A Dish Recognition Framework Using Transfer Learning

Project Code :TCMAPY629

Objective

In this work, the task of dish recognition is implemented. A novel dish recognition method based on Efficient Net architecture and transfer learning is proposed.

Abstract

Digital media dish comprehension is an intriguing issue with a significant barrier. The dish's complicated ingredient list presents a hurdle. With the advancement of deep learning, a number of useful techniques can help to partially resolve the issue. The task of dish recognition is taken into consideration in this study. It is suggested to use transfer learning and the Efficient Net architecture to create a novel dish recognition system. First, we add a number of significant layers to the EfficientNet-B0. Second, we retrain the model on a new dataset of dish images, referred to as the Food dataset, using transfer learning to make use of the best parameters discovered when pertaining the model on Image Net. The Food dataset includes pictures of Indian food that have been gathered from various sources. The proposed approach may successfully identify a dish, according to experimental findings. Additionally, it performs better than other convolutional neural network models like CNN and Mobile net. A web application is also created using the training data to assist visitors who want to learn about Indian cuisine.

KEYWORDS: Dish recognition, Nutrition analysis, Deep learning,food classification

NOTE: Without the concern of our team, please don't submit to the college. This Abstract varies based on student requirements.

Block Diagram

Specifications


H/W CONFIGURATION:

Processor:I5/Intel Processor

RAM:8GB (min)

Hard DisK:128 GB


S/W CONFIGURATION:

Operating System: Windows 10

Server-side Script:Python 3.6

IDE:PyCharm, Jupyter notebook,VS code

Libraries Used:Numpy, IO, OS, Django, Keras, pandas, tensorflow


Demo Video

mail-banner
call-banner
contact-banner
Request Video

Related Projects

Final year projects