Chapter 12. Sensitive materials and image sensors

192 Sensitive materials and image sensors



Although a number of different mechanisms have

been suggested to account for the details of latent

image formation, there is considerable experimental

evidence supporting the basic principle suggested by

Gurney and Mott in 1938. The essential feature of this

treatment is that the latent image is formed as a result

of the alternate arrival of photoelectrons and interstitial silver ions at particular sites in the crystal.

Figure 12.2 shows the important steps. Here the

process is considered as occurring in two stages:

(1)

Figure 12.1 Schematic representation of the energy levels

relevant in latent image formation



move and conduction can take place. Between these

two levels is the band gap in which electrons cannot

exist. An energy of 2.5 eV is equivalent to a

wavelength of approximately 495 nm, and corresponds to the longest wavelength of the spectral

sensitivity band of ordinary silver bromide. Also

indicated in Figure 12.1 are the relevant energy levels

of an adsorbed dye molecule suitable for sensitizing

silver bromide to longer wavelengths. It can be seen

that a photon of energy much less than 2.5 eV may,

upon absorption by the dye, promote the molecule

from its ground energy state to its first excited state.

If the excited electron can pass from the dye to the

crystal, latent image formation may proceed. In this

way it is possible to sensitize silver halides to green

and red light and even infrared radiation, although it

should be noted that in the last case electron

transitions arising from the thermal effects may cause

crystals to become developable without exposure.

Excessive fogging can result and careful storage and

usage of such materials is necessary.



Figure 12.2



(2)



The nucleation of stable but sub-developable

specks.

Their subsequent growth to just-developable

size and beyond.



The broken arrows indicate the decay of unstable

species, an important characteristic of the process.

The first stable species in the chain is the two-atom

centre, although it is generally accepted that the size

of a just developable speck is 3 to 4 atoms. This

implies that a crystal must absorb at least 3 to 4

quanta in order to become developable, but throughout the sequence there are various opportunities for

inefficiency, and as a general rule far more than this

number will be required. During most of the nucleation stage the species are able to decay, and liberated

electrons can recombine with halogen atoms formed

during exposure. If this occurs, the photographic

effect is lost. The halogen atoms may also attack the

photolytically formed silver atoms and re-form halide

ions and silver ions. Although gelatin is a halogen

acceptor and should remove the species before such

reactions can occur, its capacity is limited and its

efficiency drops with increasing exposure.

These considerations lead to possible explanations

for low-intensity and high-intensity reciprocity-law

failure. During low-intensity exposures, photo-electrons are produced at a low rate and the nucleation



Latent image nucleation (top row). Initial growth (bottom row)



Sensitive materials and image sensors



Figure 12.3



Energy levels in a p-type semiconductor



stage is prolonged. As a result the probability of

decay and recombination is relatively high. During

high-intensity exposures large numbers of electrons

and bromine atoms are simultaneously present in the

crystals. Such conditions can be expected to yield

high recombination losses. Also, nucleation may

occur at many sites in a single crystal, producing

large numbers of very small silver specks.



Image formation by

charge-coupled devices

The heart of the CCD and most electronic image

sensors is the metal oxide semiconductor (MOS),

which is illustrated in Figure 12.3. It comprises

silicon doped with impurities, such as aluminium, so

that an acceptor level of positive holes is formed in

the band gap, which is able to accept electrons excited

from the valence band. Since silicon has a band gap

of 1.1 eV it has a sensitivity up to about 1100 nm.



Figure 12.4



193



Whilst this may be an advantage for recording in to

the IR region of the spectrum in certain applications,

it is a disadvantage for imaging within the visible

region of the spectrum and would result in incorrect

tone and colour reproduction. CCD cameras have a

special IR absorbing filter included to restrict their

spectral response to the visible region.

Figure 12.4 is a simple schematic diagram of

p-type silicon MOS capacitor. The positive voltage to

the gate (Vgate ) causes mobile positive holes to

migrate to the ground electrode. The region with an

absence of positive charge is called the depletion

region. If light of sufficient energy (greater than

1.1 eV) is absorbed in the depletion region an

electron-hole pair is produced. The electron remains

in the depletion region (see Figure 12.4) whilst the

positive holes migrate to the ground electrode. The

number of electrons that can be retained in the

depletion region is the well capacity, which depends

upon the physical sizes of the oxide layer and the gate

electrode area as well as the applied voltage.

