Browsing by Author "Fernando, WSP"

Now showing 1 - 5 of 5

item: Thesis-Abstract
CEnhanced camshift kalman filter for object tracking
(2015-11-27) Fernando, WSP; Cooray, TMJA
In this thesis an enhanced Cam-shift Kalman object tracking algorithm for video surveillance and object tracking was developed. And this new algorithm was based on a modified Cam-shift tracking algorithm and the Kalman filter. This modified Cam-Shift algorithm solves a major drawback in the classical Cam-Shift algorithm such that the search area for the next frame was optimized, so that the time taken to track the object was minimized. The classical Cam-Shift algorithm for tracking performs well under perfectly maintained conditions such as light condition and without partial occlusions that constitute a good tracking method. However, under different environment conditions and with occlusions the algorithm fails. To test the performance of the enhanced Cam-Shift algorithm color of the object was selected as the feature for identifying the object, and was compared with the performance of the classical Cam-Shift algorithm. Also mean-shift algorithm was also incorporated for the comparison. In order to enhance the performance and accuracy under cluttered environment, the presence of noise and occlusions Kalman filter was combined. When the object disappears from the scene partially or fully the algorithm is capable of tracking the object. The experimental results verifies the ability of the enhanced Cam-shift Kalman object tracking algorithm in comparison to the classical Cam-Shift, which can locate the target object more effectively.
item: Conference-Extended-Abstract
Current status of relative motion detection and segmentation of moving objects for navigation using optical flow
(2007) Fernando, WSP; Udawatte, L; Pathirana, P
Motion Control is one of the most important research topics in computer vision. It is the base for many other problems such as visual tracking, structure from motion. A major interests of motion analysis is to estimate 3D motion.
item: Conference-Abstract
Influence of varying supply yarn package diameter on the pirn winding tension
(2011) Fernando, ESK; Jayawardane, TSS; Fernando, WSP
Pirn winding is an operation of winding yarn from supply yarn package onto pirns which are used for shuttle weft insertion. As the package size varies when winding pirns, the varying balloon effect caused to change the thread tension at the winding point of the pirn. As a number of pirns can be wound from a single supply yarn package, pirns are wound from different yarn package diameters. A significant change in take off tension takes place from pirn to pirn. Placing dead weights on the disc tension controller which adds a tension to take off tension may help to compensate the yarn tension variation to some extent. However, incorrect timing of compensation and stepwise compensation may create significant tension variations in pirn windings. The authors attempted to design a tension control device by applying an automatically continuously varying electromagnetic force on the dead weight, so that the variation of take off tension due to change of package diameter is compensated and investigate influence of varying supply yarn package diameter on the fmal winding tens ion of pirn.
item: Conference-Abstract
Object detection with Artificial neural networks combined with dyadic image downscaling using Haar transform
(2008) Fernando, WSP; Udawatta, L
This paper describes the methodology for identifying a moving object by a fast dyadic image downscaling combined with a neural network (a multi layer perceptron). Given a sequence of images here we basically try to detect a particular type of object (here we used cars) which can be located in different distances. Either it is closer to the or moving away from the camera.
item: Thesis-Abstract
Real-time detection and tracking of vehicles with lane detection
(3/23/2012) Fernando, WSP; Udawatta, L; Pathirana, P
In this research, a computer vision based procedure for navigating an autonomous vehicle safely in a sub-urban road under an unstructured environment was described. This was analyzed in two main areas. Namely; an on road object detection method, where we are only concerned of detecting cars, and a novel method in detecting road lane boundaries. For the detection of vehicles (cars) from an on-road image sequence taken by a monocular video capturing device in real time and an algorithm of multi resolution technique based on Haar basis functions were used for the wavelet transform, where a combination of classification was carried out with the multilayer feed forward neural network. The classification is done in a reduced dimensional space, where Principle Component Analysis (PCA) dimensional reduction technique has been applied to make the classification process much more efficient. Then, the other approach used is based on boosting which also yields very good detection rates. In general, boosting is one of the most important developments in classification methodology. It works by sequentially applying a classification algorithm to reweighed versions of the training data, followed by taking a weighted majority vote of the sequence of classifiers thus produced. For this work, a strong classifier was trained by the discrete adaboost algorithm and its variants. In this thesis, a novel algorithm for detection of lane boundaries was presented. Initially, the method fits the CIE L*a*b* transformed road chromaticity values (that is a* and b* values) to a bi-variate Gaussian model followed by the classification of road area based on Mahalanobis distance. Then, the classified road area acts as an arbitrary shaped region or a mask in order to extract blobs resulting from the filtered image by a two dimensional Gabor filter. This is considered as the first visual cue. Another visual cue of images was employed by an entropy image. Moreover, the results from color based visual cue and visual cue based on entropy were integrated following an outlier removing process. Finally, the correct road lane points are fitted with Bezier splines which act as control points that can form arbitrary shapes. The algorithm was implemented and experiments were carried out on sub-urban roads.