This technology is a deinterlacing device and method that performs initial interpolation of missing pixels using a Wiener filter with six vertical pixels, evaluates the difference between the interpolated value and surrounding pixel values using a fuzzy metric to determine pixel-specific weights, and updates the interpolated values using these weights.
In environments ranging from mobile devices to large displays that require different scan formats, displaying interlaced video directly often results in visual artifacts such as line crawling, field aliasing, and inter-line flicker. High-performance deinterlacing techniques designed to reduce these issues have historically been computationally intensive, making it difficult to achieve both high processing speeds and high image quality.
This technology utilizes a two-stage structure that refines simple linear filter results once using a fuzzy metric-based weighted average, allowing for improved image quality and faster deinterlacing results with minimal computational resources. It can be applied to real-time streaming services and mobile video playback devices, providing seamless progressive scan video even on devices with limited power and processing performance.
N/A
