Because of the holidays, TI E2E™ design support forum responses will be delayed from Dec. 25 through Jan. 2. Thank you for your patience.

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

[FAQ] SK-TDA4VM: How to create Object Classification Reference Design using Edge Impulse platform on TDA4VM-SK

Part Number: SK-TDA4VM
Other Parts Discussed in Thread: TDA4VM

Edge Impulse is a tool for creating, analyzing, and deploying machine learning to embedded targets. TI Processors, like the TDA4VM and AM62x, running the default Processor SDK for Linux can quickly and easily be used directly with Edge Impulse's tools, allowing developers to use their embedded processor to collect data and test their model.

For getting started with edge-impulse set up please refer below points from [FAQ] Edge Impulse Setup on TI Processors.

1. Set up device according to Starter Kit Quick-Start Guide
2. Install Edge Impulse tools in Linux
3. Set up an Edge Impulse account and create/clone a project 

Once you have completed edge impulse environment setup you can start following below listed steps.

1. Data Collection

Please refer below image, which has highlighted sections for better illustration.

 

  • Click on Data Acquisition tab on the left hand side, this will pop up page similar to above attached image.
  • Click on the Upload data tab on the top to start uploading data for classification project.
  • Click on Automatically split between training and testing , this will split your input samples into training and testing categories randomly.
  • Click on Enter Label and give suitable name to the uploaded samples. For examples if you have uploaded dog samples, then suitable name would be dogs.
  • Click on Begin upload tab to start uploading input samples.

2. Model Impulse Design

2.1 Create Impulse

Please refer below image, which has highlighted sections for better illustration.

  • Click on the top left Impulse Design tab, to start creating impulse design
  • Click on first tab, Create impulse present just below impulse design tab
  • Under Image data, select image resolution and resize model as mentioned in above image
  • Under processing block select image, as we are providing image data as input for this project
  • Under Learning block, select transfer learning
  • The specified classes will be visible under Output Features tab.

2.2 Image

Please refer below image, which has highlighted sections for better illustration.

  • Click on Image tab present inside Impulse Design section
  • Select RGB as Color depth, as we have provided colored images as input.

2.3 Transfer Learning

Please refer below image, which has highlighted sections for better illustration.


  • Click on Transfer learning tab, and select Target to TI TDA4VM as shown in above image
  • Choose Model training setting as per project requirement, one can refer above mentioned values inside training setting tab
  • Under Neural Network Architecture select mobileNetV2, one can opt for different model as per their preference for object classification.
  • The Model accuracy is visible on the right hand side as highlighted in above image.
  • Click on Start training tab, this will initiate model training.

After these steps, you can get stated with model inferencing on target device.

Please follow below mentioned points from [FAQ] Edge Impulse Setup on TI Processors.

6. Run Inference on your Starter Kit

7. Moving Forward with an Edge Impulse design

Object Classification Reference Design using Edge Impulse and TDA4VM-SK : 

Reference Design Name : Tomato Leaves Disease Classification using TDA4VM-SK and Edge Impulse.

Description : 

This Reference Design illustrate classic Object Classification problem use case specific to Agricultural Industry.

We have trained model on custom data specific to different tomato plant diseases, these models are trained on edge impulse platform by following steps mentioned above.

The model inference is captured by live real time camera sensor which is connected to TDA4VM-SK.

Reference Design Video : 

Let us know more about how you're using TI Processors and/or Edge Impulse for your Edge AI Applications!

If you have any issues, please don't hesitate to reach out to use on e2e.ti.com. Thanks!