This thesis proposes a robust on-line tracking method by 1) enlarging the convergence range and 2) improving the observation memory of a phase-based appearanceadaptive “Wandering Stable Lost” (WSL) tracker, which was developed by Jepson, Fleet and El-Maraghi. The resultant tracker is demonstrated to be adaptive to temporary and permanent appearance changes in the object being tracked while at t...

We consider the problem of resequencing a prearranged set of jobs on a moving assembly line with the objective of minimizing changeover costs. A changeover cost is incurred whenever two consecutive jobs do not share the same feature. Features are assigned from a set of job-specific feasible features. Resequencing is limited by the availability of offline buffers. The problem is motivated by a v...

Product lines are often employed for the facilitation of software re-use, rapid application development and increase in productivity. Despite the numerous advantages of software product lines, the task of testing them is a cumbersome process due to the fact that the number of applications that need to be tested is exponential to the number of features represented in the product line. In this pa...

Nowadays, Software Product Line Engineering (SPLE) is an emerging software engineering paradigm, which is based on the concept of reusing software artifacts gaining from the previous software development lifecycle. Researches concerning with domain analyzing, feature modeling (FM), common and variability analyzing processes have being developed for SPLE. So, this system proposes re-engineering ...

We propose a framework of a multi-object tracking method based on image classification. Aiming to build an appearance model of a target, a sparse measurement matrix is used as a projection for dimensional reduction and online feature selection to improve discriminative power. Our main idea is putting appropriate weights on the features selected to form a feature pattern. Using our method, it is...

In the lifecycle of a software product line (SPL), incremental generalization is usually required to extend the variability of existing core assets to support the new or changed application requirements. In addition, the generalization should conform to the evolved SPL requirements which are usually represented by a feature model. In this paper, we propose a featuredriven and incremental variab...

As a new pattern classification method, Nearest Feature Line (NFL) provides an effective way to tackle the sort of pattern recognition problems where only limited data are available for training. In this paper, we explore the use of NFL for speaker identification in terms of limited data and examine how the NFL performs in such a vexing problem of various mismatches between training and test. I...

Feature-based visualization of flow fields has proven as an effective tool for flow analysis. While most flow visualization techniques operate on vector field data, our visualization techniques make use of a different simulation output: Particle Tracers. Our approach solely relies on integral lines that can be easily obtained from most simulation software. The task is the visualization of dense...

In these years, with the expansion of network-based communication tools, consumer based marketing strategies using on-line discussions have received much attention. To get the benefit from on-line discussions, methodologies for analyzing on various discussion elements are demanded. In this paper, we propose a message feature map, which is a visualization method for plotting discussion messages ...

This paper presents a new monocular SLAM algorithm that uses straight lines extracted from images to represent the environment. A line is parametrized by two pairs of azimuth and elevation angles together with the two corresponding camera centres as anchors making the feature initialization relatively straightforward. There is no redundancy in the state vector as this is a minimal representatio...