The images have significant overlapping geographic area
The images must have the same spatial resolution
The images must belong to the same dates
Show Solution
The Correct Option isA, B
Solution and Explanation
Step 1: Recall the purpose of registration. The goal of image-to-image registration is purely geometric: to establish a pixel-to-pixel spatial correspondence between two images of overlapping ground coverage using a coordinate transformation estimated from matched tie points, followed by resampling of one image onto the grid of the other.
Step 2: Test each statement against this definition. Images may belong to different dates: since registration only needs common stable ground features (roads, field boundaries, buildings) to serve as tie points, multi-date registration is standard, and is exactly the pre-processing step before change detection. TRUE. Images have significant overlapping geographic area: tie points can only be selected where both images actually show the same ground location, so overlap is a mathematical necessity for solving the transformation equations. TRUE. Images must have the same spatial resolution: the resampling step is specifically designed to reconcile different pixel sizes between a reference and target image, so equal resolution is not a precondition. FALSE. Images must belong to the same dates: this directly contradicts the common multi-temporal registration workflow, so it is FALSE.
Step 3: Cross-check with a real workflow. Registering a 2020 Landsat scene to a 2023 Landsat scene of the same city to measure urban growth: different dates (confirms A), large common footprint needed for tie points (confirms B), and the scenes can even come from sensors of different resolution (confirms C is not required).
Step 4: Final answer. Statements (A) and (B) hold in general; (C) and (D) impose unnecessary or incorrect constraints. \[ \boxed{(A) \text{ and } (B) \text{ are correct}} \]
