An Automated Pain Assessment System Using SVM Classifier
Abstract
Facial expressions are a touchy, particular biomarker of the nearness and seriousness of pain, and computer vision and machine-learning techniques empower dependable, substantial estimation of pain-related facial expressions from video. Current pain evaluation techniques are imperfect and helpless against predisposition and under recognition of clinical pain. Automatic pain observing may help by giving a target and constant evaluation. Multiple methods have been devised which helps in identifying pain expression. In the existing system the algorithm Active apperance model (AAM) is used, which alters a full facial model of appearance and characterizes both shape variation and the texture of the model region. This approach is time consuming. This paper proposes a new approach by extrating features using viola jones and bounding box algorithm. The extracted features include different properties of painful images that can be classified by Support Vector Machine (SVM) Classifier to detect the pain. This approach provides better result compared to other state of the art