specific version of the gcc compiler, the compilation is likely to fail.

1.2.5. Check Python version

Make sure that when you type python --version in your shell, the version is Python 3.x.x. If not, create the symbolic link:

 sudo ln -s /usr/bin/python3.8 /usr/bin/python

1.2.6. Install the OpenMV IDE

Download OpenMV IDE from OpenMV website.
OpenMV IDE is used to develop microPython scripts and to flash the board.

1.3. Step 1 - Download and prepare the OpenMV project

In this section we clone the OpenMV project, checkout a known working version and create a branch.
Then we initialize the git submodules. This clones the OpenMV dependencies, such as microPython.

Info white.png Information
Check that there are no spaces in the path of openmv directory else the compilation will fail. You can check by running pwd command inside the openmv directory. If there are some spaces, move this directory to a path with no spaces.

1.3.1. Clone the OpenMV project

 cd $HOME/openmv_workspace
 git clone --recursive

1.3.2. Checkout a known working version

 cd openmv
 git checkout v3.9.4

1.4. Step 2 - Add the STM32Cube.AI library to OpenMV

Now that the OpenMV firmware is downloaded, we need to copy over the STM32Cube.AI runtime library and header files into the OpenMV project.

 cd $HOME/openmv_workspace/openmv/src/stm32cubeai
 mkdir -p AI/{Inc,Lib}
 mkdir data

Then copy the files from STM32Cube.AI to the AI directory:

 cp $HOME/openmv_workspace/X-CUBE-AI.6.0.0/Middlewares/ST/AI/Inc/* AI/Inc/
 cp $HOME/openmv_workspace/X-CUBE-AI.6.0.0/Middlewares/ST/AI/Lib/GCC/STM32H7/NetworkRuntime*_CM7_GCC.a AI/Lib/NetworkRuntime_CM7_GCC.a

After this operation, the AI directory should look like this

├── Inc
│   ├── ai_common_config.h
│   ├── ai_datatypes_defines.h
│   ├── ai_datatypes_format.h
│   ├── ai_datatypes_internal.h
│   ├── ai_log.h
│   ├── ai_math_helpers.h
│   ├── ai_network_inspector.h
│   ├── ai_platform.h
│   ├── ...
├── Lib
│   └── NetworkRuntime_CM7_GCC.a

1.5. Step 3 - Generate the code for a NN model

In this section, we train a convolutional neural network to recognize hand-written digits.
Then we generate a STM32 optimized C code for this network thanks to STM32Cube.AI.
These files will be added to OpenMV firmware source code.

1.5.1. Train a convolutional neural network

Info white.png Information
Alternatively, you can skip this step and use the pre-trained mnist_cnn.h5 file provided (see next chapter).

The convolutional neural network for digit classification (MNIST) from Keras will be used as an example. If you want to train the network, you need to have Keras installed.

To train the network and save the model to the disk, run the following commands:

 cd $HOME/openmv_workspace/openmv/src/stm32cubeai/example

1.5.2. STM32 optimized code generation

To generate the STM32 optimized code, use the stm32ai command line tool as follows:

 cd $HOME/openmv_workspace/openmv/src/stm32cubeai
 stm32ai generate -m example/mnist_cnn.h5 -o data/

The following files are generated in $HOME/openmv_workspace/openmv/src/stm32cubeai/data:

* network.h
* network.c
* network_data.h
* network_data.c

1.5.3. Preprocessing

If you need to do some special preprocessing before running the inference, you must modify the function ai_transform_input located into src/stm32cubeai/nn_st.c . By default, the code does the following:

  • Simple resizing (subsampling)
  • Conversion from unsigned char to float
  • Scaling pixels from [0,255] to [0, 1.0]

The provided example might just work out of the box for your application, but you may want to take a look at this function.

