scholarly journals Accuracy Assessment of Land Use Land Cover Classification using Google Earth

2015 ◽  
Vol 4 (4) ◽  
pp. 193 ◽  
Author(s):  
Abineh Tilahun

The study examines land use land cover and change detection in Chikodi taluk, Belagavi district, Karnataka. Land use land cover plays an important role in the study of global change. Due to fast urbanization there is variation in natural resources such as water body, agriculture, wasteland land etc. These environment problems are related to land use land cover changes. And for the sustainable development it is mandatory to know the interaction of human activities with the environment and to monitor the change detection. In present study for image classification Object Based Image Analysis (OBIA) method was adapted using multi-resolution segmentation for the year 1992, 1999 and 2019 imagery and classified into four different classes such as agriculture, built-up, wasteland and water-body. Random points (200) were generated in ArcGIS environment and converted points into KML layer in order to open in Google Earth. For the accuracy assessment confusion matrix was generated and result shows that overall accuracy of land use land cover for 2019 is 83% and Kappa coefficient is 0.74 which is acceptable. These outcomes of the result can provide critical input to decision making environmental management and planning the future.


Author(s):  
Deepak Patle

Land is a limited natural resource which restricts no further increase in a cultivated area. Moreover, due to the increasing population, the pressure on this resource is increasing day by day. Hence, land use/land cover (LU/LC) information is very much necessary for the best possible use by maximizing outputs sustainably from this diminishing resource such that good planning and management can be done to meet the demand of the ever-increasing population. Therefore, a study has been conducted for land use/land cover mapping using SENTINEL-2B satellite data having a fine spatial resolution of Nahra nala watershed, which is a tributary of Wainganga river situated in Balaghat district of Madhya Pradesh, India. Five land use/land cover classes were identified, namely water bodies, agricultural land, forest, habitation (built-up), and wasteland in the study area. The study area possesses forest as the predominant LU/LC class with 83.79 percent of the total geographical area of the watershed. Accuracy assessment was also applied to the final classified results based on the ground truth points or known reference pixels along with Google Earth imageries. The overall classification accuracy of 95.52% with the kappa value of 0.92 was achieved.


Sign in / Sign up

Export Citation Format

Share Document