Deep Learning Approaches for Hair Disease Classification: A Comparative Analysis of MobileNetV2 and VGG19 Architectures

This study compares the performance of two CNN architectures, MobileNetV2 and VGG19, for classifying hair diseases using a dataset of 12,000 images covering 10 different conditions, including Alopecia Areata and Male Pattern Baldness. The dataset was preprocessed and split into training, validation, and testing sets. Both models were trained using the Adam optimizer over 20 epochs. VGG19 achieved a slightly higher accuracy of 98% compared to MobileNetV2's 97%, and it also showed better precision and recall for specific conditions. However, MobileNetV2 was faster and required fewer computational resources, making it suitable for resource-constrained environments. The study highlights the trade-off between accuracy and efficiency, suggesting that the choice of model should depend on specific requirements, particularly in telemedicine and resource-limited settings.