Enhancing land-use classification accuracy with digital satellite data is a key challenge in land management, and data fusion algorithms can help improve the results. This study compares the performance of different data fusion algorithms in improving land-use classification accuracy. In this study, Landsat 9 data from 2024 was utilized as the most recent available information. During the fusion process, Landsat 9 multispectral data with a 30m resolution and panchromatic data with a 15m resolution were combined. After necessary preprocessing and corrections, a land-use map was prepared, including classes for orchards and agricultural lands, residential and industrial areas, mountainous and foothill areas, and clayey areas and covered plains. Then, the data fusion algorithms including Simple, Brovey, IHS, Gram–Schmidt, and ESRI were applied to the digital data. A land-use map was generated for each dataset using the Support Vector Machine algorithm. The methods' performance was then assessed and compared based on the overall accuracy and Kappa coefficient. The results indicated that the overall classification accuracy was 80.29% before applying data fusion algorithms and increased to 89.87% afterward, showing a 9.58% improvement. The performance varied among the evaluated algorithms, with the Simple method achieving the highest accuracy of 89.87% and demonstrating better results than the other data fusion techniques. Map accuracy assessed with the Kappa coefficient revealed that the Simple method also scored the highest, with a value of 0.82, followed by Brovey (0.78), Gram–Schmidt . (0.745), ESRI (0.63), and IHS (0.568). Overall, choosing an appropriate data fusion algorithm significantly influences classification accuracy, with the Simple method being the most effective for creating land-use maps from Landsat 9 data in the study area.