I am working on the feature extraction from the binarized license plate characters. I want to extract features and run machine learning algorithms for the classification of the characters.
Dear Yonten
please see the following paper:
Article A new approach for rotation-invariant and noise-resistant te...
It may be useful for your application.
Dear Yonten Jamtsho,
You may try Bag-of-features which is an adaption of bag-of-words. Also, you may use GLCM features for your gray-scale images. For your work, CNNs features will give better results, since they are designed for image processing tasks.
Although the application is different from your area, you can check the following papers to see how these feature extraction methods affect the performance of the classification systems for specific areas.
Conference Paper A Comparative Analysis on Fruit Freshness Classification
Conference Paper Meat Quality Assessment based on Deep Learning
- Badges
- Science topic
- Similar topics
- Biological Science
- Kinesiology
- Running