Author : Ms. Shamli S. Malvadkar and Dr. Satish N. Gujar

Making lines for refrigerator take benefit of automated inspection apparatus that relies on cameras. As anrising problem, refrigerator classification based on images from its front view is potentially priceless for industrial automation of refrigerator. However, it remains an extremely challenging task because refrigerator is commonly view against dense clutter in a conditions. In this paper, we suggest an automatic refrigerator image classification method which is based on a new planning of Convolutional Neural Network (CNN). It decide the hardships in refrigerator image classification by leveraging a data-driven mechanism and together optimizing both classification and similarity constraint. To our best knowledge, this is perhaps the first time that the deep-learning architecture is practical to the field of household appliance of the refrigerator. Due to the experiments conventional out using 31,247 images of 30 categories of refrigerators, our CNN architecture create an extremely imposing accuracy of 99.96%.

Keywords: Smart Device, Smart Refrigerator, Camera, Android Application.