56 - What are features in machine learning?

Feature engineering is an essential part of traditional machine learning. Selecting appropriate features not only speeds up the analysis time but also helps improve your analysis accuracy. This tutorial explains features from an image processing perspective.
The code from this video is available at: github.com/bnsreenu/python_fo...

[@antonberg1131]

Superb tutorials sir!

[@shyammahato7178]

good explanation of features.. thank you sir

[@Cicatka.Michal]

Just what I needed. Thank You :)

[@DigitalSreeni]

You're welcome!

[@lucycai3356]

great video!

[@yurikleb]

Great tutorials!!!!

[@DigitalSreeni]

Glad you like them!

[@harshitkaushik4584]

Great tutorials sir

[@DigitalSreeni]

Glad you like them!

[@housseynenadour2233]

I am really learning from your videos, though I am not computer vision enigneer. Can you please teach us What are different features (filter) I must know in order to process images. I really need it. Than you a lot^(inf) :)

[@user-kd1ff4gr6n]

thank you for your videos... I am coming from a geospatial background.. I am using your channel to learn image segmentation and object detection for my project.. using deep learning to detect building damages along the shores after Hurricanes ... can you please do a video on how to use the xview2 dataset for image segmentation(buildings)

[@tahirak.7565]

Great

[@prajwalbanagar3179]

how can we use Naive bayes for multi-class classification?

[@DigitalSreeni]

Approach will be very similar to what I showed using Random Forest. Define your X and Y and split into train and test X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size=0.25, random_state=0) from sklearn.naive_bayes import GaussianNB #Or other NB of your choice gnb = GaussianNB() #Instantiate the classifier y_pred = gnb.fit(X_train, y_train).predict(X_test) #fit the classifier.

[@TheHarpanOnly]

What filter did you use at 01:52 ?

[@DigitalSreeni]

Sorry, not sure. Just added a bunch of filters and did not keep track of them. Looks like some sort of band pass filter.

[@agsantiago22]

2:00 - What is that filter?

[@DigitalSreeni]

It is a band pass filter. I believe I've used Joachim Walter's FFT Filter plugin in imageJ. It removes high spatial frequencies (blurring the image) and low spatial frequencies (similar to subtracting a blurred image). It can also suppress horizontal or vertical stripes that were created by scanning an image line by line.