Edge detection is the process of segmenting an image by detecting discontinuities in brightness. Several standard segmentation methods have been widely used for edge detection. However, due to inherent quality of images, these methods prove ineffective if they are applied without any preprocessing. In this paper, an image pre-processing approach has been adopted in order to get certain parameters that are useful to perform better edge detection with the standard edge detection methods. The proposed preprocessing approach involves median filtering to reduce the noise in image and then edge detection technique is carried out. Finally, Standard edge detection methods can be applied to the resultant pre-processing image and its Simulation results are show that our pre-processed approach when used with a standard edge detection method enhances its performance.

Converting color images to grayscale is used for various reasons, like for reproducing on monochrome devices, subsequent processing. Each pixel in color image is described by a triple (R, G, B) of intensities like red, green, and blue. But how do you map that to a single value i.e. grayscale value. There are three methods to convert it. Average, Luminosity, Lightness. Different color models are used for different applications such as computer graphics, image processing, TV broadcasting, and computer vision. But still now there is no particular method for converting of grayscale to color image. In this paper a new approach was introduce to convert the grayscale image to color by using an YCbCr color space technique. Simulation results are presented to show how this approach is used to convert the grayscale to color image.

This paper presents an approach to formulating and estimating simultaneous equation based econometric models as neural network mapping problems.Conventional econometric methods are briefly surveyed.Motivation for neural network based simulation is discussed. A system of equations for the US economy is estimated using neural networks, and the results are compared with the popular Two Stage Least Squares method. The results are comparable indicating that the neural network based approach is promising. The pros and cons of this approach and possible future research are briefly discussed.

No abstract

No abstract