Volume 18, No. 6, 2021
Recognizing Images Using Deep Neural Networks
Gurpreet Kaur
Abstract
The article demonstrates the anomaly and its mathematical approaches and methodology. Pattern recognition has focused on handwritten maths symbols and equations. New and sophisticated algorithms for handwritten character identification have produced a more diverse handwritten digits data set. Yet handwritten data sets behave poorly. We use multiple instance learning (ML) to create a more advanced handwritten digit representation model that can compute handwritten digit data from diverse feature spaces. This study presents machine learning-based offline pattern identification methods. Multilayer Perception, Support Vector Machines, Convolutional Neural Networks, etc. The fundamental goal is pattern recognition efficiency. The research reveals classification algorithm accuracy varies. Machine learning technologies identify symbols and numbers. The Bayesian Network classifies segment binary images "roughly" for symbol initialization. Content categorization uses neural networks.
Pages: 8682-8691
Keywords: Pattern Recognition, Handwritten Recognition, Digit Recognition, Machine Learning, Machine Learning Algorithms, Neural Network, Classification Algorithms.