A two-layer constraint image interpolation model based on both local and global information

作者：Wu Jiying^*; Ruan Qiuqi

来源：Chinese Journal of Electronics, 2007, 16(3): 509-514.

• image interpolation
• local information
• global information
• two-layer constraint

In this paper, a novel two-layer constraint image interpolation model is proposed. The new model can process natural image, texture image and noisy image. It interpolates image under the constraint of area-sample in original un-interpolated image and based on both local and global information. Natural image processed by the new model is smooth and has clear edge. The edge pixel is processed by PDE (Partial differential equation) which enhances edge and smoothes jaggies. Pixel in smooth regions of un-interpolated image is processed by NL (Non local)means method which gets pixel's gray value by weighted averaging global information and image is smoothed. Texture image interpolated by the new model preserves entire texture pattern; jaggies in linear structure is smoothed. It is even proved noisy image can be interpolated directly to the required size using this new model. Both theoretical analysis and experiments have verified the benefits of the new interpolation model.

• 出版日期2007-7
• 单位北京交通大学