In order to record images the sensor contains a

large number of elements (pixels) in the form of a

two-dimensional array for most CCD based cameras

and a linear array for many scanning devices and

some types of studio based cameras. The sensor size,

number of picture elements and the number of grey

levels are all becoming larger as the technology

matures. At the time of writing, full frame (36 ×

24 mm) 35 mm camera format size sensors are

becoming available with 6 million pixels, although at

relatively high cost.

The CCD array consists of a series of gates.

Adjustments of the gate voltages allows transfer of

electrons from well to well. This is a function of time.

Figure 12.5 shows the way in which charges are

transferred between neighbouring wells (1). After

exposure to light, the charge in well 1 is shown in (2).

If a voltage is applied to gate 2, electrons flow from

well 1 to well 2 (3), and (4) shows the situation when

equilibrium is reached. If the voltage to gate 1 is



The basic element of an electronic image sensor: p-type silicon metal oxide semiconductor



194 Sensitive materials and image sensors



Figure 12.5 Transfer of charge between neighbouring pixels

in a CCD sensor (after Holst). The potential in the well

increases towards the bottom of the well



reduced electrons begin to be completely transferred

to well 2 (5), and in (6) complete transfer is shown.

Thus the charges forming the image are transferred

in a time-dependent way if one considers a complete

array of image elements and their gates. The clocking

sequence for a three-phase device is shown in Figure

12.6. In the diagram it can be seen that only one-third

of the pixel area is available (a fill factor of 33 per

cent) as the well capacity (three gates).

There are a number of different ways in which

CCDs can be configured for different readout modes.

Some different array architectures are given in Figure

12.7 and include linear arrays, full-frame transfer,

interline transfer and frame interline transfer. These

are provided for various applications. For example,

linear arrays are used in scanners and some studio

based still cameras; full-frame transfers are used

mainly for scientific applications; interline transfer

and frame transfer arrays tend to be used in

professional television systems and consumer

products.

The storage and transfer areas are shielded from

the light. From Figure 12.7 it can be seen that the full

frame transfer architecture consists of two almost

identical arrays, one for imaging and one for storage,

which makes them costly to produce. Interline



Figure 12.6 Clocking sequence for a typical CCD

three-phase device (after Holst). The black areas represent the

gates and the shaded areas the stored charges. Different times

are given by t1 etc.



transfer arrays, however, have shielded strips which

decreases the light sensor size and may occupy only

20 per cent of the array, i.e. have a low fill factor.

Like many commercial products, CCD detectors are a

compromise between cost, efficiency and suitability

for purpose. For example, interline transfer arrays

have low fill factors but the transfer of charge is

rapid. Manufacturers improve the fill factors of these

devices by the inclusion of microlens arrays over the

CCD, which increase the effective size of the sensor

element.

So far we have only considered arrays that respond

to light intensity and in order for CCD based devices

to record colour three methods are used. One is to

successively record each image through red, green

and blue filters to additively record colour (see

Chapter 14). This method is used for scanning colour

originals and in some studio cameras where only

static subjects are being recorded. The second

method, which is commonly used in digital cameras,

is to place a colour filter array (CFA) over the CCD.

Although many filter arrays are used in digital



Sensitive materials and image sensors



195



Figure 12.7 CCD architectures in which the arrows indicate the movement of charge and the readout output register is shown

as the horizontal box at the bottom of each diagram



cameras, a typical Bayer colour filter array is

illustrated in Figure 12.8.

The third method involves the use of a beamsplitter behind the lens which divides the input into

three separate channels which separately record the

red, green and blue signals (or other appropriate

combinations) on three filtered CCD detectors.

Whilst this is an expensive solution to the recording

of colour it allows higher resolution than is possible

with CFAs. With CFAs a single CCD is used which

reduces the effective resolution and data has to be



interpolated to provide the missing information. If a

red area is being recorded, it can be seen from Figure

12.8 that there will be pixels which will not have been

activated and the missing data has to be added.

CCDs are very complicated and sophisticated

detectors and possess a number of characteristics by

which they can be judged and compared. Some

fundamental characteristics of CCDs are summarized

in Table 12.1.



