Varying weighted spatial quality assessment for high resolution satellite image pan-sharpening

This paper focuses on spatial quality assessment of pan-sharpened imagery that contains valuable information input images. Its aim is to show fusion functions respond differently different types landscapes. It compares a an object-level procedure with conventional pixel-level-based which assigns uniform scores all image pixels To do so, after performing series pan-sharpening evaluations, weighted for assessments products, allocating spatially varying weight factors the proportional their level content proposed. All experiments are performed using five high-resolution datasets products produced by three common algorithms. The acquired from WorldView-2, QuickBird, and IKONOS. Experimental results distortion fused images class vegetation cover exceeds man-made structures, reaching more than 4% in some cases. Our can preclude illogical fidelity estimations occurring when contain land covers. Since particular structures high importance remote sensing applications, our provides purpose-oriented estimation comparison procedures.