Author(s): Ritica Thakur, Praveen Rathod, Vivek Manekar

Linked Author(s): Praveen Rathod, Vivek Manekar

Keywords: Image classification; Generalized Likelihood Image Classification Method (GLIM); Isocluster (IC); Semi-Automatic Classification Algorithm (SACA); Kappa coefficient;

Abstract: Present methodologies of image classification give moderate accuracy due to, intermixing of the pixels and classification of the pixels into any other classes. In hydrologic modeling, curve number is mainly dependent on land use and land cover (LULC) classification and in hydraulic modeling, mostly Manning's coefficient is assigned based on the image classification. Hence, it is important to classify the geospatial image accurately. The present study focused on accuracy improvement of image classification pertaining to water resources engineering problems. Isocluster (IC), Generalized Likelihood Image Classification Method (GLIM) which were popularly used in the literature were employed and compared with Index based image classification method developed in the present study using Semi-Automatic Classification Algorithm (SACA). In this study Level I classification viz. Forest, Water, Urban, Barren, and Agricultural land was carried out using these image classification methods whereas, accuracy assessment of the image classification was carried out by using the indicators-Commission Error, Omission Error, Producer accuracy, User accuracy, Kappa coefficient, and Overall accuracy. While comparing these accuracy indicators for reported classification methods, it reveals that the Semi-Automatic Classification Algorithm (SACA) based method indicated the improved performance of image classification. Hence, this study recommended the use of SACA method for image classification.

DOI: https://doi.org/10.3850/38WC092019-6661

Year: 2019