Skip to Content

Digit Recognition AI/ML Application on SAM E51 IGAT Curiosity Evaluation Kit Using MPLAB® Harmony v3: Step 10

Last modified by Microchip on 2023/11/09 09:08

Contents

Build and Program the Application

Clean and build your application by clicking on the Clean and Build clean and build icon icon.

​In case of compilation errors, recheck the steps and build the project again.

Program your application to the device by clicking on the Make and Program Make and program icon icon.
Open the Tera Term application (or another terminal console) on your PC and navigate to File > New connection.
Terra Term new connection
Click image to enlarge.
Navigate to the Serial box in the opened window and select the port number that corresponds to the Embedded Debugger (EDGB) Virtual COM Port of your connected SAM E51 IGAT Xplained Pro Evaluation Kit. Then, press OK to open a serial connection.
Terra Term new connection window
Click image to enlarge.
Open the Serial port configuration of Tera Term by navigating to Setup > Serial port.
Terra Term serial port
Click image to enlarge.
In the Serial Port window, verify that the Baud rate is set to 115200 and other elements are set as shown, then press OK.
115200, 8, none, 1
Click image to enlarge.
This shows what you should see on the SAM E51 IGAT Curiosity Evaluation Kit display.
Display reset screen
Click image to enlarge.
Draw any digit (from 0 to 9) on the SAM E51 IGAT Curiosity Evaluation Kit display as shown.
digit drawn
Click image to enlarge.
The recognized digit will be displayed as shown.
result screen
Click image to enlarge.

The scores for the corresponding 10 digits (indexes) will be displayed on the Tera-Term as shown.

You can repeat the different digits (between 0 to 9) drawing on the display to verify the digit recognition.

Tera Term output
Click image to enlarge.

Back to Top

Results

You successfully created a Digit Recognition Artificial AI/ML application on a SAM E51 Integrated Graphics and Touch Curiosity Evaluation Kit. Also, you observed that the application recognized handwritten digits (numbers 0 to 9) drawn on the display of the SAM E51 IGAT Curiosity Evaluation kit, and the result was displayed on the same display screen.

Back to Top

Analysis

You have successfully created your first AI/ML application using MPLAB® Harmony v3 on the SAM E51 microcontroller. Your application used the TensorFlow model for convolutional neural network problems, Touch Libraries, Legato Graphic Libraries, System Services, and all the fundamental elements like PLIBs and drivers that go into building a real-time application. Your application successfully recognized handwritten digits (numbers 0 to 9), and the result was displayed on the display screen.

In this application, you used MPLAB Harmony Configurator (MCC) and the MPLAB Harmony v3 Framework to configure the SAM E51. You verified the clock configurator to set up the CPU clock and peripheral clocks. You configured several peripherals, drivers, and the Timer System Service. You also configured the Touch Library and designed Display Graphics on the Legato Graphics Composer. You used the Pin Configurator to set up the pins for Touch Display and several peripheral pins.

Back to Top

Conclusions

This tutorial provided you with the training of training a TensorFlow Machine Learning model and deploying it on an MCU for an AI/ML real-time application on a SAM E51 microcontroller with MPLAB Harmony v3 Framework. As a next step, you may customize this application and reconfigure some of the components used in this tutorial. You could also add new components (train a different model for another problem, MPLAB Harmony services, middleware components, drivers, etc.) to enhance this application to realize your end application.

Back to Top