음식 메뉴 추천을 위한 이미지 인식 기반 필터링 시스템
Image Recognition-Based Filtering System for Food Menu Recommendation
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
This study proposes a personalized food filtering system based on image recognition. The system analyzes food images uploaded by users to identify their nutritional content, and based on this information, it suggests a list of meals that either supplement the lacking nutrients or adjust excess nutrients. The proposed food object detection model, which uses YOLO and ResNet, achieved an outstanding performance with a Mean Average Precision (mAP) of 0.99 and an accuracy of 0.94. Using the nutritional information of the detected food, an initial filtering was performed based on the balance score and risk score, and a secondary filtering is performed using ChatGPT to provide practical and personalized meal recommendations tailored to the user's preferences and dietary goals. A user satisfaction survey was conducted with 30 participants to evaluate the system, and User evaluations showed an average satisfaction score of 4.45, demonstrating the effectiveness of the system in offering a balanced, personalized, and nutritionally sound meal filtering system.
Abstract
This study proposes a personalized food filtering system based on image recognition. The system analyzes food images uploaded by users to identify their nutritional content, and based on this information, it suggests a list of meals that either supplement the lacking nutrients or adjust excess nutrients. The proposed food object detection model, which uses YOLO and ResNet, achieved an outstanding performance with a Mean Average Precision (mAP) of 0.99 and an accuracy of 0.94. Using the nutritional information of the detected food, an initial filtering was performed based on the balance score and risk score, and a secondary filtering is performed using ChatGPT to provide practical and personalized meal recommendations tailored to the user's preferences and dietary goals. A user satisfaction survey was conducted with 30 participants to evaluate the system, and User evaluations showed an average satisfaction score of 4.45, demonstrating the effectiveness of the system in offering a balanced, personalized, and nutritionally sound meal filtering system.