Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Abnormal Activity Monitoring Using Kalman Filter with Adaptive Background Updation
https://orcid.org/http://orcid.org/0000-0001-6352-1207
Object detection and tracking system is an emerging topic now a day. It has variety of applications in military and surveillance systems. The main objective of this project is object detection (subjective object), feature extraction and tracking of target object using Kalman filter in real time. A camera is interfaced with P.C which uses Matlab to detect and capture the image of the scene. If the object is present in the image it will be isolated and its features will be extracted. Then it will be continuously tracked using Kalman filter. We present a real-time system for recognizing human activities, applying the Radon transform (RT), Principal Component Analysis (PCA), and Linear Discriminant Analysis (LDA) enhances low-frequency components. While PCA captures their global representation in a reduced number of eigenvectors. Our proposed approach computes radon projections in various directions to extract directional features from video sequences. PCA is then employed to reduce the dimensionality of the radon shape features, followed by LDA to enhance class separation. Our aim objective is to develop an effective real-time recognition system by combining both local and global features. To evaluate the system, we have created a dataset containing normal and abnormal activities. ANN (Artificial Neural Networks) are utilized to recognize different humanoid movements/activities in real-time recognition with 78% for human activities. Experimental outcomes demonstrate superior recognition performance of our system compared to hi-tech methods.
Abnormal Activity Monitoring Using Kalman Filter with Adaptive Background Updation
https://orcid.org/http://orcid.org/0000-0001-6352-1207
Object detection and tracking system is an emerging topic now a day. It has variety of applications in military and surveillance systems. The main objective of this project is object detection (subjective object), feature extraction and tracking of target object using Kalman filter in real time. A camera is interfaced with P.C which uses Matlab to detect and capture the image of the scene. If the object is present in the image it will be isolated and its features will be extracted. Then it will be continuously tracked using Kalman filter. We present a real-time system for recognizing human activities, applying the Radon transform (RT), Principal Component Analysis (PCA), and Linear Discriminant Analysis (LDA) enhances low-frequency components. While PCA captures their global representation in a reduced number of eigenvectors. Our proposed approach computes radon projections in various directions to extract directional features from video sequences. PCA is then employed to reduce the dimensionality of the radon shape features, followed by LDA to enhance class separation. Our aim objective is to develop an effective real-time recognition system by combining both local and global features. To evaluate the system, we have created a dataset containing normal and abnormal activities. ANN (Artificial Neural Networks) are utilized to recognize different humanoid movements/activities in real-time recognition with 78% for human activities. Experimental outcomes demonstrate superior recognition performance of our system compared to hi-tech methods.
Abnormal Activity Monitoring Using Kalman Filter with Adaptive Background Updation
https://orcid.org/http://orcid.org/0000-0001-6352-1207
Object detection and tracking system is an emerging topic now a day. It has variety of applications in military and surveillance systems. The main objective of this project is object detection (subjective object), feature extraction and tracking of target object using Kalman filter in real time. A camera is interfaced with P.C which uses Matlab to detect and capture the image of the scene. If the object is present in the image it will be isolated and its features will be extracted. Then it will be continuously tracked using Kalman filter. We present a real-time system for recognizing human activities, applying the Radon transform (RT), Principal Component Analysis (PCA), and Linear Discriminant Analysis (LDA) enhances low-frequency components. While PCA captures their global representation in a reduced number of eigenvectors. Our proposed approach computes radon projections in various directions to extract directional features from video sequences. PCA is then employed to reduce the dimensionality of the radon shape features, followed by LDA to enhance class separation. Our aim objective is to develop an effective real-time recognition system by combining both local and global features. To evaluate the system, we have created a dataset containing normal and abnormal activities. ANN (Artificial Neural Networks) are utilized to recognize different humanoid movements/activities in real-time recognition with 78% for human activities. Experimental outcomes demonstrate superior recognition performance of our system compared to hi-tech methods.
Abnormal Activity Monitoring Using Kalman Filter with Adaptive Background Updation
Lecture Notes in Civil Engineering
Mansour, Yasser (Herausgeber:in) / Subramaniam, Umashankar (Herausgeber:in) / Mustaffa, Zahiraniza (Herausgeber:in) / Abdelhadi, Abdelhakim (Herausgeber:in) / Al-Atroush, Mohamed (Herausgeber:in) / Abowardah, Eman (Herausgeber:in) / Javed, Mubashar (Autor:in) / Munir, Mehr E. (Autor:in)
Proceedings of the International Conference on Sustainability: Developments and Innovations ; 2024 ; Riyadh, Saudi Arabia
27.10.2024
7 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Kalman Filter , Abnormal Activity Monitoring , Image Processing Engineering , Building Construction and Design , Geoengineering, Foundations, Hydraulics , Sustainable Architecture/Green Buildings , Engineering Economics, Organization, Logistics, Marketing , Energy Policy, Economics and Management , Renewable and Green Energy
Mammogram segmentation using maximal cell strength updation in cellular automata
| Springer Verlag | 2015
An Optimal Adaptive Kalman Filter
| Online Contents | 2006
An Optimal Adaptive Kalman Filter
| Online Contents | 2006
Adaptive Kalman Filter Approach for Road Geometry Estimation
| British Library Conference Proceedings | 2003