OpenMV Cam M7 Text Recognition

[alpha91]

Hi all, I am new to OpenMV Cam and I am require to use OpenMV M7 to detect text and character and send the output to Raspberry Pi

I find this OpenMV from Youtube

Re-loadable CNN on the OpenMV Cam M7/H7 with CMSIS-NN

and OpenMV side having this tutorial openmv/ml/cmsisnn at more_nn_nets · kwagyeman/openmv · GitHub

I dont understand here is, is that i must setup the environment to achieve the text reading function of OpenMV M7?
Or is there anyone use M7 cam to recognize text before can guide me on this?

[kwagyeman]

It doesn’t run on the M7. It runs on the H7 model. If you want to run it on the M7 you have to re-train the network to be smaller.

If you buy the OpenMV Cam H7 we have the demo ready to run and included with the IDE.

[alpha91]

It doesn’t run on the M7. It runs on the H7 model. If you want to run it on the M7 you have to re-train the network to be smaller.

If you buy the OpenMV Cam H7 we have the demo ready to run and included with the IDE.

Oh… I didnt know that and i already get the M7.
Is there any guide on how to re-train the network?
To be honest, i just get to understand the Neuron network thing and still trying to figure how it actually works.
I am totally beginner in this

[iabdalkader]

Is there any guide on how to re-train the network?

Yes, to retrain a network just follow the guide for the network you want:

https://github.com/openmv/openmv/blob/master/ml/cmsisnn/README.md

Of course you need to reduce the size of the network to fit the F7 (maybe just reduce IP1 to 50 or something).

This is a good introduction to NNs:
https://adeshpande3.github.io/adeshpande3.github.io/A-Beginner’s-Guide-To-Understanding-Convolutional-Neural-Networks/

A Beginner's Guide To Understanding Convolutional Neural Networks Part 2

And this is a practical tutorial:

A Practical Introduction to Deep Learning with Caffe and Python // Adil...

[alpha91]

Is there any guide on how to re-train the network?

Yes, to retrain a network just follow the guide for the network you want:

https://github.com/openmv/openmv/blob/master/ml/cmsisnn/README.md

Of course you need to reduce the size of the network to fit the F7 (maybe just reduce IP1 to 50 or something).

This is a good introduction to NNs:
https://adeshpande3.github.io/adeshpande3.github.io/A-Beginner’s-Guide-To-Understanding-Convolutional-Neural-Networks/
A Beginner's Guide To Understanding Convolutional Neural Networks Part 2 – Adit Deshpande – Engineering at Forward | UCLA CS '19

And this is a practical tutorial:
A Practical Introduction to Deep Learning with Caffe and Python // Adil Moujahid // Bridging Tech and Art

Hi, alright. thanks for sharing.

[alpha91]

Is there any guide on how to re-train the network?

Yes, to retrain a network just follow the guide for the network you want:

https://github.com/openmv/openmv/blob/master/ml/cmsisnn/README.md

Of course you need to reduce the size of the network to fit the F7 (maybe just reduce IP1 to 50 or something).

This is a good introduction to NNs:
https://adeshpande3.github.io/adeshpande3.github.io/A-Beginner’s-Guide-To-Understanding-Convolutional-Neural-Networks/
A Beginner's Guide To Understanding Convolutional Neural Networks Part 2 – Adit Deshpande – Engineering at Forward | UCLA CS '19

And this is a practical tutorial:
A Practical Introduction to Deep Learning with Caffe and Python // Adil Moujahid // Bridging Tech and Art

Hi, sorry. Can I ask for setting up the Ubuntu for deep learning is for what purpose?
Is the OpenMV cam have to work with that environment?
Actually what I thinking about the text recognition function is that I need to write a Python program and load into OpenMV M7, which my OpenMV camera is a standalone device for this function. Then I can write a function for it to send out the reading result via I2C.

Is my concept on this correct?

[iabdalkader]

It’s for retraining the network, you said you wanted to retrain the network to work on the F7.

[alpha91]

I see… alright. thanks