International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 04 | Apr 2024
p-ISSN: 2395-0072
www.irjet.net
FOOD IMAGE RECOGNITION AND CALORIE PREDICTION Nisha P.K1, Basil Kunjumon 2, Ajin Johnson3, Abhiram K Rajan4, Adams Jacob 5 1Asst.Professor,Dept.of Computer Science and Engineering, Sree Narayana Gurukulam College of Engineering,
Kadayiruppu, Kerala, India
2,3,4,5 Dept.of Computer Science and Engineering, Sree Narayana Gurukulam College of Engineering, Kadayiruppu,
Kerala, India ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - As the number of health issues linked to
recognized food items. This dual approach not only addresses the challenge of food identification but also provides valuable insights into the potential impact on an individual's overall dietary intake. The project's methodology involves the use of convolutional neural networks (CNNs) for food image recognition, training the model on a comprehensive dataset that encompasses a wide variety of cuisines and presentation styles. Subsequently, a calorie prediction model will be implemented, utilizing regression algorithms trained on a dataset containing nutritional information for a diverse range of food items. The integration of these two components will create a holistic system capable of automating the process of calorie estimation from food images.
Key Words: Food Recognition, Calorie Estimation
This project is anticipated to contribute to the emerging field of health-tech by providing a practical solution to the challenges associated with dietary monitoring. The outcomes of this research could have a substantial impact on promoting healthier eating habits, aiding individuals in achieving their fitness and wellness goals. As we embark on this exploration at the intersection of artificial intelligence and nutrition, we aim to pave the way for a more technologically empowered and health conscious society
obesity and overeating rises, individuals are increasingly mindful of their dietary habits to stave off conditions such as hypertension, diabetes, and cardiovascular diseases associated with excess weight According to data from the World Health Organization (WHO), a staggering 2.8 million people succumb annually to complications related to being overweight or obese. A pivotal aspect of any effective dietary regimen is monitoring calorie intake. Therefore, we propose a novel deep learning methodology for estimating the caloric content of food items depicted in user-captured images. Our approach employs a layered framework encompassing Image Acquisition, Food Item Classification, Surface Area Detection, and Calorie Prediction.
1.INTRODUCTION In the modern era, where technology intertwines seamlessly with daily life, there is a growing interest in leveraging artificial intelligence (AI) to address health and nutrition challenges. One such area that has gained significant attention is food image recognition coupled with calorie prediction. This project delves into the innovative realm of utilizing advanced computer vision techniques to recognize and analyze food items from images, subsequently predicting their caloric content. The significance of this project lies in its potential to revolutionize the way individuals manage their dietary habits and make informed choices about their nutrition. With the increasing prevalence of smartphones and the pervasive use of social media, people frequently capture and share images of their meals. However, manually tracking nutritional information remains a cumbersome task. By harnessing the power of AI, we aim to streamline this process and empower individuals to effortlessly obtain accurate and timely information about the caloric content of their meals. The primary objectives of this project include developing a robust food image recognition system capable of identifying various food items from images captured in diverse real-world scenarios. Additionally, the project aims to implement a calorie prediction model that leverages machine learning algorithms to estimate the nutritional content of the
© 2024, IRJET
|
Impact Factor value: 8.226
2. LITERATURE SURVEY In the paper [1] “Image-based Thai Food recognition and calorie estimation using Machine Learning” by Rattikorn Sombutkaew , Orachat Chitsobhuk. This paper contributes to the existing body of literature by introducing a novel calorie estimation system embedded within an Android mobile application. The proposed system utilizes a fusion of techniques, employing a mobile camera to capture food images and leveraging the depth information from the AR Core library. The segmentation of the food area is accomplished through the implementation of a fine-tuned Mask R-CNN, trained on a specific Thai food image dataset. Furthermore, the study incorporates various machine learning methodologies, including Linear Regression, Support Vector Regression, K-Nearest Neighbor, and Deep Neural Network, to estimate the quantity of food calories within each captured image. Notably, the Deep Neural Network emerges as the most promising model, exhibiting
|
ISO 9001:2008 Certified Journal
|
Page 1855