Você está na página 1de 12

IMAGE PROCESSING USING FPGA

AUTHORS

S.Elanchezhiyan Chezhiyan741992@gmail.com B.E. III year ECE ARUNAI ENGINEERING COLLEGE THIRUVANNAMALAI

S.Mehanathan connectmehanathan@live.in

AGENDA INTRODUCTION EXISTING ARCHITECTURES DSPs ROLE IN IMAGE PROCESSING WHY FPGA? FPGA A BYPASS TO DSP APPLICATIONS OF FPGA Radiographic imaging Molecular imaging Magnetic Resonance Imaging (MRI) Video imaging CONTRAST BETWEEN FPGA AND DSP

CONCLUSION

ABSTRACT The field of digital image processing has experienced continuous and significant expansion in recent years. The usefulness of this technology is apparent in many different disciplines covering medicine through remote sensing. The advances and wide availability of image processing hardware has further enhanced the usefulness of image processing. In this paper we look in the trends of video and image processing that are forcing developers to re-examine the architectures they have used in the past. This paper will now discuss the emerging architecture FPGA (Field Programmable Gate Array) which has significant advantage over DSPs in image processing. An FPGA is an uncommitted sea of gates programmed by connecting large number of gates together to form multipliers, adders, registers and so forth. DSP has to work hard to transfer the data without any loss. On the other hand, FPGA dedicates logic for receiving data, so can maintain high rates of I/O. There are many high sampling rate applications that an FPGA does easily. FPGA are Using the Xilinx Core Generator this can be done at a block-diagram level. Many blocks can be very high level

ranging from a single gate to an FIR or FFT. FPGAs are widely used for video and image processing applications, such as broadcast infrastructure, medical imaging. HD videoconferencing, video surveillance, and military imaging. Video and image processing solutions for FPGAs include optimized development tools and kits, reference designs, video compression IP, and interface and system IP, as well as video and image processing IP suite. These solutions can improve cost, performance, and productivity for many video and imaging applications. INTRODUCTION Image Processing is a technique to enhance raw images received from cameras/sensors placed on satellites, space probes and aircrafts or pictures taken in normal day to day life for various applications. Various techniques have been developed in Image processing during the last four to five decades. Most of the techniques are developed for enhancing images obtained from unmanned space crafts, space probes and military reconnaissance flights. Image processing systems are becoming popular due to easy availability of powerful personnel computers, large size memory devices, graphic softwares etc.

The common steps in image processing are image scanning, storing, enhancing and interpretation. The schematic diagram of image scanner-digitizer is shown in figure below,

has grown from 5000 gates to over 100 million. Modern ASICs often include entire 32-bit processors, memory blocks including ROM, RAM, EEPROM, FLASH and other large building blocks. Such an ASIC is often termed as SoC (system-ona-chip). Designers of digital ASICs use a hardware description language (HDL), such as Verilog or VHDL, to describe the functionality of ASICs.

DSPs ROLE IN IMAGE PROCESSING: Digital signal processing (DSP) or media processors has advantage of ultimate choice depending on end equipment requirements and solution availability. It has the characteristics such as high performance, flexibility, easy upgradability, low development cost, and a migration path to lower cost as the application matures. The DSP is a specialised microprocessor typically programmed in c, perhaps with assembly code for performance. It is well suited to extremely complex maths-intensive tasks, with conditional processing. It is limited in performance by the clock rate, and the number of useful operations it can do per clock. As an example, a TMS320C6201 has two multipliers and a 200MHz clock, so can achieve 400M multiplies per second. A Typical DSP processor block diagram is given below,

There are two methods of image processing 1. Analog image processing 2. Digital image processing The Principle advantage of Digital Signal Processing over Analog Signal Processing is its versatility, repeatability and preservation of original data precision. EXISTING ARCHITECTURES An application-specific integrated circuit (ASIC) is an integrated circuit (IC) customized for a particular use, rather than intended for general-purpose use. For example, a chip designed solely to run a cell phone is an ASIC. Applicationspecific standard products (ASSPs) are intermediate between ASICs and industry standard integrated circuits like 7400 or the 4000 series. As feature sizes have shrunk and design tools improved over the years, the maximum complexity possible in an ASIC

Most general-purpose microprocessors and operating systems can execute DSP algorithms successfully, but are not suitable for using portable devices such as mobile phones and PDAs because of power supply and space constraints. A specialized digital signal processor, however, will tend to provide a lower cost solution, with better performance, lower latency, and, no requirements for specialized coolers or large batteries.