Production of light-sensitive

materials and sensors



Figure 12.8



The Bayer colour filter array (CFA)



The principal materials used in the preparation of a

photographic emulsion which is coated on a base to

form the sensitive layer are silver nitrate, alkali-metal

halides and gelatin, and all these must satisfy

stringent purity tests. The gelatin must be carefully

chosen, since it is a complex mixture of substances

obtained from the hides and bones of animals, and,

although silver salts form the light-sensitive material,

gelatin plays a very important part both physically

and chemically.

Gelatin is a protein derived from collagen, the

major protein component of the skin and bones of

animals. It is a lyophilic or hydrophilic colloid



196 Sensitive materials and image sensors

Table 12.1 Performance characteristics of CCD sensors

Characteristic



Comment



Detector size



dimensions of sensor area (mm × mm)



Pixel size



physical dimensions of pixels expressed as μm × μm for each pixel, or as numbers of vertical and

horizontal pixels for whole sensor area



Fill factor



fraction of pixel area which is sensitive to light (%)



Full well capacity



maximum charge (or signal) which the sensor element can hold (J/cm2, or number of electrons)



Noise



many sources: shot noise accompanies generation of photoelectrons and dark current electrons; pattern

noise due to pixel to pixel variations; reset noise when reading the charge and added to by the

amplifier; quantization noise arising from analogue to digital conversions. Expressed as rms electrons

or rms volts



Dark current



due to thermal generation of current (number of electrons or nA/cm2 )



Dynamic range



ratio of full well capacity to noise (dB)



Scanning mode



interlaced or progressive



Resolution



relates to detector size, pixel size and fill factor (see Chapter 24) expressed as lines/mm, dpi,

cycles/mm. Often expressed as numbers of pixels or file size. Applies to the imaging system output



Quantum efficiency



ratio of number of electron-hole pairs produced to number of photons absorbed (maximum value unity)



Spectral response



plots of quantum efficiency against wavelength



Architecture



structure and readout modes of detector arrays



made up of long chain molecules which are soluble in water. It possesses a number of unique

properties that make it particularly suitable as a

binder for silver halide microcrystals. The combination of many different physical and chemical

properties in one type of chemical compound

explains why it has not so far been possible to

replace gelatin by a synthetic polymer as a binder

for silver halides. Certain synthetic polymers may

be used in conjunction with gelatin as gelatin

extenders but not as a substitute. Gelatin has

remarkable properties as a binding agent which are

summarized as follows.

(1)



(2)



Dispersed with water, it forms a convenient

medium in which solutions of silver nitrate and

alkali halides can be brought together to form

crystals of insoluble silver halides. These crystals remain suspended in the gelatin in a fine

state of division, i.e. it acts as a protective

colloid.

Warmed in water, it forms a solution which will

flow, and, by cooling an aqueous solution of

gelatin it sets to a firm gel. Thus it is possible

to cause an emulsion to set firmly almost

immediately after it has been coated on the

base, aided by cooling. The water is removed

in a current of warm air to form a coated layer

that is reasonably strong and resistant to

abrasion.



(3)

(4)



(5)

(6)



(7)



(8)



When wetted, gelatin swells and allows processing solutions to penetrate.

Gelatin is an active binding agent, containing

traces of sodium thiosulphate sensitizers and

nucleic acid degradation products (restrainers)

that influence the speed of emulsions.

It acts as a halogen acceptor, as described

earlier.

The properties of gelatin can be modified, by

chemical reactions such as hardening or crosslinking of the gelatin chains, to make it tougher,

and reduce its tendency to swell.

It enables developer solutions to distinguish

between exposed and unexposed silver halide

crystals. In the absence of gelatin, silver halides

are reduced to metallic silver by a developer

solution regardless of whether they bear a latent

image.

Gelatin confers stability on the latent image.



Gelatins vary greatly in their photographic properties. The ‘blending’ of gelatins originating from

different sources helps in producing a binder with the

required properties and in obtaining consistency.

Modern photographic emulsion technology tends to

use ‘inert’ gelatin, i.e. gelatin containing less than

five parts per million of sulphur sensitizers and

nucleic-acid-type restrainers. A more recent development by gelatin manufacturers is to provide gelatins

which are substantially free of all sensitizers and



Sensitive materials and image sensors



197



restrainers (‘empty’ gelatin). These gelatins may then

be doped to the required degree by the addition of

very small amounts of the appropriate chemicals by

the emulsion manufacturer.

