In this study, a spotlight filter for capturing the local structure information in tracing object's contour is proposed for object contour extraction. By searching the possible contour path between the initial source and destination contour points, the spotlight filter is able to capture the local structure information along the path, filter out the edge candidate pixels with weak evidence of being on contour and extract the correct contour. The spotlight filtering can then be operated in parallel on all pairs of initial source and destination contour points of an object to achieve a closed contour of the object. The proposed spotlight filter is tested on a synthetic object image with noise and a real medical image and better results are obtained in terms of accuracy and computational efficiency by comparing with other contour following methods.

Mesh generation and adaption for solving dopant diffusion in process simulation is a difficult task, complicated by both the moving boundaries of oxide growth and the time dependence of the solutions. For both computational and ease of use reasons, automatic mesh generation and discretization error control is desirable. This paper describes an approach based on local error estimates to refine the mesh. The results presented in this paper extend our previous work to two-dimensional problems. The implementation of this approach is done in a process simulator, the Florida Object-Oriented Process Simulator. Several test cases are described to demonstrate the effectiveness of the algorithms.

The authors point out that modern numerical process simulators are becoming increasingly complicated in both physical models and domain shape. Grid generation is difficult for these simulators because of the inherently transient nature of the problems being solved. Here, adaptive grid refinement is considered for use in solving diffusion problems. Higher order approximations to the discretized diffusion flux are also studied. Several methods of both adaptive grid refinement and discretization are investigated and compared in terms of CPU time and final discretization error. All the methods are directly applied to the one-dimensional version of SUPREM-IV.>