摘要
通过 2幅已知的灰度图像 ,用数学形态学的方法确定出插值图像的轮廓 .通过求出轮廓内任意插值点与中心点的位置关系 ,分别找出其在 2幅源图像上的对应点 .如果 2个对应点的灰度值较为接近 ,可通过线性插值直接求出此插值点的灰度 ;如 2个对应点的灰度值相差较大 ,则判断插值点的相对位置与哪个对应点的相对位置接近 ,并认定插值点的灰度值即为此对应点的灰度值 .对轮廓内所有的点采用上述算法 ,即可求得最终的插值图像 .算法将图像的形状信息和灰度信息较好地结合起来 .
Mathematical morphology is used to acquire the contour of the interpolated image. Then, to each point in this contour, the corresponding points in both original images are found. If the grey value of both of the corresponding points are close to each other, linear interpolation is used to calculate the grey value of the interpolated point. But if the two corresponding points have distinct grey value, a method to calculate the interpolated point's grey value is presented. Once all point's grey value are acquired, the final interpolated image is accomplished. The experimental results show that this new method is effective.
出处
《东南大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2003年第2期222-225,共4页
Journal of Southeast University：Natural Science Edition