1.6. Step 4 - Compile

1.6.1. Build MicroPython cross-compiler

MicroPython cross-compiler is used to pre-compile Python scripts to .mpy files which can then be included (frozen) into the firmware/executable for a port. To build mpy-cross use:

 cd $HOME/openmv_workspace/openmv/src/micropython/mpy-cross

1.6.2. Build the firmware

  • Lower the heap section in RAM allowing more space for our neural network activation buffers.
    • For OpenMV H7: Edit src/omv/boards/OPENMV4/omv_boardconfig.h, find OMV_HEAP_SIZE and set to 230K.
    • For OpenMV H7 Plus: Edit src/omv/boards/OPENMV4P/omv_boardconfig.h, find OMV_HEAP_SIZE and set to 230K.
  • Execute the following command to compile:
    • For OpenMV H7, it is the default target board, no need to define TARGET
 cd $HOME/openmv_workspace/openmv/src/
 make clean
 make CUBEAI=1
  • For OpenMV H7 Plus, add the TARGET=OPENMV4P to make
Info white.png Information
This may take a while, you can speed up the process by adding -j4 or more (depending on your CPU) to the make command, but it can be the right time to take a coffee.
Info white.png Information
If the compilation fails with a message saying that the .heap section overflows RAM1, you can edit the file src/omv/boards/OPENMV4/omv_boardconfig.h or src/omv/boards/OPENMV4P/omv_boardconfig.h and further lower the OMV_HEAP_SIZE by a few kilobytes and try to build again.

Do not forget to run make clean between builds.

1.7. Step 5 - Flash the firmware

  • Plug the OpenMV camera to the computer using a micro-USB to USB cable.
  • Open OpenMV IDE
  • From the toolbar select Tools > Run Bootloader
  • Select the firmware file (It is located in openmv/src/build/bin/firmware.bin) and follow the instructions
Info white.png Information
For Windows users, the firmware is located here:
  • Once this is done, you can click the Connect button located at the bottom left of the IDE window

1.8. Step 6 - Program with microPython

  • Open OpenMV IDE, and click the Connect button located at the bottom left of the IDE window
  • Create a new microPython script File > New File
  • You can start from this example script running the MNIST neural network we have embedded in the firmware
Copyright (c) 2019 STMicroelectronics
This work is licensed under the MIT license

# STM32Cube.AI on OpenMV MNIST Example

import sensor, image, time, nn_st

sensor.reset()                      # Reset and initialize the sensor.
sensor.set_pixformat(sensor.GRAYSCALE) # Set pixel format to Grayscale
sensor.set_framesize(sensor.QQQVGA)   # Set frame size to 80x60
sensor.skip_frames(time = 2000)     # Wait for settings take effect.
clock = time.clock()                # Create a clock object to track the FPS.

# [STM32Cube.AI] Initialize the network
net = nn_st.loadnnst('network')

nn_input_sz = 28 # The NN input is 28x28

    clock.tick()             # Update the FPS clock.
    img = sensor.snapshot()  # Take a picture and return the image.

    # Crop in the middle (avoids vignetting)

    # Binarize the image 
    img.midpoint(2, bias=0.5, threshold=True, offset=5, invert=True)

    # [STM32Cube.AI] Run the inference
    out = net.predict(img)
    print('Network argmax output: {}'.format( out.index(max(out)) ))
    img.draw_string(0, 0, str(out.index(max(out))))
    print('FPS {}'.format(clock.fps())) # Note: OpenMV Cam runs about half as fast when connected

Take a white sheet of paper and draw numbers with a black pen, point the camera towards the paper. The code must yield the following output:

Output from camera

2. Documentation of microPython STM32Cube.AI wrapper

This section provides information about the 2 microPython functions added the the OpenMV microPython framework in order to be able to initialize and run STM32Cube.AI optimized neural network inference.

2.1. loadnnst


Initialize the network named network_name.


  • network_name : String, usually 'network'


  • A network object, used to make predictions


import nn_st
net = nn_set.loadnnst('network')

2.2. predict

out = net.predict(img)

Runs a network prediction with img as input.


  • img : Image object, from the image module of nn_st. Usually taken from sensor.snapshot()


  • Network predictions as an python list


Copyright (c) 2019 STMicroelectronics
This work is licensed under the MIT license

import sensor, image, nn_st

# Init the sensor

# Init the network
net = nn_st.loadnnst('network')

# Capture a frame
img = sensor.snapshot()

# Do the prediction
output = net.predict(img)