What is it about?
Convolutional Neural Network(CNN) is good for image classification. In this study CNNs are trained with original patterns plus artificially generated patterns. Patterns are generated through simple rotation based approach. Two different models are investigated and both outperformed conventional CNN where training performed with available patterns.
Featured Image
Why is it important?
Our finding show that performance of CNN may improve with artificial patterns.
Perspectives
This study spot light on a direction of performance improvement of CNN.
M. A. H. Akhand
Read the Original
This page is a summary of: Convolutional Neural Network Training incorporating Rotation-Based Generated Patterns and Handwritten Numeral Recognition of Major Indian Scripts, IETE Journal of Research, July 2017, Taylor & Francis,
DOI: 10.1080/03772063.2017.1351322.
You can read the full text:
Contributors
The following have contributed to this page
