• Mr Utpal Shrivastava
• Dr. Vikas Thada

Deep learning is rapidly growing field for classification of digital images. Convolution Neural network(CNN) has become the de-facto standard for image classification which is subfield of deep learning. The paper look at optimizing the classification of images using CNN and various deep learning optimization algorithms. The dataset comprises three different classes of images: Rock, Paper and Scissor. Various deep learning optimization algorithm such as stochastic gradient descent, ADAM, RMSProp, Adamax, Adadelta have been tried with varying number of epochs and most optimized one is chosen for predicting unseen images. Optimization algorithm look at the internal parameters of the network such as learning rate, weights, number of neurons etc. To improve and updates the internal parameters of neural network, different optimization methods are used The framework used is Keras with tensorflow as backend.

deep learning, convolution, optimization, pooling ,image classification

"Optimized Multi class classification of Images Using Deep Learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 5, page no.98-100, May-2019, Available :http://www.jetir.org/papers/JETIR1905814.pdf

"Optimized Multi class classification of Images Using Deep Learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 5, page no. pp98-100, May-2019, Available at : http://www.jetir.org/papers/JETIR1905814.pdf