A Comparison of Shadow Detection Removal and Reconstruction Methods

Author(s):

Chithra K , THEJUS ENGINEERING COLLEGE; Rahul Ramachandran, THEJUS ENGINEERING COLLEGE; Aleena T. A., THEJUS ENGINEERING COLLEGE

Keywords:

IOOPL, Kmeans clustering, Shadow detection, Shadow removal, Reconstruction

Abstract:

Nowadays image processing is a developing area. In this area there are variety of sub areas like remote sensing, biomedical image processing etc. Even now many problems are faced during its processing. The atmosphere, land, and water of the Earth are remarkably complex and do not provide themselves well to being recorded by remote sensing devices. They have constraints like spectral, spatial and radiometric resolution. It creates a range of errors like geometric, atmospheric and topographic in the sensor data. Such errors ease the quality of recorded data and in turn affect the accuracy. Hence, employing image pre-processing operation is necessary in order to construct corrected image or at least to reduce impacts of errors. One of the most common type of error seen is shadow. This is the main cause of misclassification and uncertainty in extracting land cover information. In this paper a comparison of two methods based on IOOPL and K-MEANS CLUSTERING for shadow detection, removal and reconstruction is done. Experiment shows that using K-MEANS method the shadow detection and removal is more perfect and data can be perfectly reconstructed.

Other Details:

Manuscript Id	:	IJSTEV1I10144
Published in	:	Volume : 1, Issue : 10
Publication Date	:	01/05/2015
Page(s)	:	308-311

Download Article