Facial Emotion Recognition Using Canny Edge Detection Operator and Histogram of Oriented Gradients

• Published in: Journal of Multimedia Information System Vol. 12; no. 1; pp. 1 - 12
• Main Authors: Jo, Heesun, Kwon, Beom
• Format: Journal Article
• Language: English
• Published: 한국멀티미디어학회 31.03.2025
• Subjects: 컴퓨터학
• ISSN: 2383-7632; 2383-7632
• DOI: 10.33851/JMIS.2025.12.1.1

Facial emotion recognition (FER) has received considerable attention from researchers due to its wide range of potential applications, such as human-computer interaction, marketing, customer service, education, security, and mental health care. In this study, we propose a method for recognizing human emotions from facial images using the Canny edge detection operator and histogram of oriented gradients (HOG). To extract contour and wrinkle information corresponding to emotional states from facial images, the Canny edge detection operator is applied to detect edge features. These contour and wrinkle patterns are critical because they provide valuable cues that reflect subtle changes in facial expressions, which are essential for accurately identifying emotions. Then, HOG is applied to the edge-detected image to quantify the edge features and use them as features for FER. To demonstrate the effectiveness of the proposed features in the FER task, we conducted a perfor-mance evaluation on four machine learning (ML) models using two publicly available FER datasets: JAFFE (Japanese Female Facial Expres-sion) and CK+ (Extended Cohn-Kanade). The experimental results showed that all four ML models achieved state-of-the-art performance on the test set when trained using our proposed features. KCI Citation Count: 0