Fully Pipelined Low-Cost and High-Quality Color Demosaicking VLSI Design for Real-Time Video Applications
Abstract : This paper presents a fully pipelined color demosaicking design. To improve the quality of reconstructed images, a linear deviation compensation scheme was created to increase the correlation between the interpolated and neighboring pixels. Furthermore, immediately interpolated green color pixels are first to be used in hardware-oriented color demosaicking algorithms, which efficiently promoted the quality of the reconstructed image. A boundary detector and boundary mirror machine were added to improve the quality of pixels located in boundaries. In addition, a hardware sharing technique was used to reduce the hardware costs of three interpolators. The VLSI architecture in this work contains only 4.97 K gate counts and the core area is 60,229 um2 synthesized by using 0.18-um CMOS process. The operating frequency of this work is 200 MHz by consuming 4.76 mW. Compared with the previous lowcomplexity designs, this work has the benefits in terms of low cost, low power consumption, and high performance.
? The demosaicking procedure is enclosed within the camera, consequently an honest demosaicking methodology applicable for inexpensive VLSI implementation is required and in addition the hardware worth is extremely necessary.
? The worth of VLSI implementation depends upon totally on the desired memory and process quality.
? However, the resolution of camera images increased from VGA (640X480) to presently 4K extremist HD (4096X2160). the big the image, the higher the road buffering is needed.
? The desired line buffering can become the foremost very important component of the hardware value.
? Besides, another novel graph based regularization framework was presented in which a weight matrix was built to measure similarity and a static Laplacian was used to solve a variational problem.
? Whereas spatial correlation is the only estimation groundwork for everyday picture interpolation, spectral correlation between the shade channels additionally comes into play for the demosaicing problem.
? The easiest strategy to the demosaicing problem is to deal with color channels separately and fill in lacking pixels in every channel the use of a spatially invariant interpolation technique like bilinear or bicubic interpolation.
? The desired line buffering can become the foremost essential issue of the hardware value.
• The one-dimensional ‘‘ideal’’ interpolation is the multiplication with a rect function in the frequency domain and can be realized in the spatial area through a convolution with the sinc function.
• A need to keep the interpolation filter kernel space-limited to a small measurement and also extract as an awful lot data from the neighborhood as possible.
• It is found that most implementations are designed with hardware implementation in thinking paying terrific attention to the need for pipelining, system latency, and throughput per clock cycle.
• The larger the neighborhood, subjected to the increase in the pipeline, the higher the latency and per chance lesser the throughput.
? A color filter array (CFA) technique is an efficient and compact method to obtain a multispectral image both on CCD and CMOS image sensors.
? Some high-performance and high-quality color demosaicking and interpolation algorithms have been proposed.
? It utilized edge information to avoid averaging noncorrelated color differences and improved demosaicking performance successfully.
? In this paper, a novel cost-efficient and high-performance hardware-oriented color demosaicking algorithm is proposed for VLSI implementation.
? Moreover, a novel linear deviation compensation is also used to promote the performance of red-blue color interpolation.
|