Rahul Despande - Copy

7/28/2019 Rahul Despande - Copy

1/27

VISVESVARAYA TECHNOLOGICAL UNIVERSITY

BELGAUM

VIDYAVARDHAKA COLLEGE OF ENGINEERING

MYSORE

SEMINAR REPORT ON

2D Analog Filters for Real Time Video Signal Processing

By

NAME : Rahul Deshpande

USN : 4VV08EC075

BRANCH : ELECTRONICS AND COMMUNICATION

SUBCODE : 06EC86

SEMESTER : VIII


2/27

VIDYAVARDHAKA COLLEGE OF ENGINEERING

MYSORE

SEMINAR REPORT ON

2D Analog Filters for Real Time Video Signal Processing

NAME OF THE CANDIDATE : Rahul Deshpande

USN : 4VV08EC075

DATE MAXIMUM

MARKS

PRESENTATION REPORT TOTAL

MARKS50

.....................................

(Signature of the student)

.................................. .................... ................................

(Dr. L. BASAVARAJ) (SHISHIRA HANUMANTAPPA) (CHETHANA K S)

HOD/GUIDE COORDINATOR GUIDE


3/27

VIDYAVARDHAKA COLLEGE OF ENGINEERING,

MYSORE

DEPARTMENT OF ELECTRONICS AND

COMMUNICATION

CERTIFICATE

Certified that the seminar entitled 2D Analog Filters for Real Time Video Signal

Processing is a bonafide work carried out by Rahul Deshpande (4VV08EC075) in

partial fulfillment for the award of degree of Bachelor of Engineering in

Vidyavardhaka College of Engineering of the Visvesvaraya Technological

University, Belgaum during the year 2012.

It is certified that all corrections/suggestions indicated for internal assessment

has been incorporated in the report deposited in the departmental library. The seminar

report has been approved as it satisfies the academic requirements in respect of

seminar report prescribed for the Bachelor of Engineering degree.


4/27

Abstract

A practical hardware design of a two-dimensional (2D) analog filter is

explained. The structure is implemented using charge coupled device (CCD) analog

shift registers and wideband operational amplifiers. The operation of the filter is

demonstrated by processing TV video images in real time. The 2D analog approach is

evaluated by comparison with a 2D distributed arithmetic digital filter. The analog

approach offers realization at lower cost, less power consumption, higher resolution,

and inherent true real-time capability independent of filter order.


5/27

CONTENTS

1. Introduction 11.1.2D Images . 11.2.2D Video ... 11.3.2D analog filters 1

2. Why analog filter? ......... 32.1.Application of analog filters ..... 4

3. Scanning Principles 63.1. Analog television. 63.2.Progressive Scanning .. 63.3. Interlaced scanning .. 7

4. Derivation of filter functions......................... 115. Practical Realization of 2D analog filters 13

5.1.Charge-coupled device . .. 135.2.Working principle of CCD 135.3.Design of 1H delay line (CCD) 145.4.Design of analog processor section .. 16

6. Filtering of images . 187. Evaluation .. 20

7.1.Advantages over digital filters . 207.2.Extension to higher order. 21References .. 22


6/27

1.Introduction1.1.2D Images

In general, a 2D image can be described as a function of two independent

spatial variables and time in the scene f(x, y, t) An image is converted to an

electrical time varying signal for transmission by the process of periodic

horizontal scanning. To prevent flickering in the display, conventional practice

has been to introduce interlaced scanning with all the even lines being scanned

first, followed by all the odd lines, producing two alternate fields of lines for each

picture frame.

1.2.2D Video.Video is the technology of electronically capturing, recording, processing,

storing, transmitting, and reconstructing a sequence of still images representing

scenes in motion.

Video technology was first developed for cathode ray tube (CRT) television

systems, but several new technologies for video display devices have since been

invented.

