پیوندهای مفید
اخبار و اعلانات
آمار
| تعداد نشریات | 127 |
| تعداد شمارهها | 7,237 |
| تعداد مقالات | 77,625 |
| تعداد مشاهده مقاله | 161,094,071 |
| تعداد دریافت فایل اصل مقاله | 120,790,287 |
Pan-sharpening Solutions for Khayyam Satellite Imagery: A Comparative Study on Pixel-Level Image Fusion Approaches | ||
| Earth Observation and Geomatics Engineering | ||
| دوره 10، شماره 1، مهر 2026، صفحه 29-40 اصل مقاله (1.19 M) | ||
| نوع مقاله: Original Article | ||
| شناسه دیجیتال (DOI): 10.22059/eoge.2026.410000.1212 | ||
| نویسندگان | ||
| Farhad Samadzadegan1؛ Ahmad Toosi* 1؛ Hadiseh Hasani2؛ Omran Amini3 | ||
| 1School of Surveying and Geospatial Engineering, University College of Engineering, University of Tehran, Tehran, Iran | ||
| 2Department of Geodesy and Surveying Engineering, Tafresh University, Tafresh, Iran | ||
| 3Iranian Space Research Center (ISRC), Tehran, Iran | ||
| چکیده | ||
| Objective: This study presents the first comprehensive evaluation of pansharpening methods on Iran's Khayyam satellite imagery, launched in August 2022. Method: Twenty-one algorithms across traditional methods (including component substitution, multi-resolution analysis, and hybrid approaches) and advanced deep learning networks were compared using Khayyam data from Yazd-Iran, Semnan-Iran, and Baotou-China, with 0.8-meter panchromatic and 3.2-meter multispectral resolution. Traditional methods included 18 algorithms such as PCA, Gram-Schmidt variants, MTF-GLP methods, wavelet-based approaches, and hybrid combinations of wavelet and Gram-Schmidt techniques. Due to unique spectral characteristics, three deep learning networks (PNN, PanNet, PSGAN) were trained on Khayyam data. Quality assessment was conducted using visual analysis and quantitative metrics, including SAM, ERGAS, SSIM, PSNR, and UIQI for spectral quality, and SCC for spatial quality. Results: Results showed that traditional methods like GS-Wavelet and PRACS achieved superior performance with PSNR values exceeding 76 dB and SSIM above 0.99. Deep learning methods showed limited performance despite sensor-specific training, requiring extensive training data specific to Khayyam's sensor characteristics. Conclusions: No single method excelled universally, revealing trade-offs between spatial enhancement and spectral preservation. Baseline performance metrics for Khayyam imagery pansharpening were established, and insights for optimal method selection were provided. | ||
| کلیدواژهها | ||
| Remote sensing؛ Khayyam Satellite؛ Pan-sharpening؛ Spatial resolution؛ Deep learning | ||
|
آمار تعداد مشاهده مقاله: 30 تعداد دریافت فایل اصل مقاله: 9 |
||
سامانه مدیریت نشریات علمی. قدرت گرفته از سیناوب