Malaysian Noodle Images Classification System Using CNN and Transfer Learning

Abstract-Image Recognition is a term used to describe a set of algorithms and technologies that attempt to analyze images and understand the hidden representations of features behind them and apply these learned representations for different tasks like classifying images into distinct categories automatically, understanding which objects are present and where in an image, etc. These technologies leverage various traditional computer vision methods as well as machine learning and deep learning algorithms to achieve the required results for solving such problems. This paper shows a recognition model for classifying Malaysian Noodle images. Convolutional Neural Network (CNN) algorithms, a deep learning technique extensively applied to image recognition were used for this task. The model uses a deep learning process that was trained on natural images (AlexNet and SqueezeNet dataset) and was fine-tuned to generate the predictive Noodle model, which comprised approximately 4308 images. The dataset was divided into ten groups/categories of Noodles images which include the following: Mee Bee Hoon Goreng, Mee Bee Hoon Sup, Mee Goreng, Mee Koay Teow Goreng, Mee Koay Teow Sup, Mee Laksa Goreng, Mee Laksa Sup, Mee Maggi Goreng, Mee Maggi Sup, Mee Sup. The trained model achieved high accuracy on the test set, demonstrating the feasibility of this approach.