Frame rate, the number of still pictures per unit of time of video, ranges from

six or eight frames per second (frame/s) for old mechanical cameras to 120 or

more frames per second for new professional cameras.

1.3.2D analog filtersA hardware design for the physical realization of 2D analog filters has been

developed. Filters of this type are inherently capable of operating directly onraster scanned television images in real time. Here, real time operation means that

processing is done at the same rate as the sampling rate which could be as high as

40 million pixels/second for high definition television.

In the past analog processing techniques such as noise scoring, edge peaking,

and comb filter separation of luminance and chrominance signals have been based


7/27

on one-dimensional time domain approaches implemented as simple FIR

structures. These have been quite limited as to the type of processing and

enhancement operations that could be implemented. In contrast the 2D analog

approach is capable of realizing the general transfer function (IIR or FIR) for 2D

(spatial) filtering which makes it possible to develop filters of all types.

The use of recursive 2D analog structures to directly filter analog raster

scanned images can provide a more effective solution than digital filters. Recently

motion adaptive digital filters have been used for in high definition television

video processing. They require delays of one or more field periods which are

accomplished by means of frame-stores. since pixels in separate fields are

combined, this type of filtering is referred to as temporal and can only be

performed on those pixels for which no motion (in the scene being viewed) has

occurred between fields thus the development of this type of filter is complicated

by the inclusion of circuitry that implements the motion detection algorithm. 2D

analog filters require only line delays, analog summers, inverters and integrators.

They do not require expensive frame-stores, A/D and D/A converters and pre-

filters, or motion detection circuitry.


8/27

2.Why analog filter?When an analog filter is implemented, it is done prior to the analog-to-digital

conversion. In contrast, when a digital filter is implemented, it is done after the

conversion from analog- to-digital has occurred. Analog filtering can remove

noise superimposed on the analog signal before it reaches the Analog-to-Digital

Converter. In particular, this includes extraneous noise peaks. Digital filtering

cannot eliminate these peaks riding on the analog signal. Consequently, noise

peaks riding on signals near full scale have the potential to saturate the analog

modulator of the A/D Converter. This is true even when the average value of the

signal is within limits.

Additionally, analog filtering is more suitable for higher speed systems, i.e.,

above approximately 5kHz. In these types of systems, an analog filter can reduce

noise in the out-of-band frequency region. This, in turn, reduces fold back signals.

The task of obtaining high resolution is placed on the A/D Converter. In contrast,

a digital filter, by definition uses oversampling and averaging techniques to

reduce in band and out of band noise.

IN SPITE OF THE BEATING, there are still many applications where analog

filters should, or must, be used. This is not related to the actual performance of the

filter (i.e., what goes in and what comes out), but to the general advantages that

analog circuits have over digital techniques. The first advantage is speed: digital

is slow; analog is fast. For example, a personal computer can only filter data at

about 10,000 samples per second, using FFT convolution. Even simple op amps

can operate at 100 kHz to 1 MHz, 10 to 100 times as fast when compared to a

classical digital system.

The second inherent advantage of analog over digital is dynamic range.

This comes in two flavors. Amplitude dynamic range is the ratio between the

largest signal that can be passed through a system, and the inherent noise of the

system. For instance, a 12 bit ADC has a saturation level of 4095, and an rms

quantization noise of 0.29 digital numbers, for a dynamic range of about 14000.

In comparison, a standard op amp has a saturation voltage of about 20 volts and

an internal noise of about 2 microvolts, for a dynamic range of about ten million.


9/27

Just as before, a simple hardware for example if we consider operation amplifier

devastates the digital system.

The other flavor is frequency dynamic range. For example, it is easy to

design an op amp circuit to simultaneously handle frequencies between 0.01 Hz

and 100 kHz (seven decades). When this is tried with a digital system, thecomputer becomes swamped with data. At 200 kHz, it takes 20 million points

to capture one complete cycle at 0.01 Hz.

