An Empirical Study on Hair Strength Classification using CNN and Random Forest Ensembles

This study offers a hybrid CNN-Random Forest model for hair strength categorization. The model attains a significant degree of accuracy, namely 76.77 ^% , hence showcasing its effectiveness in precisely classifying hair into several levels of strength. The model's ability to accurately classify different levels of strength is further supported by the precision, recall, and F1-scores. The architectural design consists of two convolutional layers, which are further followed by pooling, flattening, and Random Forest classification. Although the current approach has achieved success, there are opportunities for further improvement. These may include refining the parameters of the model and integrating a wider range of datasets to strengthen its ability to generalise. This study represents a notable advancement in the field of automated and accurate evaluation of hair quality. The ramifications of this phenomenon have wide-ranging effects across several sectors, including cosmetics, healthcare, and hair care management. These implications have the potential to bring about advantages in areas such as product creation, clinical diagnosis, and overall management of hair care. The shown precision highlights the potential of machine learning methodologies in transforming the realm of hair care and establishes a robust basis for future progressions in this area.