The essential steps involved in the preparation of

photographic emulsions consist of the following:

(1)



Solutions of silver nitrate and soluble halides

are mixed in the presence of gelatin under

carefully controlled conditions, where they react

to form silver halide and a soluble nitrate. This

stage is called emulsification.

(2) This emulsion is subjected to a heat treatment in

the presence of the gelatin in a solution in which

the silver halide is slightly soluble. During this

stage the crystals of silver halide grow to the

size and distribution which will determine the

characteristics of the final material, particularly

in terms of speed, contrast and graininess. This

stage is known as the first (physical or Ostwald)

ripening.

(3) The emulsion is then washed to remove the byproducts of emulsification. In the earliest

method of doing this the emulsion was chilled

and set to a jelly, shredded or cut to a small size

and then washed in water. In modern practice,

washing is achieved by causing the emulsion to

settle to the bottom of the vessel by the addition

of a coagulant to the warm solution. The liquid

can then be removed by decanting etc.

(4) The emulsion is subjected to a second heat

treatment in the presence of a sulphur sensitizer.

No grain growth occurs (or should occur) during

this stage, but sensitivity specks of silver

sulphide are formed on the grain surface and

maximum sensitivity (speed) is reached. This

stage is known as the second ripening, digestion

or chemical sensitization.

(5) Sensitizing dyes, stabilizing reagents, hardeners,

wetting agents, etc., are now added.

All these operations are capable of wide variation

and it is in the modification of these steps that much

progress has been made in recent years, so that

emulsions of extremely diverse characteristics have

been produced. In a given photographic emulsion the

silver halide crystals vary both in size and in the

distribution of their sizes. They are said to be

polydisperse unless special conditions are used in

their preparation such that they are all of the same

size (monodisperse). Not only does the crystal size

and distribution of crystal sizes affect the photographic emulsion, but the shape of the crystals, the

nature of the halides used, their relative amounts and

even the distribution of the halides within a single

crystal have pronounced effects.

Manufacturers of sensitive materials are able to

control all these and many other factors, and to tailormake emulsions so that the material is optimized with



Figure 12.9 Examples of types of emulsion grains

(crystals).(a) Twin grains (Agfa). (b) Double structure (Fuji).

(c) Tabular or T-grains (Kodak). (d) Cubic (Konica)



respect to speed and granularity for a specific

application, as well as having the appropriate characteristics of tone and colour reproduction. In the past

ten years much of the research and development work

has led to the manufacture of colour negative films

and papers of high sensitivity, low granularity and

high image quality. For optimizing the sensitivity,

resolution and granularity of colour negative films the

individual manufacturers adopt their own individual

methods.

Some examples of the various sophisticated ways

of employing specific types of silver halide crystal in

colour negative films are illustrated in Figure 12.9.



198 Sensitive materials and image sensors



T-grains (flat tabular crystals) allow maximum

sensitizing dye to be adsorbed to the grain surface,

and encourage very efficient absorption of light, as

well as possessing other useful optical and chemical

properties that give higher resolution and lower

graininess than might be expected from crystals of

large surface area. Double-structure crystals optimize

both absorption of light and graininess. Monodisperse

crystals of appropriate size minimize light scatter

within the layer, and so improve image resolution.

Much of this sophisticated emulsion technology,

originally devised for colour materials, has also been

applied to the modern generation of black-and-white

materials. Examples include Ilford’s XP-2, a monochrome film that yields a dye image and is processed

in colour-processing chemicals, Kodak’s ‘T max’

films using Kodak’s ‘T-grain’ technology, and Fuji’s

Neopan films which use their ‘high-efficiency’ lightabsorption grain technology.



The support

The finished emulsion is coated on to a support,

usually referred to as the base, most commonly film or

paper. The base used in the manufacture of films is

usually a cellulose ester, commonly triacetate or

acetate-butyrate, although manufacturers also use

newer synthetic polymers which offer important

advantages in properties, particularly dimensional

stability. This makes them of special value in the fields

of graphic arts and aerial survey. The first of these

newer materials to be used widely was polystyrene;

but the most outstanding dimensionally stable base

material to date is polyethyleneterepthalate, a polyester and the raw material of Terylene fabric. As a film,

polyethylene terephthalate has exceptionally high

strength, much less sensitivity to moisture than

cellulose-derivative films, and an unusually small

variation in size with temperature (see Table 12.2).

