Abstract: In response to the hectic pace of modern life, there's a growing need for a smartphone web app that streamlines meal preparation. Our project aims to address this need by developing a sophisticated recipe recommendation system powered by technologies such as computer vision and machine learning. The primary objective is to simplify the culinary experience for users who often find themselves uncertain about what to cook with the ingredients they have on hand. By leveraging computer vision techniques, our system can accurately identify the ingredients available to the user. This information is then processed using machine learning algorithms to generate tailored recipe suggestions. This approach eliminates the need for extensive meal planning or manual recipe searches, saving users valuable time and effort. To tackle this, we prepared an ingredient dataset containing image 12,558 images across 15 food ingredient classes. The YOLOv8 object detection model was used to detect and classify food ingredients. Additionally, the recommendation system was built using machine learning. In the end, we achieved an accuracy of 96%, which is quite impressive.
Keywords: Object Detection, YOLOv8, FastAPI, TF-IDF, Word2Vec.
Cite:
Hency Jostan Dsouza, K Sthuthi Nayak, Krishii Kirti Karkera, Melan Varghese, Mr. Shreejith K B, "Ingredient Detection and Recipe Recommendation Using Deep Learning", IJARCCE International Journal of Advanced Research in Computer and Communication Engineering, vol. 13, no. 3, 2024, Crossref https://doi.org/10.17148/IJARCCE.2024.133100.
|
DOI:
10.17148/IJARCCE.2024.133100
