Efficient Movement Compensation and Detection Algorithm using Blob Detection and Modified Kalman Filter
Sridevi N1, M Meenakshi2
1Sridevi N*, Department of EIE, Dr. Ambedkar Institute of Technology, Bangalore, Karnataka, India. Affiliated to VTU, Belagavi, Karnataka.
2M Meenakshi, Department of EIE, Dr. Ambedkar Institute of Technology, Bangalore, Karnataka, India.
Manuscript received on June 08, 2020. | Revised Manuscript received on June 25, 2020. | Manuscript published on June 30, 2020. | PP: 1187-1193 | Volume-9 Issue-5, June 2020. | Retrieval Number: E1041069520/2020©BEIESP | DOI: 10.35940/ijeat.E1041.069520
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In various real time applications such as security and surveillance etc., detection of movement from video sequence is commonly used. In such applications, time required to detect the movement and its accuracy is very crucial. In this paper, an efficient motion compensation and detection algorithm using Blob detection and modified Kalman filter techniques is proposed. The method is mainly based on Kalman filtering technique which is modified to compensate and detect the unwanted movement caused by the camera. Also the shadow effect caused by the variation in the intensity of light and object is removed using thresholding technique. Accuracy of movement detection is improved by implementing the blob detection method. The experimental results obtained from the developed algorithm is compared with few methods existing in the literature for validation.
Keywords: Motion estimation, Motion detection, Motion compensation, Kalman Filter, Thresholding, Three Step Search, Blob detection.