الصفحة 1
الصفحة 1
Food lens = فود لينس
Food lens is an innovative application designed to revolutionize dietary management by leveraging advanced image recognition and nutritional analysis. The primary objective of this senior project is to develop a user-friendly tool that identifies various foods through a camera interface and provides detailed nutritional information. This application not only enhances the user's understanding of their dietary intake but also assists in achieving personalized health and fitness goals. The core functionality of Food Lens involves the integration of a robust image recognition system capable of accurately identifying a wide range of foods. Upon identification, the application retrieves comprehensive nutritional data, including calorie content, macronutrient distribution (proteins, fats, carbohydrates), and essential micronutrients (vitamins and minerals). This data is then seamlessly integrated into the user's dietary profile. Food Lens is designed to track the user's daily caloric intake and compare it against personalized recommendations based on age, gender, weight, height, and activity level. By maintaining a dynamic record of consumed foods, the application provides real-time feedback on the user’s nutritional progress. This feature is particularly beneficial for individuals aiming to manage weight, address dietary restrictions, or improve overall health.
Deepfake detection = اكتشاف التزييف العميق
In the rapidly evolving era of artificial intelligence, addressing the escalating threats of deepfake technology becomes a necessity because of the increasing sophistication of AI algorithms in generating deceptive content, and since it threatens the integrity of information across diverse data. The main objective is to build a sophisticated AI-driven system to detect different types of deepfake in text, audio, and images. In English text deepfake detection, multiple pre-trained tokenizers have been used, but XLNET and BERT stand out with identifying objects outside the dataset with an accuracy of 0.9809 and both have been generalized & trained using LSTM. In Arabic text deepfake detection, Arabert has been trained using LSTM which led with an accuracy of 99.53% by generalizing the model. Both English and Arabic datasets have been generated to enhance the accuracy and effectiveness of the models. Audio deepfake detection has been generalized too, using Random Forest with an accuracy of 98.259%.