Being insoluble in all common solvents, it cannot be

fabricated by the traditional method of ‘casting’

(spreading a thick solution of the film-former on a

moving polished band or drum, evaporating off the



solvent and stripping off the dry skin). Instead, the

melted resin is extruded, i.e. forced through a die, to

form a ribbon which is then stretched while heated to

several times its initial length and width. A heat-setting

treatment locks the structure in the stretched condition,

and the film is then stable throughout the range of

temperatures encountered by photographic film. Polystyrene is extruded in a somewhat similar manner.

Film base differs in thickness according to the

particular product and type of base, most bases in

general use coming within the range of 0.08 mm to

0.25 mm. Roll films are generally coated on 0.08 mm

base, miniature films on 0.13 mm base, and sheet

films on 0.10 mm to 0.25 mm base. Polyethylene

terephthalate base can usually be somewhat thinner

than the corresponding cellulosic base because of its

higher strength.

Glass plates have now almost entirely been

replaced by films but they are still used occasionally

in a few specialized branches such as astronomy,

holography and spectroscopy, because of their dimensional stability and rigidity, although the dimensional

stability of modern safety film bases and the application of electronic sensors is such that the use of glass

plates is now rarely essential. An additional advantage of using plates is that they lay flat in the camera;

however, these advantages are outweighed by their

disadvantages of high cost, fragility, weight, storage

space requirements and loading difficulties.

The paper used for the base of photographic and

digital hard copy must be particularly pure. Photographic base paper is, therefore, manufactured at

special mills where the greatest care is taken to ensure

its purity. Before photographic paper is coated with

emulsion, the base is usually coated with a paste of

gelatin and a white pigment known as baryta (barium

sulphate), to provide a pure white foundation for the

emulsion, giving maximum reflection. However,

most modern paper bases are not coated with baryta

but are coated on both sides with a layer of

polyethylene and are known as PE or RC – polyethylene or resin-coated papers respectively (Figure

12.10). The upper polyethylene layer contains titanium dioxide as the white pigment and optical



Table 12.2 Properties of supports

Glass

Thermal coefficient of expansion (per °C)

Humidity coefficient of expansion (per % RH)



Paper



Cellulose triacetate

0.0055%



<0.002%



0.003 to 0.014%



0.005 to 0.010%



0.002 to 0.004%



0.001%

0.000



Polyethylene terephthalate



Water absorbed (at 50% RH, 21 °C)



0.0%



7.0%



1.5%



0.5%



Processing size change



0.00%



–0.2 to –0.8%



<–0.1%



+0.03%



Tensile strength at break (N cm–2)



13 730



690



9650



17 240



Sensitive materials and image sensors



199



Figure 12.11 Schematic diagram of a coating machine



Coating the photographic emulsion

Figure 12.10 Construction of photographic papers.

(a) Baryta paper. (b) Polyethylene or resin-coated paper



whiteners. They are impermeable to water, which

prevents the paper base from absorbing water and

processing chemicals. This results in substantially

shorter washing and drying times than are required by

traditional baryta-coated fibre-based papers.

Different papers are required for the hard copy

output of digital images, particularly if ‘photographic

quality’ is a requirement. In addition to the use of

heavyweight bases of 264 g/m2, papers for ink-jet

printing require special surface coatings which guarantee homogeneous surface penetration, minimum

image spread, high gloss and brightness. Coated

paper surfaces may use relatively thick coatings

(20 μm) of silica in a starch binder. Dye diffusion

thermal transfer (D2T2) papers also have special

surface coatings to optimize dye uptake, stability,

gloss and brightness and the ability to withstand the

high temperatures required for the dye transfer.



Figure 12.12



The coating of modern materials is a complex task

and the coating methods in current use are the subject

of commercial secrecy. One of the earliest forms of

coating flexible supports was ‘dip’ or ‘trough’

coating (see Figure 12.11), but this method has been

replaced by one or other of the coating techniques

outlined below. Dip coating is a slow method because

the faster coating results in thicker emulsion layers

which are difficult to dry, and may have undesirable

photographic properties.

