Food Image Recognition using Inception v3

Abstract

: In today's visually-driven world, food photography has become an integral part of our dining experience. From Instagram posts to restaurant menus, images of food are everywhere. But what if these images could do more than just whet our appetites? Enter food image recognition, a fascinating application of machine learning that's transforming the food industry.  But imagine a world where you could instantly identify a dish simply by pointing your phone at it. That future is becoming increasingly attainable thanks to advancements in machine learning, particularly with models like Inception v3. Inception v3 provides a solid foundation for food image recognition applications, offering a good balance between accuracy and computational efficiency. The proposed method’s objective is to study various methods for Indian food image recognition. The accuracy achieved by proposed method is 93%. The proposed method is compared with the YOLO V5 and YOLO V8 and we concluded that the proposed method has better outcomes than YOLO. The results available with proposed system is better than YOLO’s.