Video de-interlacing algorithm based on motion adaptive spatial-temporal technique
Shih‐Chang Hsia, Ming-Kun Cheng · Journal of Electronic Imaging · 2013
Television (TV) systems use an interlaced scanning method because transmission bandwidth is limited. Although technological developments can solve the problem of bandwidth, the interlaced scanning approach is still used in video broadcasting due to the high cost of updating equipment. However, the current LCD displays employ progressive scanning screens. The de-interlacing technique that converts the interlaced signal to a progressive signal is required for a TV receiver to create the display. In this study, adaptive spatial and temporal processing is proposed in implementing a high-performance de-interlace algorithm. First, the spatial de-interlace method is proposed for the conditional interpolation using edge detection to improve the quality. Then the motion adaptive de-interlace technique is proposed based on motion detection with multifield references. With this method, we can differentiate between the moving states of the current field using forward or backward prediction. Then the missing lines can be interpolated based on motion features from the adaptation of spatial and temporal pixels. Results demonstrate that the proposed algorithm can achieve a better tradeoff between complexity and performance compared with the competing methods.