Vehicle Counting And Classification Using Kalman Filter And Pixel Scanner Technique And Its Verification With Optical Flow Estimation

Vehicle tracking is important in traffic monitoring systems. The behaviors of regions of moving vehicles are complicated, since the regions may combine or break during the tracking due to mistakes in vehicle detection and tracking or vehicles’ overlapping with each other, and as a result, region matching simply according to similarities between successive frames is not enough to achieve reliable results. This paper proposes a novel tracking strategy that can robustly track and classify the objects within a fixed environment. We define a robust model-based tracker and classifier using kalman filtering combined with pixel scanner. The tracking is done by fitting successively more elaborate models on the tracked region and the segmentation is done by extracting the regions of the image that are consistent with the computed model of the target. We adopt a competitive and efficient dynamic Kalman filtering to adaptively update the object model by adding new stable features as well as deleting inactive features. In the next stage we need to check each and every frame for object recognition. This work introduce a diagonal pixel scanner to identify the objects. The result is verified further by implementing optical flow analysis. The tracking, counting and classification of object/vehicle have produced very consistent result. The average accuracy with short length video clipping is greater than 98%. Keywords-Kalman filter, pixel scanner, object classification and object counting.