Fall Detection Using HOG Feature Extraction and Adaptive Boosting Technique
Satwik Devle, Yogesh Jadhav, Krishna Kokate, Diptee Vishwanath Chikmurge, Sunita Barve · 2023
This research holds significant importance as it focuses on the development of a reliable and accurate fall detection system, addressing a critical need for the elderly and individuals with disabilities who are more vulnerable to fall-related incidents. The objective of this study is to utilize the Fall Detection Dataset from Kaggle to create an effective fall detection system using the Histogram of Oriented Gradients (HOG) method for feature extraction from accelerometer and gyroscope measurements. To enhance accuracy, various classification algorithms, such as LogisticRegression, K-nearest neighbors (KNN), and decision tree (DT), are explored. Additionally, ensemble learning AdaBoosting along with cross-validation techniques are employed to further improve the model's performance. By overcoming the limitations of existing fall detection systems, this research aims to provide valuable insights and contribute to the development of trustworthy fall detection systems that can offer timely assistance and enhance the safety and well-being of vulnerable individuals.