2.1.Application of analog filters.1. Data acquisition systems:

This application note investigates the design of analog filters that reduce theinfluence of extraneous noise in data acquisition systems. These types of systems

primarily utilize low-pass filters, digital filters or a combination of both. With the

analog low-pass filter, high frequency noise and interference can be removed

from the signal path prior to the analog-to-digital (A/D) conversion. In this

manner, the digital output code of the conversion does not contain undesirable

aliased harmonic information. In contrast, a digital filter can be utilized to reduce

in-band frequency noise by using averaging techniques.

2. Audio processing systems: Audio processing covers many diverse fields, all

involved in presenting sound to human listeners. Three areas are prominent: (1)

high fidelity music reproduction, such as in audio compact discs, (2) voice

telecommunications, another name for telephone networks, and (3) synthetic

speech, where computers generate and recognize human voice patterns. While

these applications have different goals and problems, they are linked by a

common umpire: the human ear. Digital Signal Processing has produced

revolutionary changes in these and other areas of audio processing, Human

Hearing, Timbre, Sound Quality vs. Data Rate, High Fidelity Audio,

Companding.

3. Data conversion systems.

4. Video processing.


10/27

5. Analog filter can be used to shape high speed digital PWM output.

6. Image formation and display.


11/27

3.Scanning Principles.3.1. Analog television.

Analog television is the analog transmission that involves the broadcasting of

encoded analog audio and analog video signal, in which the message conveyed by

the broadcast signal is a function of deliberate variations in the amplitude and/or

frequency of the signal. All broadcast television systems preceding digital

transmission of digital television (DTV) were systems utilizing analog signals.

Analog television may be wireless or can require copper wire used by cable

converters.

A cathode-ray tube (CRT) television displays an image by scanning a beam of

electrons across the screen in a pattern of horizontal lines known as a raster. At

the end of each line the beam returns to the start of the next line; at the end of the

last line it returns to the top of the screen. As it passes each point the intensity of

the beam is varied, varying the luminance of that point. A color television system

is identical except that an additional signal known as chrominance controls the

color of the spot.

Today, two different techniques are available to render the video: interlaced

scanning and progressive scanning. Which technique is selected will depend on

the application and purpose of the video system, and particularly whether the

system is required to capture moving objects and to allow viewing of details

within a moving image.

3.2.Progressive Scanning.Progressive scanning (alternatively referred to as noninterlaced scanning) is a

way of displaying, storing, or transmitting moving images in which all the lines ofeach frame are drawn in sequence. This is in contrast to interlaced video used in

traditional analog television systems where only the odd lines, then the even lines

of each frame (each image called a video field) are drawn alternately.

Progressive scanning, as opposed to interlaced, scans the entire picture line by

line every sixteenth of a second. In other words, captured images are not split into


12/27

separate fields like in interlaced scanning. Computer monitors do not need

interlace to show the picture on the screen. It puts them on one line at a time in

perfect order i.e. 1, 2, 3, 4, 5, 6, 7 etc. so there is virtually no "flickering" effect.

As such, in a surveillance application, it can be critical in viewing detail within a

moving image such as a person running away. However, a high quality monitor is

required to get the best out of this type of scan. The Fig 3.2.1 shows the example

of progressive scan.

Fig 3.2.1: Progressive scanning.

3.3.Interlaced scanning.This is in contrast to interlaced video used in analog television systems where

only the odd lines, then the even lines of each frame (each image called a videofield) are drawn alternately. Since we are concentrating on analog filters, we shall

study in brief about interlaced scanning.

TV calls one picture a Frame - it breaks each frame up into two interlaced

fields. Each Field is comprised of 262.5 horizontal lines which are scanned onto

the screen, left to right; each line is scanned below the previous line. There is one

odd Field (Field 1) and one even Field (Field 2). The odd field scans lines 1, 3, 5,

etc and the even field scans lines 0, 2, 4, etc as shown in Fig 3.3.1 - hence the

term interlaced. The two field's interlaced lines mesh perfectly to create one full

frame of lines 0, 1,2,3,4, etc.

