多相同最小能量值线路下基于线裁剪的图像缩放算法研究

The Study on an Image Resizing Algorithm Based on Seam Carving with Lots of Same Minimum Energy Value Paths

针对线裁剪图像缩放算法中可能同时存在多条能量值最小的线路且当其数量多于待提取的线路时,选择不同的提取路线将产生不一样的缩放效果的情况,提出一种改进的能量值线路提取方法,通过增加约束条件计算具有相同最小能量值线路的位置偏移差,在能量值相等的情况下优先提取位置偏移差最小的线路。实验结果证明,改进的算法提取的线路定位更为合理,视觉关注重点区域出现扭曲变形的问题得到改善,可较好地应用于各种风格属性的图像缩放,特别是简笔画,能获得更好的裁剪结果。

To solve the problem that there are lots of optimum seams with same minimum energy values and its amount is more than to be extracted,and by choosing different extracted seams will lead to different resizing results,an improved adaptive image resizing algorithm has been proposed based on Seam Carving.The improve algorithm calculates the position offset of seams and select the minimum one.The experimental results show that the seams extracted by the improved algorithm are more reasonable,and it has fewer aliasing problems in the details of image's key objects.It is better applying to all kinds of images resizing,and it has a better effect especially in the seam carving of simple stroke drawing.