Adaptive General Scale Interpolation Based on Weighted Autoregressive Models
The autoregressive (AR) model has been widely used in signal processing for its effective estimation, especially in image processing. Several dedicated two× interpolation algorithms adopt the AR model to explain the strong correlation between low-resolution (LR) pixels and high-resolution (HR) pixels. However, these AR model-based methods closely rely on the mounted relative position between LR pixels and HR pixels that are nonexistent in the final scale interpolation. During this paper, we tend to gift an adaptive general scale interpolation algorithm that is capable of arbitrary scaling factors considering the nonstationarity of natural pictures. Different from different dedicated two× interpolation methods, the proposed AR terms are modeled by pixels with their adjacent unknown HR neighbors. To atone for the knowledge loss caused by mismatches of AR models, we think about a weighting scheme suitable for general scale things based on the pixel similarity to increase accuracy of the estimation. Comprehensive experiments demonstrate the effectiveness of the proposed technique on general scaling factors. The most gain of peak signal-to-noise ratio is two.07 dB compared with phase adaptive gradient angle in 1.five× enlargements. To evaluate the performance in resolution adaptive video coding, we have a tendency to have additionally tested our method on Joint Scalable Video Model codec and obtained higher subjective quality and rate-distortion performance.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here