CLASSIFICATION OF GARMENTS FROM FASHION MNIST DATASET USING ALEXNET CNN ARCHITECTURE

Abstract

Clothing in many cultures mainly reflects very similar characteristics for human such as a social status, lifestyle and gender. Now a day’s popular fashion images are considered as attraction for humans for buying new trending things of fashion products. Fashion is considered as self-expression of the human and it can context in many ways like footwear, lifestyle, makeup and hairstyle. For the classification of clothing different techniques are applied and one of the methods is CNN which is known as Convolutional Neural Network. So, CNN is type of artificial intelligence system. It is used in image recognition and processing by varying different layers. Alexnet CNN model has been used for predicting accuracy with the use of Fashion Mnist dataset. In this model, 10 classes have been used of T-shirt, Trouser, Pullover, Dress, Coat, sandal, Shirt, Sneaker, Bag, and Ankle Boot. 25 Epochs has been applied for better result in all models.