COVID-19 Visual Insights Personal Face Mask Detection with IBM Cloud Annotation and Watson Machine Learning

The camera vision object detection is one of the related technology in computer vision such as multimedia image processing and video processing. Today, the AI visual recognition software could help us to watch and monitor in distance. It can be applied in such as Factories, hospitals, food supply chain, transportation, mobile robotics, fly machine robots, etc.

In these tough covid19 times, wearing Mask is a responsibility for everyone in public, the reason behind the mask is for the protection of others by preventing the coronavirus from spreading out. Therefore, Mask detection can be found in a variety of solutions such as Safety Mask Alert.

With machine supervised learning and deep learning object detection algorithms, we can label our input to know the output, the key values of data.

Our custom model can be applied in public places, it focus on detection for wearing mask | No mask.

How does it works?

Step 1: Prepare your own custom model datasets and start label objects using IBM Cloud Annotations.

An example of training datasets:

Step 2: Create Watson Machine Learning Instance to train your model datasets.

Step 3: Create IBM Cloud Object Storage to store your labeled objects datasets into the bucket.

Step 4: Installation.

1- Install node.js to install cloud annotations, It can be found in here: https://nodejs.org/en/download/

2- To train your model you need to install the Cloud Annotations CLI using this command:

npm install -g cloud-annotations

Make sure cacli command is working, it is a package comes with cloud annotations, cacli used for training your models datasets in ML.

If not install with cloud annotations, then you can install it with Homebrew instate, if you are using macOS:

$brew install cacli

Then, use the commands cacli to access to Watson machine learning and Cloud object storage services which it has your labeled datasets.

Some features of using cacli:

cacli login 
cacli train
cacli progress <model ID>
cacli list
cacli download <model ID>

You will be asked to provide resources for IBM services WML and COS, such as instance Id, URL, and locations.. etc. it can be found in service credentials.

Step 5: Start training your model

cacli train

Then, you can watch your model progress by passing cacli progress:

cacli progress <model ID>

Step 6: Download the model

once the model training is completed, start download it.

cacli download <model ID>

Step 7: To continue deploy and use the model, it can be found in the resources below:

conclusion

To conclude, If you want to run the same experiment, you can find an interesting Topics, Tutorials, Workshops, Demos related to AI vision and computer vision such as deep learning classification and object detection. you can find it on https://developer.ibm.com, All resources are below in references.

References

[1] https://developer.ibm.com/tutorials/iot-deep-learning-anomaly- detection-2/

[2] https://developer.ibm.com/videos/build-custom-model-watson-visual-recognition-object-detection/

[3] https://cloud.annotations.ai/login

[4] https://developer.ibm.com/?s=object+detection+

[5] https://developer.ibm.com/patterns/create-a-real-time-object-detection- app-using-watson-machine-learning/

Master of CS | CUA D.C. Area U.S.A | IBM Developer Advocate, Passionate about DevOps,containers, serverless. https://www.linkedin.com/in/mofaqalthiyabi

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store