Comparative Study of Three Image Enhancement Techniques for Geospatial Data

Peter Ekow Baffoe

doi:10.11648/j.ajmcm.20190402.13

Abstract

Processing images for Geomatic works is one of the most difficult techniques. The image enhancement algorithms have direct effect on the quality of images. It is normally done to improve visual appearance and provide a better technique for future automated image processing. Sources of mages include satellite, photography and aerial photogrammetry that are used for geospatial data processing. These images suffer from poor contrast and noise. To use these images effectively, there is the need to enhance the contrast and remove the noise from the image to increase its quality. There are different techniques for image enhancement but this study focused on image interpolation. This multi-resolution technique is useful for variety of fields where fine and minor details are important. In this research, the Nearest Neighbor, Bilinear and Bicubic image interpolation algorithm were compared. Using the aforementioned techniques, two images were enhanced in order to compare their strengths and processing speed. The results of the algorithm of Nearest Neighbor had low computational time, low complexity of algorithm and poor image quality. On the other hand, the algorithms of Bilinear and Bicubic had average and high computational time, average and high complexity of algorithm and average and good image quality respectively.

Highlights

Images are major assets in geospatial data acquisition and processing
This study focuses on the non-adaptive methods of interpolation which consist mainly of the nearest neighbor, the bilinear and the bicubic methods
There are a number of interpolation algorithms that could be used to enhance an image

Summary

Introduction

Images are major assets in geospatial data acquisition and processing They are normally obtained through photography, photogrammetry or remote sensing. In order to enhance images for geospatial use many algorithms have been proposed [5,6,7] Appropriate choices of such techniques are influenced by the image modality, task at hand and viewing conditions of an application. The study of images reveals interesting features of water flow, remains concentration, geomorphology and bathymetric patterns, to name a few [10] These features are more clearly observable in images that are digitally enhanced to overcome the problem of moving targets, deficiency of light and obscure surroundings [11]. Peter Ekow Baffoe: Comparative Study of Three Image Enhancement Techniques for Geospatial Data along specific directions are calculated using directional weights. This study focuses on the non-adaptive methods of interpolation which consist mainly of the nearest neighbor, the bilinear and the bicubic methods

Acquisition of Images

Image Interpolations

First Set of Results for Images A-A and B-B

Discussion

Conclusion