Deploy transformed model on an Arm Cortex-M processor

The last thing to do before being able to actually run your image recognition application on your device is to compile the files we have just created.

Let's test the image acquisition application:

cd cmsisnn_demo/ 
mbed compile -m DISCO_F746NG -t GCC_ARM --source .  --source ../ML-examples/cmsisnn-cifar10/code/m7 --source ../ML-examples/cmsisnn-cifar10/camera_demo/camera_app/ --source ../CMSIS_5/CMSIS/NN/Include --source ../CMSIS_5/CMSIS/NN/Source --source ../CMSIS_5/CMSIS/DSP/Include --source ../CMSIS_5/CMSIS/DSP/Source --source ../CMSIS_5/CMSIS/Core/Include -j8 

Copy the generated .bin file to the board, directly by dragging and dropping into the board that will appear as a USB device.

Note: that this model is very easily affected by changes in light conditions, we have chosen it only for simplicity.

This is what you should see:

Note: The image classification is being done on the resized image. For this reason, make sure to keep the image you are trying to classify centred in the small square. 

Previous Next