FPGA implementation of multi-dimensional Kalman filter for object tracking and motion detection

A method of abnormal motion detection for intelligent video surveillance is presented, which includes object intrusion detection, object overlong stay detection and object overpopulation detection. Background subtraction algorithm is used to detect moving objects in video streams. Kalman filter is applied for object tracking. By the construction of relation matrix, the tracking process is divided into five statuses for prediction and estimation, which are object disappearing, object separating, new object appearing, object sheltering and object matching. The object parameters and predictive information in the next frame which is used to track moving objects is established by Kalman filter. Then, three types of abnormal motion detection are implemented. The relative position of alarm area or guard line with the rectangle boxes of the moving objects is used to detect whether the object is invading. The existing time of the moving objects in monitor area is counted to detect whether the object is staying too long. Moving objects in the monitor area are classified and counted to detect whether the objects are too much. Alarm will be triggered when abnormal motion detection as defined is detected in the monitor area.

Download Full-text

Spatio-Temporal Context, Correlation Filter and Measurement Estimation Collaboration Based Visual Object Tracking

Sensors ◽

10.3390/s21082841 ◽

2021 ◽

Vol 21 (8) ◽

pp. 2841

Author(s):

Khizer Mehmood ◽

Abdul Jalil ◽

Ahmad Ali ◽

Baber Khan ◽

Maria Murad ◽

...

Keyword(s):

Kalman Filter ◽

Object Tracking ◽

Environmental Changes ◽

Correlation Filter ◽

Temporal Context ◽

Visual Object ◽

Target Model ◽

Change Of Scale ◽

Tracking Model ◽

Spatio Temporal

Despite eminent progress in recent years, various challenges associated with object tracking algorithms such as scale variations, partial or full occlusions, background clutters, illumination variations are still required to be resolved with improved estimation for real-time applications. This paper proposes a robust and fast algorithm for object tracking based on spatio-temporal context (STC). A pyramid representation-based scale correlation filter is incorporated to overcome the STC’s inability on the rapid change of scale of target. It learns appearance induced by variations in the target scale sampled at a different set of scales. During occlusion, most correlation filter trackers start drifting due to the wrong update of samples. To prevent the target model from drift, an occlusion detection and handling mechanism are incorporated. Occlusion is detected from the peak correlation score of the response map. It continuously predicts target location during occlusion and passes it to the STC tracking model. After the successful detection of occlusion, an extended Kalman filter is used for occlusion handling. This decreases the chance of tracking failure as the Kalman filter continuously updates itself and the tracking model. Further improvement to the model is provided by fusion with average peak to correlation energy (APCE) criteria, which automatically update the target model to deal with environmental changes. Extensive calculations on the benchmark datasets indicate the efficacy of the proposed tracking method with state of the art in terms of performance analysis.

Download Full-text