• Jyoti Mahur
• Rachna Sharma
• Chandani Srivastava

Passive infrared sensors are inexpensive gadgets that are often utilized as simple yet effective person-presence triggers. The objective of this work is to extend the conventional application of the detector from motion detection to motion identification and activity classification. Specifically, by adjusting the effective polarization of the sensing elements in a very PIR detector, it is possible to determine the relative direction of movement of an item traveling on the motion plane of the PIR sensor. In this work, we introduce a novel method that uses two pairs of orthogonally aligned PIR sensors to determine the relative direction of human movement (in eight evenly spaced directions). Six participants, each traveling in eight different directions, provided data for our newly designed data gathering equipment, which is equipped with four dual sensing element PIR sensors and new lenses. Based on the collected PIR signals, we have performed a classification study utilizing popular machine learning techniques as instance-based learning and support vector machines. Based on the data set gathered from two orthogonally aligned PIR sensors with different lenses, we were able to identify the direction of movement with an accuracy of over 97%, according to our findings. Furthermore, we found that we could use machine learning techniques to attain recognition accuracy between 88% and 94% by employing a smaller feature set consisting of three peak values for each PIR detector.

PIR, motion sensor, machine learning, occupancy sensing, SVM, motion detection

"Motion Direction Identification Using Pyroelectric Infrared Sensors Utilizing Machine Learning ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 11, page no.e98-e102, November-2023, Available :http://www.jetir.org/papers/JETIR2311415.pdf

"Motion Direction Identification Using Pyroelectric Infrared Sensors Utilizing Machine Learning ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 11, page no. ppe98-e102, November-2023, Available at : http://www.jetir.org/papers/JETIR2311415.pdf