Covariance-based adaptive deinterlacing method using edge map

Abstract

The purpose of this article is to discuss deinterlacing results in a computationally constrained and varied environment. The proposed covariance-based adaptive deinterlacing method using edge map (CADEM) consists of two methods: the modified edge-based line averaging (MELA) method for plain regions and the covariance-based adaptive deinterlacing (CAD) method along the edges. The proposed CADEM uses the edge map of the interlaced input image for assigning the appropriate method between MELA and the modified CAD (MCAD) methods. We first introduce the MCAD method. The principle idea of the MCAD is based on the correspondence between the high-resolution covariance and the low-resolution covariance. The MCAD estimates the local covariance coefficients from an interlaced image using Wiener filtering theory and then uses these optimal minimum mean squared error interpolation coefficients to obtain a deinterlaced image. However, the MCAD method, though more robust than most known methods, was not found to be very fast compared with the others. To alleviate this issue, we propose an adaptive selection approach rather than using only one MCAD algorithm. The proposed hybrid approach of switching between the MELA and MCAD is proposed to reduce the overall computational load. A reliable condition to be used for switching the schemes is established by the edge map composed of binary image. The results of computer simulations showed that the proposed methods outperformed a number of methods presented in the literature.