Satellite Imagery Quality Evaluation using Image Quality Metrics for Quantitative Cadastral Analysis

Satellite imageries are provided in both panchromatic and multispectral images. The former have higher spatial resolutions while the latter have relatively lower spatial resolutions but are rich in spectral information. The two types of images have complementary features of spatial and spectral resolutions as such could be fused or merged to produce a better imagery that would be used in land use classification, cadastral feature analysis, change detection, maps updates, hazards monitoring and many other land related applications as well as for general quantitative image analysis. Image fusion quality evaluation approaches are divided into two main categories: qualitative and quantitative evaluation approaches. In qualitative approach, quantifying image quality is done through subjective means by human experts, whereas quantitative evaluation is based on objective performance assessments of the fusion processes. Registration of images that may be used as reference in image quality measure is a crucial step in Satellite image processing. Errors due to registration of reference images give rise to local errors in merging processes and these results in significant color and other distortions in the fused imagery. Before using the images in cadastral aspects of object extraction or recognition, the quality assessment of these image data is highly crucial and cannot be over emphasized. The paper attempts to measure the quality of high resolution multispectral images of a region that has a variety of artificial and natural patterns using some well known image quality assessment metrics or predefined quality indicators. Experimental comparisons, using Matlab as a tool, between sets of noised and referenced images of different high resolution images, demonstrated the effectiveness of the proposed method. Finally, the necessary metrically good satellite images for further quantitative cadastral analysis were practically obtained. Keywords— Satellite Imagery, Image Fusion, Quality metrics, Cadastral Analysis, Spatial resolution