Hardware Implementation Image Processing Algorithms on FPGA
Importance of Image and its analysis:
Today Digital Image Processing has a variety of applications such as remote sensing, automated industrial inspection, medical technology for X-Ray enhancement, surveillance and many more. As the image sizes become larger, software tools have become inefficient to analyze large amount of data. As we look for solutions, Image processing comes up as a fitting reply.
Image processing: Definition:
Image processing is used to modify pictures for improvisation purposes, such as image enhancement, contrast, brightness as well as to extract information. Besides, analysis, detection and changing the original structure like composition, image editing are also possible using as image enhancement.
Advantage of Field Programmable Gate Array (FPGA):
There are two types of technologies available for hardware design-
- Full custom hardware design– It is also known as Application Specific Integrated Circuits (ASIC) and semi custom hardware device, which comprises programmable devices like Digital signal processors (DSPs) and Field Programmable Gate Arrays (FPGA’s). Full custom ASIC design offers high speed and lower power dissipation, but the complexity and the cost associated with the design are very high. The ASIC design is not reprogrammable and is time consuming. ASIC designs are used in high volume commercial applications.
- 2. Digital signal processor- DSPs are a class of hardware devices that fall somewhere between an ASIC and a PC in terms of the performance and design complexity. DSPs are specialized for microprocessors, typically programmed in C, or with assembly code for improved performance. It is well suited to extremely complex math intensive tasks such as image processing. Knowledge of hardware design is still required, but the learning curve is much lower than the other design choices. Field Programmable Gate Arrays are reconfigurable devices.
Hardware implementation of Image processing algorithms:
Images can be processed by optical, photographic, and electronic means, but image processing using digital computers is the most common method. It is mainly because digital methods are fast, flexible, and accurate. Software based image processing algorithms have many limitations and are hence unable to offer desired results. This is a lacuna that Hardware based image processing can rectify.
Hardware implementation is one of the most important tools for real time image processing application. Hardware based image processing is very fast and provides accurate analysis compared to software implementation. Moreover, Hardware design techniques such as parallelism and pipelining techniques can be developed on FPGA, which is not possible in dedicated DSP designs.