In order to increase the coating speed this method

has been modified by the use of an air-knife, which is

an accurately machined slot directing a flow of air

downwards onto the coated layer and increases the

amount of emulsion running back into the coating

trough. Coating by this method results in the use of

more concentrated emulsions, faster coating speed

and thinner coated layers. Other more modern coating

methods employ accurately machined slots through

which emulsion is pumped directly onto the support

(slot applicator or extrusions coating) or after flowing

down a slab or over a weir onto the support (cascade

coating) (Figure 12.12). Such coating methods allow

coating speeds to be far higher than were possible



Cross-section of a multiple cascade coating head (Agfa)



200 Sensitive materials and image sensors



with the more traditional methods. This allows very

high coating speeds to coat base material approximately 1.4 metres wide. Modern monochrome materials have more than one layer coated; colour

materials may have as many as fourteen layers, while

instant self-developing colour-print films have an

even more complex structure. In modern coating

technology, many layers are coated in a single pass of

the base through the coating machine, either by using

multiple slots or by using a number of coating

stations, or by a combination of both. The coated

material is chilled to set the emulsion, after which

(with films and papers) a protective layer, termed a

non-stress superheat, is applied to reduce the effects

of abrasion. The emulsion is then dried.

With fast negative materials it is usually impossible

to obtain the desired properties in any single emulsion. Two emulsions may then be prepared which

together exhibit the desired characteristics. These

may be mixed and coated as a single emulsion, or

they may be applied as two separate layers, an

undercoat and a top coat. For the highest speeds the

top coat consists of the fastest component emulsion

while the undercoat comprises a slower layer. This

arrangement gives the required combination of speed,

contrast and exposure latitude.



Table 12.3 Some common sizes of sheet film and papers

Size (cm)

10.2

10.5

14.8

16.5

20.3

21.0

25.4

29.7

40.6



×

×

×

×

×

×

×

×

×



12.7

14.8

21.0

21.6

25.0

29.7

30.5

42.0

50.8



Size (inches)

4

4

5.8

6.5

8

8

10

11.7

16



(A6)

(A5)

(A4)

(A3)



×

×

×

×

×

×

×

×

×



5

5.5

8.3

8.5

10

11.5

12

16.5

20



MOS techniques. Their fabrication involves deposition of many layers of silicon compounds, implantation of specific ions and a metallization layer on ultrapure silicon wafers. Theuwissen (1996) gives details

of 29 stages involved in the production of a typical

CCD sensor. This explains their relatively high cost

and difficulty in mass producing a 24 × 36 mm, or

larger, full frame imager because of the increase in

defects as the sensor size increases and the high

wastage rate.



Sizes and formats of photographic

and electronic sensors and media



Charge-coupled devices (CCDs)

Photographic sensors are complex to manufacture

and require a number of stages and components of

high purity. They can be produced in large lengths

and widths to provide homogeneous sensors of large

area which are subsequently cut to appropriate sizes.

CCD sensors, however, are far more difficult to

produce and are batch multi-stage processes based on



Sheet films and papers are supplied in a great many

sizes. The nominal sizes in most common use are

given in Table 12.3.

Roll and films are made in several sizes, identified

by code numbers. A number of sizes are given in

Table 12.4. Where more than one format is given in



Table 12.4 Roll film sizes

Size coding



Film width

(mm)



Nominal image size

(mm)



110



16



13 × 17



12/20



Single perforations, cartridge loaded



APS



24



17 × 30



15/25/40



Advanced Photo System, different aspect

ratios possible, data recorded on magnetic

strip, processed film remains in cartridge



126



35



26 × 26



12/20



Single perforations, cartridge loaded



135



35



24 × 36



12–36



Double perforations, cassette loaded



120



62



45

60

60

60



×

×

×

×



16/15

12

10

8



Unperforated, rolled in backing paper



220



62



70 mm



70



60

60

70

90



Number of images



As for 120 but double number of images



Notes



Unperforated film with leader and trailer

Double perforations, cassette loaded



Sensitive materials and image sensors



201



Table 12.5 Sizes and resolutions of typical CCD sensors

Approximate size

(H × V)

(mm × mm)



No of pixels

(H × V)



File size*

(mB)



5.3 × 4(8 mm, 1/3 in)



320

640

756

1024



×

×

×

×



240

480

504

768



0.2

0.9

1.1

2.2



8.8 × 6.6 (2/3 in)



1280 × 960



3.7



14 × 9.3



1524 × 1012



4.4



24 × 36 (35 mm)



3072 × 2048



18



*Assumes no data compression, 8 bits/colour and 3 channels.



