Allergy Management: Easily input their allergies and receive customized alerts and recommendations tailored to their specific needs.
Ingredient Analysis: Quickly scan product labels to identify potential allergens, making grocery shopping safer and more efficient.
Safe Product Recommendations: Access a database of allergen-free products, helping users confidently choose items that align with their dietary restrictions.
Daily Life Navigation: Navigate social events, restaurants, and travel with ease by leveraging the app's comprehensive allergy information and guidance.
Health Tracking: Keep track of allergic reactions, symptoms, and triggers over time, empowering users to make informed decisions about their health and well-being.
One challenge we encountered during the development of the Allerginie app was integrating the image recognition feature for ingredient analysis. We initially faced issues with accurately parsing and extracting allergen information from product labels due to variations in label designs and languages.
To overcome this hurdle, we implemented a multi-step approach:
Data Preprocessing: We developed algorithms to preprocess and standardize images of product labels, including resizing, cropping, and enhancing image quality to improve recognition accuracy.
Machine Learning Model: We trained a custom machine learning model using a combination of computer vision techniques and natural language processing to extract text from product labels.
Data Validation: We implemented robust data validation techniques to verify and cross-reference extracted text against known allergen databases to ensure accuracy.
User Feedback Loop: We incorporated a user feedback loop within the app, allowing users to report inaccuracies or missing information, which helped refine and improve the recognition algorithm over time.
Discussion