Image registration is used to detect changes in lakes by using large volumes of remotely sensed images. Alaska is c overed by ice

snow

and lakes all year

which make it difficult to find valid Ground Control Points（ GCPs） using remotely sensed images. This adds to the difficulty of image registration used in lake change detection and analysis. Based on lake change regulation

we propose that the deepest point of a lake is the most stable point during changes in the shape of that lake

and that this point can be used as a GCP in image registration. The center point of a polygon has been applied to many fields. Compared with the centroid of a polygon

the center point of the Largest Inner Circle（ LIC） of a polygon is always in the interior of the polygon

and the distance from this point to all the edges of the polygon remains the furthest

which meets the requirements of numerous applications. The center point of a polygon can be computed using the Voronoi diagram of the polygon. After analyzing the Voronoi d iagram computing procedure for a simple polygon

this paper presents the Voronoi method of generating a complex polygon

and then analyzes the algorithm complexity. We use the proposed algorithm to seek the LIC of many lakes in Alaska. Results show that the proposed algorithm can compute the center points of all the LICs of the lakes with high efficiency and achieve perfect registration effect based on these points.