Table 12.4, the size achieved in any given instance

depends upon the design of the camera used. In

addition, with modern cameras and encoding of films

different image sizes, selected by the choice of

camera or camera settings, are now becoming more

popular.

In 1996 a new format was introduced known as the

Advanced Photo System (APS). This was defined

jointly by Eastman Kodak Company, Fuji Photo Film

Co. Ltd., Canon Inc., Minolta Co. Ltd. and Nikon

Corporation. In comparison with previous cartridge

systems, APS has the unusual features of storing the

processed film in the cassette and incorporating a

magnetic coating for recording information about the

photographer’s choice of formats. The film format

and the cassette is smaller than 35 mm, which has

allowed the design of more compact cameras with a

very much simpler loading system (see Figure

12.14).

Generally, for image sensors size is related to

quality; the larger the area the better the quality.

Manufactures of photographic sensitive materials

have improved the quality of their products over a

long period of time such that smaller format materials

can be introduced which are of acceptable quality to

the consumer, the most recent one being the APS

technology mentioned above. Electronic sensors are a

much more recent innovation and have had a shorter

period of evolution of less than 30 years compared

with 150 years for photographic materials, but are

advancing rapidly. Table 12.5 lists some typical CCD

sensor sizes, their number of pixels and file sizes.

Because of the complex multi-stage manufacturing

process and structure of CCDs it is very difficult to

mass produce large area CCDs. The failure rate is

high, which makes them expensive and the larger the

area and the greater the number of pixels the higher is

the chance of defects which also makes the production of high resolution (large numbers of pixels in a

given area) costly.



Figure 12.13



DX coding of 35 mm film



Film coding

DX coding was introduced more than a decade ago

for 35 mm films which provides information for the

photographer and the processing laboratory. This

system of coding is shown in Figure 12.13. The

cassette has printed on it an auto-sensing code which

enables certain cameras to set the film speed on the

camera to that of the film automatically by making

appropriate electrical contact with the pattern on the

cassette. Also printed on the cassette is a machinereadable bar code. A raster pattern punched into the

film leader and a bar code along the edge of the film

provide data about the film for the processing

laboratory. More recently, APS was introduced (see

Table 12.4 and Figure 12.14) and provides a magnetic

recording coding system for the photographer and for

the processing laboratory. In addition to the magnetic



202



Sensitive materials and image sensors



Figure 12.14



The Advanced Photo System (APS)



data for recording information at the time of exposure, the end of the cartridge has visual indicators of

the exposure status of the film to show if the film is

unexposed, partially exposed, fully exposed but not

processed, or exposed and processed (see Figure

12.14).



File formats

There are a vast number of different file formats which

are used for handling, manipulating and storing digital

images, many of which arose from the requirements of

computer graphics, driven by manufacturers of hardware and software. Virtually all graphics programs

save files in their own proprietary or native format but

most modern programs are able to read a variety of

different formats. This is now being driven by the endusers who demand inter-system compatibility, the

gradual emergence of standards and manufacturers

joining together to provide file formats that are more

universal and more appropriate to the requirements of

photorealistic imaging.



There are two basic overall types of files for

storing image data – these are vector graphics and

bitmap (raster) image structures. Vector graphics files

comprise data in the form of vectors that mathematically define lines and curves by their geometrical

formulae but cannot be used for photorealistic

imagery. This format has the advantage that it can be

re-sized or re-scaled without distortion, the images

are of good quality on any output device and is most

suitable for bold sharp graphics rather than for

continuous tone photographic types of imaging. Also

it results in smaller file sizes with less need for data

compression. Examples of vector graphics formats

include: CDR (CorelDRAW), CGM (Computer

Graphics Metaphile) and WMF (Microsoft Windows

Metafile). A more complete listing is given in Table

12.6. Bitmap or raster images divide the image in to

a grid of equally sized squares (pixels). Each pixel is

defined by a specific location and colour (see Figures

1.1 and 1.2). This file format is especially suitable for

photographic types of imagery, which involve continuous tone with subtleties of colour and tone but

suffer from the disadvantage of being resolution