Piestova I. Technique for vegetation condition estimation in urban area using multispectral satellite imagery

The thesis is devoted to development of theoretical and methodological foundations of computer-aided estimation of vegetation condition. The analysis of modern methods for ground-based vegetation survey demonstrates ones complexity and low efficiency in urban area studies. The paper formulates the basic requirements for the multispectral space imagery and justifies the necessary operations for data preprocessing. The theoretical background for estimating the vegetation condition by multispectral satellite imagery is presented. The mandatory ground-based measurements must be performed for this purpose according to special procedure described. All necessary ground-truth measurement was carried out in statistically significant amount. The method for the vegetation amount estimation on the basis of robust nonlinear regression between vegetation parameters and remote sensed indices was developed. All necessary algorithms for remote analysis of vegetation condition are developed. All algorithms are summarized into complete technique for vegetation condition estimation in urban area using multispectral satellite imagery.