There are 525 horizontal lines total in each frame, 262.5 lines per field - but

only 91% of them are visible, the scanning beam is turned off for all invisible

lines.


13/27

The final number of visible lines after cropping is approximately 480. The

Television screen displays 60 fields each second. Since each frame is divided into

2 fields (even and odd), we define the 60 fields as 30 pairs of fields, and each pair

is called a Frame. Therefore the Television screen displays 30 Frames each

second. Due to the persistence of vision, a moving image is seen.

Fig 3.3.1: Interlaced scanning, example.

Full Lines (cycles) vs Visible Lines as shown in Fig 3.2.2 where in

there is a full line, which is the same as one complete horizontal scan cycle. It

includes both the visible scanning and retrace.

Fig 3.3.2: Full Lines (cycles) vs Visible Lines

The complete horizontal scan cycle is 63.4uSec (15,750 cycles per

sec). 53uSec for the left-to-right scan, and 10uSec for the right-to-left

retrace. The common divider is 12, which breaks up the cycle into segments of

5.25uSec. Therefore the visible scan is 10/12 of one cycle and the retrace scan is

2/12 of one cycle. One-half of a visible line is 5/12 of one cycle.


14/27

Fig 3.3.3: Field 1 and Field 2.

From the Fig 3.3.3, we see that:

Field 1 has a half line at the end. After that Field 2 begins scanning.

Field 2 has a half line at the beginning, and a retrace line. After that it scans linesnormally, beginning at the left.

Field 2 begins scanning from the middle to insure that its lines fit exactly in

between the lines from Field 1.

The entire scanning may look like as shown in Fig 3.3.4 and an example

image is shown in Fig 3.3.4.

Fig 3.3.4: Interlaced scanning pattern after complete scan.


15/27

Fig 3.3.5: Interlaced scanning, example image.


16/27

4.Derivation of filter functions.Consider each field of the interlaced picture to be a separate image. Them

each line can be expressed as a function x(t,nT)of a continuous variable t, over the

horizontal scan period and a discrete variable nT which designates the nth line in

the field. In general a 2D analog filter can be represented as a liner time-invariant

system for which the filtered image signal y(t,nT) is given by the convolution of

the input signal x(t,nT) with the impulse response h(t,nT), ie. y(t,nT) = h(t,nT) *

x(t,nT) where the corresponding transfer function is given by:

H(s,z) =

=

------- (1)

Or

Y(s,z) =

where i+j0 ------- (2)

Equation (2) can be written in a line recursive form, which in turn can be

realized with analog circuitry.

A plot of an Ideal 2D filter can be seen in the Fig 4.1 and plot of practical 2D

filter can be seen in Fig 4.2.


17/27

Fig 4.1: Ideal 2D filter.

Fig 4.2: Practical 2D filter.


18/27

5.Practical Realization of 2D analog filters.5.1.Charge-coupled device.

A charge-coupled device (CCD) is a device for the movement of electric

charges between capacitors. This is achieved by a shift signal which uses an

electric field for "shifting" the signals between capacitive stages within the device

one at a time.

The CCD is a major technology in some digital imaging sensors to move the

light energy related charge within the device to an area where the charge can be

manipulated, for example conversion into a digital value. In a CCD image sensor,

pixels are represented by p-doped MOS capacitors. These capacitors are biased

above the threshold for inversion when image acquisition begins, allowing the

conversion of incoming photons into electron charges at the semiconductor-oxide

interface; the CCD is then used to read out these charges.

5.2.Working principle of CCDIn a CCD for capturing images, there is a photoactive region (an epitaxial

layer of silicon), and a transmission region made out of a shift register (the CCD).