Modern signal processors yield greater performance. This is due in part to both technological and architectural advancements like lower design rules, Fast access to level cache, DMA circuitry and a wider bus system. Generally, DSPs are dedicated integrated circuits; however DSP functionality can also be produced by using Field Programmable Gate Array [FPGA]. WHY FPGA? An FPGA is an uncommitted sea of gates. The device is programmed by connecting the gates together to form multipliers, registers, adders and so forth. Using the Xilinx core generator this can be done at a block diagram level. Many blocks can be very high level ranging from a single gate to an FIR or FFT. Their performance is limited by the number of gates they have and the clock rate. FPGAs have the following characteristics that make them very appealing for video and image processing architectures: High performance: HD processing can be implemented in a single FPGA. Flexibility: FPGAs provide the ability to upgrade architectures quickly to meet evolving requirements, while scalability allows use of FPGAs in low cost and high performance systems. Low development cost: Video development kits from very low cost to include the software tools required to develop a video system using FPGAs.

The architecture of a digital signal processor is optimized specifically for digital signal processing. Most also support some of the features as an applications processor or microcontroller, since signal processing is rarely the only task of a system.

Video and Image Processing Solution: This includes optimized DSP design flows, Image processing suite, and interface and third party video compression IP, and video reference designs. In Addition, Adaptive filters which are the heart of Active Vibrato Control System are traditionally implemented using DSP processors due to their ability to perform fast floating-point arithmetic. General purpose DSPs are optimized for common signal processing operations. DSP systems based on softwares are flexible, but due to sequential nature of microprocessors, suffer from insufficient processing capability. They are the fastest method to get an algorithm running because they offer a comprehensive development environment, with tools for code analysis, debugging, and rapid prototyping. Disadvantages of DSPs is that ultimately they execute instructions serially, setting an upper limit on the chips throughput. Fastest DSP caters for billion MACs (multiply-accumulate/sec). On the other hand, FPGAs are Reconfigurable hardware devices where in it is possible to change the topology of electronic circuits at runtime. FPGAs incorporate arrays of dedicated multipliers, embedded memory and high-speed I/O that make them ideal for DSP applications. Also in recent times due to its growing die size as well as incorporating the embedded DSP block, the FPGA devices have become a serious role in the signal processing field. Although it is not yet feasible to use floating-point arithmetic in modern FPGAs, it is sufficient to use fixed-point arithmetic and still achieve tap-weight convergence for adaptive filters.

In transversal adaptive filters, each tap, as well as components for updating each filter coefficient, requires a multiplier and an adder. By instantiating the required number of multipliers and an adder, the performance of FPGA based filter can increase significantly compared to DSP processors. FPGAs allow great deal of parallelism to be incorporated. Also the new generation of FPGAs like virtex-4 from Xilinx has embedded DSP slices with in the device, which have dedicated circuitry to perform additions and multiplications, which are fundamental for any DSP applications.

With the following areas of application, we may now discuss how FPGA is advantageous over DSP. When sample rates grow above a few MHz, a DSP has to work very hard to transfer the data without any loss. This is because the processor must use shared resources like memory busses, or even the processor core which can be prevented

from taking interrupts for some time. An FPGA on the other hand dedicates logic for receiving the data, so can maintain, so can maintain high rates of I/O. A DSP is optimised for use of external memory, so a large data set can be used in the processing. FPGAs have a limited amount of internal storage so need to operate on smaller data sets. However FPGA modules with external memory can be used to eliminate this restriction. A DSP is designed to offer simple re-use of the processing units, for example a multiplier used for calculating an FIR can be re-used by another routine that calculates FFTs. This is much more difficult to achieve in an FPGA, but in general there will be more multipliers available in the FPGA. The DSP can take a standard c program and run it. This c code can have a high level of branching and decision making. For example, the protocol stacks of communication systems. This is easy to implement in FPGA. Most signal processing systems start life as a block diagram of some sort. Actually translating the block diagram to the FPGA may will be simpler than converting it to c code for the DSP.

filter), operating at a sample rate of 50-100MHz. At these rates DSP processor would find it extremely difficult to do anything. It is now FPGA implementing it in a simple structure where a sample rate of 100MHz is achievable and even the smallest FPGA will have a lot of resource left for further processing.

