• Meghna Chhanwal
• Laksha Wadhwani
• Dr. Makarand Velankar

Understanding human emotions through facial expressions is a significant aspect of human-computer interaction and artificial intelligence. This research delves into the development of a sophisticated emotion detection system employing a tandem approach utilizing Haar Cascade and Convolutional Neural Network (CNN) algorithms. The study initially explores the intricacies of facial emotion recognition, acknowledging the complexity posed by human facial features. Leveraging the Facial Expression Recognition (FER) dataset comprising over 26,000 images, this research trains and tests a model on a diverse range of emotions such as anger, happiness, neutrality, sadness, and surprise. Haar Cascade, renowned for its object detection capabilities, serves as a foundational tool for facial recognition. Its utilization involves meticulous image analysis, harnessing features, and integral image calculations to discern facial attributes. Additionally, the integration of CNN, recognized for its prowess in pattern recognition, enriches the system's capacity to interpret emotions from facial cues. This CNN architecture comprises convolutional, pooling, and fully connected layers, optimizing the extraction and understanding of intricate emotional expressions. Moreover, this paper examines the limitations and computational challenges inherent in both algorithms. Haar Cascade's computational intensity in processing large datasets and CNN's need for substantial computational resources underscore the nuanced trade-offs in algorithm selection. The research findings showcase the system's performance in recognizing emotions from facial expressions. Despite initial successes in training the CNN model with a considerable dataset, observations revealed a stabilization in learning after a specific number of iterations, indicating potential saturation in model improvement. this paper elucidates the efficacy and limitations of integrating Haar Cascade and CNN algorithms for facial emotion recognition. The findings contribute to the ongoing discourse on emotion detection systems, underscoring the need for nuanced approaches to harness the potential of facial expression analysis in human-computer interaction and emotional

Convolutional Neural Network, Pattern Recognition, FER Dataset, Haar Cascade, Music Recommendation

"Emotion Based Music Recommendation System Using Face Detection", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 12, page no.h1-h5, December-2023, Available :http://www.jetir.org/papers/JETIR2312701.pdf

"Emotion Based Music Recommendation System Using Face Detection", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 12, page no. pph1-h5, December-2023, Available at : http://www.jetir.org/papers/JETIR2312701.pdf