An image is projected through a lens onto the capacitor array (the photoactive

region), causing each capacitor to accumulate an electric charge proportional to

the light intensity at that location. A one-dimensional array, used in line-scan

cameras, captures a single slice of the image, while a two-dimensional array, used

in video and still cameras, captures a two-dimensional picture corresponding to

the scene projected onto the focal plane of the sensor. Once the array has been

exposed to the image, a control circuit causes each capacitor to transfer its

contents to its neighbor (operating as a shift register). The last capacitor in the

array dumps its charge into a charge amplifier, which converts the charge into a

voltage. By repeating this process, the controlling circuit converts the entire

contents of the array in the semiconductor to a sequence of voltages. In a digital

device, these voltages are then sampled, digitized, and usually stored in memory;

in an analog device (such as an analog video camera), they are processed into a


19/27

continuous analog signal (e.g. by feeding the output of the charge amplifier into a

low-pass filter) which is then processed and fed out to other circuits for

transmission, recording, or other processing. The internal structure of CCD is as

shown in Fig 5.2.1.

Fig 5.2.1: Charge Coupled Device.

5.3.Design of 1H delay line (CCD)Line delays corresponding to one horizontal line scanning period (1H),

which is usually in micro seconds in the NTSC system can be considered, 1H

delay lines can be generated using charged coupled devices (CCD) which is also

known as analog shift registers as shown in Fig 5.3.1.


20/27

Fig 5.3.1: Block Diagram of Line Delay.

Each CCD operates to delay signals in the baseband or video frequency range

(e.g. 0 to vicinity of 5 MHz). A Fairchild Weston CCD321, fabricated in the

buried-channel NMOS, Clock driver circuit driven by a crystal oscillator which

provides two phase symmetric waveforms 1 and 2 to the CCD.

The type of clock driven circuits used will be function of the type of CCD

chosen and are typically based on the TTL or CMOS family integrated circuit

devices. For the FairchildWeston CCD321, which has a charge injection port at

its input and a sample-and-hold circuit in its output amplifier, the two-phase

system of clocks 1 and 2 is applied to the device to effect charge injection at

the input as well as inter stage charge transport and clocking of the a CCD offers

the advantage of reducing clock frequency feed through components in the output

signal. Any of these undesirable frequency components that remain in the output

may be further suppressed by a 5Mhz low pass filter circuit.


21/27

5.4.Design of analog processor section.This subsystem computes y(t,nT) recursively from the direct and delayed

input and the delay output, as shown in Fig 5.4.1, using the built-in coefficient

values corresponding the application.

Fig 5.4.1: Analog processor section.

Eliminating DC Offsets.

Dc offsets voltages are added to signals by the DC errors of the amplifier and

by bias level shifts. In a filtering application the signals are AC. Thus all elements

in the design were AC coupled as a straightforward method of removing DC

offsets.

Inverting Amplifier.

An inverter is required in the process section wherever a signal must undergo

a sign change with unity gain. Conventional op-amp inverters based on theLM318 wide-band op-amps can be used. A capacitor of value say 4.7pF in

parallel with the feedback resistors is required to prevent oscillations in the output

due to stray capacitance.

Summing Amplifier.


22/27

Summing amplifier in the processing unit were based on the LM318 op-amp

used in the inverting configuration in which the inverting input is the summing

node. If the voltages 1 V1, 2 V2, ., n Vn set by input attenuators are applied

to the inverting input through 10K resistors the summed output voltage is

(Rf/10K) (1 V1, 2 V2, ., n Vn) (where Rf is the feedback resistance). A

given filter coefficient is obtained as the factor (Rf/10K) i.

Integrator.

The processing unit incorporated a conventional single pole op-amp circuit

(based on the LM318). For an input signal Vi, the output is given by

V0 = - 1/ (RC) dt.

The value of the time constant RC is selected so that the peak output voltage

falls within the dynamic range of the op-amp for the lowest video frequency

component in the input signal. The integrator is set to a zero initial condition at

the start of each line scan (say every 63.5us) by means of a fast analog switch of

4066 CMOS IC type connected in parallel with capacitor C f. The sync pulse,

which occurs at the beginning of each line scan period, is separated from the

video signal, limited to 12 VpK and applied to the control input of the analog

switch.