2. Digital radio receiver baseband processing. Some receiver types would require FFTs for signal acquisition, and then matched filters once a signal is acquired. Both blocks can be easily implemented by either approach. However, there is a mode change from signal acquisition to signal reception. It may be well suited for DSP, as the FPGA would need to implement both blocks simultaneously. Note that the RF processing is better in an FPGA, so this likely to be a mixed system.

FPGA A BYPASS TO DSP We may now discuss some examples to elucidate how FPGA bypasses DSP in its application. 1. First decimation filters in digital wireless receiver. Typically a CIC filter (Cascaded Integrator Comb

3. Image Processing Here, most of the operations on an image are simple and very repetitive when best implemented in an FPGA. However, an imaging pipeline is often used to identify Regions of Interest in an object being inspected. These Blobs can be of varying sizes, and subsequent processing tends to be more complex. The algorithms used are often adaptive, depending on what the blob turns out to be So a

FPGA-based approach may be better for the back end of the imaging pipeline.

medical imaging, but rather are a part of pathology. As a discipline and it its widest sense, it is a part of biological imaging and incorporates radiology, nuclear medicine, investigative radiological sciences, endoscopy, thermograph, medical photography and microscopy. Several factors should be considered in the efficient development of flexible medical imaging equipment: Development of imaging algorithms requires high-level intuitive modelling tools for continual improvements in digital signal processing (DSP). The performance needs for near real time analysis require system plotforms that scale with both software (CPUs) and hardware (configurable logic).these processing platforms must meet various performance price points and be capable of bridging the fusion of imaging modalities. System architects and design engineers need to quickly partition and deug algorithms on these platforms, using the latest tools and intellectual property (IP) libraries to speed their deployment and improve profitability. Distributed vector processing is an algorithm that enables faster computations. The S-transforms, revealing frequency variation over both space and time. Applications include texture analysis and noise filtering. However, ST is computationally intensive, making conventional CPU implementation too

APPLICATIONS OF FPGA There are sea full of applications in implementing FPGA for its simplicity and effectiveness. Image processing is used in various applications such as: 1. Remote Sensing 2. Medical Imaging 3. Non-destructive Evaluation 4. Textiles 5. Forensic studies 6. Material Science 7. Military 8. Film industry 9. Document processing 10. Graphic arts 11. Printing industry

Lets now discuss on a medical imaging using FPGA in detail. MEDICAL IMAGING Medical imaging is the technique and process used to create images of the human body for clinical purposes or medical science. Although imaging of removed organs and tissues can be performed for medical reasons, such procedures are not usually referred to as

slow. This problem is addressed by combing vector processor in conjunction with parallel computations implemented in FPGAs. The typical block diagram of a general medical system is given below:

Now, we will discuss some of the trends and key developments driving the integration of programmable logic into medical image equipment. RADIOGRAPHIC IMAGING: The conversion from computed radiology (DR) is well underway. The benefits of digital flat-panel detectors allow images to be processed within seconds, thus increasing productivity, while eliminating the handling of cassettes and associated chemicals for film processing. In addition, portable DR systems are increasing in acceptance. Many of these smaller profile systems offer tether less flat-panel detectors using wireless connectivity removing the cables that are failure points and sources of workplace injuries. FPGAs can support the changing protocol stacks from wired to wireless and manage the power sequencing and monitoring in these digital detectors.

MOLECULAR IMAGING: X-ray or CT scans provide images of anatomical structures; molecular imaging is well suited for images of cellular functions and metabolic processes. Two commonly used modalities are singlephoton-emission computed tomography (SPECT) and PET platforms. Advances in PET include increased detector counts; expanded detector pairing areas, shortly coincidence windows, time-

of-flight technology, and more advanced algorithmic modelling. Significant increases in raw data require faster processing capabilities and pre-processing decisions earlier along the data path. In many systems, FPGAs are used in the aggregation stages of the electrical signals from the photo sensors and data converters. The intrinsic programmable nature of FPGAs allows newer designs to increase pre-processing decisions along the data path, thus reducing the amount of data delivered to the reconstruction stage.

