Deep Learning Landscape Evaluation System Integrating Poetic Emotion and Visual Features

The AI-powered Landscape & Emotion Detection system integrates advanced deep learning models for landscape classification and emotion detection from poetic texts. In the landscape prediction page, users upload an image and select their mood from a dropdown. The system utilizes pre-trained models like ResNet and MobileNet to classify the image into various landscape categories such as forest, mountain, sea, and street. Based on the user’s selected mood, the system then provides a personalized poetry suggestion to enhance the emotional experience.On the emotion detection page, users provide poetry, and the system predicts the emotion using models trained on Random Forest (RF), SVC, CNN, GRU, BERT, and DistilBERT. These models analyze the sentiment and emotion conveyed in the text and provide suggestions for places to visit based on the predicted mood, such as calming or energizing locations.The backend models were trained using state-of-the-art techniques for image classification and natural language processing (NLP). The frontend was developed using Flask, HTML, and CSS, offering an intuitive and interactive user interface. This system combines AI-driven visual intelligence and emotion recognition, providing users with personalized feedback and mood-enhancing suggestions.

Keywords:

AI, Landscape Classification, Emotion Detection, Deep Learning, CNN, ResNet, MobileNet, Random Forest, SVC, BERT, Flask, Poetry Analysis, Mood-based Suggestions, NLP, Image Classification.