Classification Model based on Ordered Features

I am trying to build a classifier for a specific card dataset let's say cards or no cards. I am using Mobilenet trained on the Imagenet dataset as my classifier and further training it on my dataset. I am able to train it, and its performance is quite good on the dataset. Let's say my card has for different regions of interest as shown below:-

It is able to perfectly recognize the above-passed image as a card.

But I am able to fool my classifier if I change the ordering of the regions of the card and then giving it to the classifier as shown below:-

I want the classifier to learn the exact occurrences of features of cards as in First Image so that if I alter them it should be able to reject that card (like in the second image).

Is there any way to achieve this?

Topic classifier machine-learning

Category Data Science

1
answered at

You could try a recurrent neural network classifier.

Alternatively, and perhaps more easily, augment your dataset to contain images of cards with changed ordering of the card images, and train on the resulting dataset.

About

Geeks Mental is a community that publishes articles and tutorials about Web, Android, Data Science, new techniques and Linux security.