Evaluation of pixel-based and object-oriented classification methods in preparing land cover maps using Landsat satellite images in Idrisi Terrset software (Case study: Doviraj region of Ilam Province)
Volume 7, Issue 1, Spring 2026, Pages 209-223
Saleh Arekhi
Abstract Background and Objective: In recent years, the preparation of land cover maps through digital classification of remote sensing data has been considered as a suitable alternative for the preparation of this type of maps. Remote sensing is one of the new and efficient technologies in extracting land cover, updating maps, and detecting land cover changes. In this dirction, the aim of present study is to compare two methods of pixel-based and object-oriented classification in land cover mapping with using OLI image in Doviraj region, Ilam province using Idrisi Teerrset software.
Research Method: After the supply of related image and implementing geometric and radiometric corrections on image, we applied two classification methods to land cover mapping. To assess the accuracy of classification methods, we used of indices of overall accuracy, kappa coefficient, producer accuracy and user accuracy. The results show that the object-oriented classification method has more resolution than the pixel-based classification method.
Findings and Conclusion: The results of accuracy show that method of object-oriented in two indices of overall accuracy and kappa coefficient with (respectively) 94 percent and 0/93 percent was more accuracy then to pixel-based classification method with overall and Kappa coefficient 88 percent and 0/87 percent. The result of this study suggests that from object-oriented classification method use to production of land ucover map.
Assessing the impact of urban and rural development and land use changes on urban and rural surface temperatures (LST) using Landsat images from 2003 to 2023, case study area of Aslandoz Township
Volume 7, Issue 1, Spring 2026, Pages 380-393
Lotfollah Maleki Masoomabad, Maryam Mohammadzadeh Shishehgaean
Abstract Background and Objective: Land surface temperature is a fundamental indicator in monitoring energy balance models at the land surface at regional and global scales. Given the limitations of meteorological stations, remote sensing can be a suitable alternative for estimating land surface temperature. The main objective of this research is to monitor land surface temperature and land use changes and urban development using satellite images from 2003 and 2023. The results of the 20-year land use map showed that very visible urban development was observed throughout the entire Aslandoz Township. If this development is done with minimizing damage to the environment and forests, it will be very positive.
Methodology: In this study, Landsat satellite images from 2003 and 2023 were used to investigate and assess the feasibility of using solar radiation energy. ENVI software was used to perform calculations using ArcGIS software to prepare maps. One of the applications of thermal images is to prepare thermal maps to determine isothermal zones.
Results and Findings: The results of the LST map showed that the land surface temperature in Aslandoz Township has increased significantly, so that in 2003 the land surface temperature was 42.49, and in 2023 this figure has increased to 49.92. The results of the integration of land use maps and land surface temperature showed that the lowest temperature in both years was attributed to water and forest use, which is due to high humidity, and the highest temperature was attributed to soil and residential use. Soil use in 2003 was 42.09 and in 2023 was 49.86, and residential use in 2003 was 42.57, and in 2023 the highest value for residential use was 42.57, which is a significant figure. The reason for this very high temperature and the difference in use is that if an area is devoid of vegetation or has sparse coverage, evaporation and transpiration are low and the temperature increases.
