Aiming at the problem that traditional video image interpolation algorithm is difficult to achieve high definition when dealing with complex and texture-rich images

and the output image is prone to aliasing and blurring effects

a video image interpolation algorithm is proposed; it utilizes inter-frame and intra-frame spatial redundancy information

which searches for non-local regions to obtain neighbor patches

and reconstructs high-resolution（HR） by collaging low-resolution（LR） patches and linear combinations of neighbor patches on different sampling grids.The experimental results show that the image reconstruction accuracy of the algorithm is high

and the calculation of the whole algorithm is small due to the simple structure of model.