A Comparative Study of Edge Preserving Smoothing Filters in Pyramid Based Image Fusion
Journal Title: International Journal of Modern Engineering Research (IJMER) - Year 2015, Vol 5, Issue 4
Abstract
Image fusion is the technique of merging or combining two or more images to give a single image with enhanced information content. Most of the standalone image fusion techniques have their own advantages as well as disadvantages. So it is always desirable to pursue a combination of two or more techniques to give better results. The fusion technique employed is a combination of the existing Pyramid based fusion and Edge preserving smoothing filter based fusion rules. First an edge preserving smoothing filter such as Bilateral filter or Guided filter is used to obtain base and detail layers of the input images. The base and detail layers are separately fused using Pyramid fusion rule. The fusion of base layers are controlled by a weighting factor which takes into consideration the exposure, contrast, and saturation measures of the input images. The fused base and detail layers are combined to give the final image. The work also gives a comparative study of the performance of the two common edge preserving smoothing filters - Bilateral filter and Guided filter. Commonly used Image quality metrics such as PSNR and SSIM are employed for analyzing the experimental results. An average PSNR value of 62.68 is obtained with Guided filter and 60.03 with Bilateral filter
Estimate Capital and Operating Costs for Railway Transportation in the Arab Republic of Egypt
It is necessary to estimate accurately transportation costs (capital and operational) for Egyptian National Railway (ENR) when developing transport tariffs in order to calculate the necessary subsidy to decre...
Influence of Hadoop in Big Data Analysis and Its Aspects
This paper is an effort to present the basic understanding of BIG DATA and HADOOP and its usefulness to an organization from the performance perspective. Along-with the introduction of BIG DATA, the important p...
Dynamic Spectrum Allocation in Wireless sensor Networks
Radio frequency spectrum is considered the most expensive and scarce resource among all wireless network resources, and it is closely followed by the energy consumption, especially in low energy, battery powered...
MCDM Techniques for the Selection of Material Handling Equipment in the Automobile Industry
Material Handling Equipments are utilized in different shops of an automobile industry. For culling congruous Material Handling Equipment, it is felt that some Multi Criteria Decision Making Methods must be use...
A Neural Network Model of Face Detection for Active Vision Implementation
We suggest an active vision design in a neural style to achieve face detection, or face recognition in gaze, understanding network structures on impaired face recognition in developmental disorders such as autism s...