23/27

6.Filtering of images.The operation of the prototype 2D analog filter can be demonstrated with an

application drawn form, a phase contrast filter. In the phase contrast filtering

technique, this enhances high frequency components in the image, the filter

transfer function H(s,z) has magnitude response which is flat and a phase

response that causes those frequency components in the input signal that are

above a given critical frequency c to be shifted -180 degrees out of phase so that

after the original image is subtracted, the frequency components below c will be

removed while those above will be double in magnitude.

In order to determine the real time operation of the 2D analog filter on TV

images, the prototype is inserted into television receiver circuitry as shown in Fig

6.1.

Fig 6.1: Prototype of 2D analog filter.

The separated sync signal is brought out from the circuitry, limited to 12 VpK

and connected to the control input of the analog switch in the integrator section of

the filter. The detected video signal is available at the emitter follower at

approximately a one volt peak-to-peak level. The signal is 2D filtered by the

prototype and sent to the final video stage, resulting in a phase contrast enhanced

image on the TV screen. An example before and after pictures showing the result

of filtering are given in the Fig 6.2(a & b).


24/27

Fig 6.2: (a) Before Filtering (b) After filtering


25/27

7.Evaluation.A 2d analog filter can be constructed with conventional components and

applied to the processing of the TV images in real time. The type of filtering done

is determined by coefficient settings. The resolution of the filtered picture is N x

M. A digital filter architecture which can be realized with hardware of

approximately the same order of complexity as for a 2D analog filter is the

distributed arithmetic architecture. A comparisonbetween the analog and digital

approaches in terms of hardware complexity, speed, and cost is provided next.

7.1.Advantages over digital filter.In both approaches the hardware complexity increases linearly with order. The

analog approach benefits from modularity in extending order. The analog

approaches benefits from modularity in extended order. The analog filter was

realized using only op-amps, CCDs, analog switches, TTL gates and passive

components.

The analog approach is capable of real time performance irrespective of filter

order. The digital filter prototype package count upto 100 ICs process images of

size up to N x M (pixels) at a speed of X kpixels/s. The 2D analog filter processesimages with N x M at a rate of N x M x S = Y Mpixels/s and requires a package

count of 40 ICs for a 2 x 2 implementation with an overall power dissipation of

10W. A 2D analog prototype for a 2 x 2 (say) structure would cost less than 2D

digital distributed arithmetic prototype which is not capable of real time

processing. Faster logic families such an ECL would require higher throughout

distributed arithmetic filter. This would increase the cost and power consumption

considerably.


26/27

7.2.Extension to higher order.2D analog filters can be demonstrated with a 1 x 1 recursive structure. Higher

order filters can be realized using the line recursive structure. Extension to

higher order can be done by adding more of the basic modules analog delays

and line recursive processors. Furthermore, modules of each type can be

reproduced identically, for greater ease of fabrication. Processor modules will

differ only in filter coefficients.


27/27

References.

[1] Kaufman, H.J., Sid-Ahmed, M.A, "2-D analog filters for real time videosignal processing", Consumer Electronics, IEEE Transactions, May 1990.

[2] Sid-Ahmed, M.A, Two-dimensional analog filters: a new form of

realization, Circuits and Systems, IEEE Transactions, Jan 1989

[3] Parag Havaldar, Gerard Medioni, "Multimedia Systems: Algorithms,

Standards, and Industry Practices", July 21, 2009 | ISBN-10: 1418835943.

[5] Lim, Jae S., Two-Dimensional Signal and Image Processing, Englewood

Cliffs, NJ, Prentice Hall, 1990, pp. 202-213.

[4] http://en.wikipedia.org/wiki/Analog_television

Documents

Rahul Despande - Copy