MAGNETIC RESONANCE IMAGING (MRI): Magnetic resonance imaging (MRI) reconstruction creates images crosssectional images of the human body. First 2D slice reconstructions are developed using Fast Fourier Transform (FFT) functions of the frequency domain data. Sharpening and spatial de-blurring techniques are employed, reducing noise while sharpening the image. Many of these are building blocks within the VIP and DSP libraries available for implementation in FPGAs. From this, 3D volumes reconstructions are generated from interpolation techniques.

more commonly used. These techniques involve simultaneous acquisition and registration of optic-acoustic and ultrasound images, segmentation of vascular and skeletal images, and the use of correlation-based enhancement filter to reduce false positives when diagnosing lung disorders. Additionally, cardiac motion estimation constitutes aid in quantifying the elasticity and contractibility of the heart muscle. Localised areas exhibiting abnormal movements are indicative of schematic heart regions, where insufficient circulation exits. A developing algorithm includes the quantitative evaluation of elasticity from a series of ultrasound images using spatial-temporal registration techniques to find the deformation field with respect to reference frame. Key VIP and DSP building block function applied here include 2D filtering, despeckling, correlation, and smoothing.

VIDEO IMAGING: In vascular imaging, the gold standard has been radiopaque contrast angiography from X-ray images of blood vessels, using salt-based contrast agents. Today, CT angiography, time-offight/phase-contrast MR angiography, and duplex/intravascular ultra-son0graphy are

performance when compared to pure DSP processor base implementations. FPGA and DSP represent two very different approaches to signal processingeach good at different things. There are many high sampling rate applications that an FPGA does easily, while the DSP could not. Equally, there are many complex software problems that the FPGA cannot address. As a result, the ideal system is often to split the work between FPGAs and DSPs. This is easily accomplished using HERON system, and in many cases can be done simply using I/O and processor modules without any dedicated FPGA resource. CONCLUSION When developing innovative communication systems, the quicker we have implementation of video and image processing. Now research projects are concentrating on powerful algorithms that can efficiently process the images. A successful area developing for hypothetical algorithm are DSPs and FPGAs. There contrast with reference to their own advantages and disadvantages, an HERON system can be developed for implementing both on a single processor. With the developing architecture, FPGAs are great fits for video and image processing applications, such as broadcast infrastructure, medical imaging. HD videoconferencing, video surveillance, and military imaging. Video and image processing solutions for FPGAs include optimized development tools and kits, reference designs, video compression IP, and interface and system IP, as well as video and image processing IP suite. These solutions can improve cost, performance, and productivity for many video and imaging applications in future. REFERENCES: Alteras Video and Image Processing Solutions website:

CONTRAST BETWEEN FPGA AND DSP FPGA devices provide a reconfigurable DSP solutions for various DSP applications. FPGA devices incorporate a variety of embedded features such as embedded processors, DSP blocks, and Memory blocks. These device features provide very high DSP capability in FPGAs compared to DSP processors. Using FPGAs, DSP designers can customize their hardware for optimal implementation of their applications. Using embedded processors such as Nios embedded processor; FPGAs also offer a software-based design flow similar to the traditional DSP software design flow. In addition, FPGAs offer a design flow using the SoPC builder tool that enables a software design flow to be combined with hardware acceleration. Using this design flow, a DSP designer can implement a complete DSP system in an FPGA and thereby develop a cost-effective, highperformance DSP system. Altera FPGAs also offer a design flow based on the DSP builder tool that allows a DSP system engineer designing in MATLAB/Simulink to implement a system on an FPGA without learning HDL. The DSP builder tool flow can also be combined with the SoPC builder tool flow for implementing a complete DSP system. Hence FPGAs from Altera provide the benefits of system integration, flexibility while partitioning the system, lower system costs, and higher

www.altera.com\video_imaging Alteras Video and Image Processing Reference Design www.altera.com/endmarkets/refdesigns/syssol/broadcast/ref-postprocessing.html Fundamentals Of Digital Image Processing - Anil K. Jain, PrenticeHall, 1989. Digital Image Processing Chellappa, 2ndEdition, IEEE Computer Society Press, 1992. Digital Image Processing - Woods, Addison Wesley, 1992.

Você também pode gostar