Genetic Algorithm Based Optimal Component Sizing for an Electric Vehicle

Publisher:
IEEE
Publication Type:
Conference Proceeding
Citation:
IECON 2013 - 39th Annual Conference of the IEEE Industrial Electronics Society, 2013, pp. 7331 - 7336
Issue Date:
2013-01
Full metadata record
Files in This Item:
Filename Description Size
Thumbnail2013001163OK.pdf Published version222 kB
Adobe PDF
The unsharp masking filter (UMF) has been widely used in image processing front ends for contrast enhancement. The filter, being easy to implement, is based on the concept of augmenting a scaled and high-passed version of the image to itself. The UMF performance is critically dependent on the generation of the highpassed signal to be added as well as its associated scale factor. However, the optimal choice of filter parameters still remains a challenging task due to possible intensity clipping problems where the filtered pixel magnitude is vulnerable to be out of the permitted display ranges. In this research, an adaptive scheme is formulated such that the scale is derived from the pixel intensity of the input image. Specifically, pixels in the mid-range intensity will be assigned a larger scaling factor according to a Gaussian-like profile. In addition, the optimal profile coefficients and the width of the high-pass generator window are determined by adopting the particle swarm optimization algorithm. Satisfactory simulation results obtained from a collection of a large set of images have shown the effectiveness of the proposed image contrast enhancement approach.
Please use this identifier to cite or link to this item: