Scalp Eye: A Machine Learning-Based Scalp Hair Inspection and Diagnosis System for Scalp Health

This paper introduces an automated system for scalp hair inspection and classification utilizing image processing and machine learning techniques, implemented in MATLAB. The system provides a graphical user interface with a selection dialogue box allowing users to choose between 'Zoom In' and 'Zoom Out' image processing options. The 'Zoom Out' feature reads scalp images, applies noise reduction, adjusts image quality, and extracts features through Histogram of Oriented Gradients (HOG). These features are then classified using a Support Vector Machine (SVM) model trained on a labelled dataset. Once classified, the results are displayed to the user and sent via email. The 'Zoom In' option follows a similar workflow, with a different dataset and corresponding SVM model. This automation allows for efficient and accurate diagnosis of various scalp and hair conditions. The integration of email functionality facilitates easy sharing of classification results with medical professionals or patients. The proposed system has the potential to improve diagnostic accuracy and reduce manual inspection efforts in dermatology and trichology applications.

Keywords: Pre-Processing, support vector machine, machine learning, Feature Extraction, Classification, Accuracy.