Image Processing Reference
In-Depth Information
a CA local rule [55]. CA can also be considered as a Very-Large-Scale Integration
(VLSI) architecture. Many special computing machines have been developed based
on the CA architecture. Computational structures for VLSI realization, which have
been developed based on CA architecture, are more suitable in terms of circuit de-
sign and layout, ease of mask generation, silicon-area utilization, and maximization
of clock speed [54, 60]. Applications include image or video processing in several
cases e.g. in [4, 25].
The VLSI technology is now a mature integrated circuit technology, that has
made possible the development of smaller, faster, and cheaper special-purpose dig-
ital processing devices. This technology also contributed in the construction of dig-
ital systems, capable of performing the complex digital signal processing tasks that
were usually too difficult or too expensive to be performed by analog circuitry.
Today, many of the functions usually performed by analog means in the past, are
realized by less expensive and more reliable digital hardware. VLSI design and im-
plementation played a crucial role in turning theory into reality. Therefore, both
signal processing and VLSI technologies were developed in parallel during the last
decades [47, 48, 57, 63].
Hardware implementations of image and video processing algorithms by using
VLSI technology, have been extensively used in the past and have been proven as
one of the most viable solution for improving the performance of image and video
processing systems. In such systems, non-linear image processing filters and math-
ematical morphology operators are usually used as parts of more sophisticated al-
gorithms. Some examples of successful implementations are: a separable 2-D FIR
filter for finite input matrix which has been implemented using VLSI [37]. A uni-
versal filter architecture that provides a multi-functional image processing solution
for applications that require the output of nonlinear filters, such as edge preservation
smoothing, noise filtering, and image segmentation [15]. A VLSI filter architecture
for video noise reduction based on a nonlinear rational operator filter [49] and an
implementation using Application Specific Integrated Circuits (ASICs) suitable for
performing non-linear image processing operations [11].
Nowadays, electronic device sizes continue to shrink deeper into the nanometer
regime but physical limitations of conventional electronics including power con-
sumption, interconnect and lithography have become increasingly difficult to over-
come. It is soon expected that nanoelectronic circuits will gradually replace the
CMOS circuits. There are several emerging nanoelectronic technologies, which
have been proposed by researchers during the last years. Carbon nanotubes, sili-
con nanowires, Quantum-dot Cellular Automata, single-electron transistors (SETs)
and circuits, resonant tunnelling diodes, and single-molecule devices are some of the
most promising emerging technologies as reported by ITRS [19]. The QCA com-
putational scheme uses highly pipelined architectures and extremely high speeds. It
does not use electric current flow to codify the information, but the positions of elec-
trical charges in the interior of the QCA cells. An efficient design of circuits based
on QCA technology would lead to the reduction of computational complexity and
power consumption. Therefore, QCA is considered as one of the most promising
emerging nanoelectronic technologies [28, 29, 44, 62].
Search WWH ::
Custom Search