Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 39
--> 4 Description and Assessment of Features The importance of assisting visually disabled people was not addressed at the time the current U.S. banknotes were first issued in 1929, and the notes were manufactured using then-available technology. In a world of mainly monochromatic commercial print, the 1929 designs with their characteristic green backs were easily recognizable. As the twentieth century comes to an end, everyone is aware of visual and tactile presentations that distinguish and sell items in daily life. Television and supermarkets are models on how to mix a multitude of similar products and yet achieve separation by techniques of brand identification. The awareness that different groupings of the population can be specifically addressed is now well understood. It is against this background that the committee has looked at features that could be incorporated into the U.S. banknotes to aid visually disabled people to recognize the presence of a banknote, identifying its denomination and, if possible, providing secure authentication. A list of such features was drawn up from a variety of sources. The features were ones used or recommended by foreign currency issuers (see Appendix D), discussed in previous studies, suggested by participants in the workshop held by the committee, and identified by committee members. Some features on the original list were currently considered too impractical for complete evaluation. These features are listed in Appendix E. To make the evaluation easier to comprehend, the committee decided that the best method for grouping the features was to address how information would be obtained by the user. By combining the target population with the methods of identification, the following four major groupings were identified: visual/tactile features, visual-only features, tactile-only features, and machine-readable features. Figures 4-1 and 4-2 show examples of some of the various features that the committee considered; the features illustrated in these figures will be discussed in their appropriate chapter section. Visual/Tactile Features Features in this category give both visual and tactile clues to the denomination and would be usable by both blind and sighted people. These include features that may be incorporated into the substrate during manufacture, for example, changes in the physical dimensions of the note or features that involve modification of the substrate after manufacture, such as modification of the comers or edges or the addition of holes.
OCR for page 40
--> Figure 4-1 Illustrations of various ways to use size to indicate banknote denomination: (a) length only; (b) height only; (c) both length and height; (d) hybrid use of length and height. Size Conceptually, the simplest visual/tactile approach is to have different-sized notes, where the size indicates the denomination (see Figure 4-1). This is an established practice in over 120 countries in the world (see table in Appendix D). The length, height, or both dimensions of the note can be changed depending on denomination. The usual practice is to have the higher denominations of a larger size, so that the crime of "raising" the value cannot be achieved. Figure 4-1 illustrates the types of dimensional changes possible. Figures4-1a and 4-1b show single-dimension variations of length and height, respectively. Figure 4-1c shows an increase in both length and height for each increase in denomination. A combination of length changes with alternating changes in height is shown as a "hybrid model" in Figure 4-1d.
OCR for page 41
--> Figure 4-2 Illustrations of various features evaluated: (a) comer cuts; (b) edge notches; (c) holes; (d) large numerals, uncluttered background; (e) distinctive shapes; (f) coarse patterns. (Features in (a) and (b) have been exaggerated for illustration purposes.) In countries where the currency is size-denominated, organizations representing and supporting blind people, such as the Royal National Institute of the Blind and the European Blind Union, recommend continuation of this practice (EBU, 1994; Farber, 1994; King, 1994; Jarvis, 1994). The use of this feature in U.S. banknotes has been an ongoing recommendation by those who are familiar with the use of different-sized banknotes and is considered to be extremely useful once an individual is trained to use it. Participants in the committee's workshop also recommended the use of size-denominated banknotes, based on their experiences in traveling and living in other countries and in handling coins and other size-differentiated objects (Hill, 1994; Miller, 1994; Wallach, 1994; Abbott, 1994). Sizing banknotes by denomination does not provide any clue to the orientation of the bill other than identification of length versus height. To give orientation information, an additional
OCR for page 42
--> feature would be necessary. Such a feature would not need to give information about a bill's denomination. The committee found no technical literature evaluating the effectiveness of denomination differentiation by size other than a mention in a paper evaluating the effectiveness of the tactile marks used on Dutch currency (Koeze, 1990). In that paper, one subject who identified all denominations correctly 100 percent of the time used banknote size, although distinctive tactile markings were available. The BEP report (1983) indicates that there was research done by the Central Bank of Switzerland that indicated the usefulness of size-denominated banknotes for blind people. Generally, there is little published literature directly related to banknote features due to the confidentiality with which currency design, handling, and production are treated. To the committee's knowledge, careful psychophysical measurements have not been reported on size discrimination and identification in a context directly applicable to currency bills. Tactile judgments of the size of bills involve active exploration with the hands and fingers. This type of "active touch" makes use of sensory inputs from both touch receptors on the skin (cutaneous sense) and positional receptors in the muscles and joints of the hands and fingers (kinesthetic sense). Basic psychophysical studies have tended to isolate each of these two components of tactile perception for separate study. For example, Durlach et al. (1989) performed a detailed study of tactile length discrimination and identification using judgments of distance between the thumb and forefinger (the kinesthetic sense). For a length of 80 mm, the Weber fraction was 3 percent (2.4 mm), but the Weber fraction increased slightly for shorter lengths. The Weber fraction of 3 percent was consistent with earlier measurements of length discrimination and represents the committee's best available estimate of a lower bound of appropriate length differences for banknote design. In the same study, Durlach et al. also confirmed the point made in Chapter 2 regarding absolute judgment; even if length differences exceed the discrimination threshold, errors of absolute judgment occur when the number of alternatives exceeds about four to seven. Based on the limited technical data and the strong recommendations for this feature from blind communities in Europe, some very preliminary experiments were carried out by two members of the committee to examine these aspects (Legge, 1994; Brabyn, 1994). Legge concentrated on size variation in length only. Six "banknotes" with 6 mm differences between "denominations" (mean length of 156 mm), could be identified with 50 percent accuracy. When the step size was increased to 12 mm, identification improved to 75 percent accurate. Brabyn showed that using four sizes of notes that varied in size in two dimensions—length by 7 mm and height by 5 mm—a 90 percent success rate for identification could be achieved in a short learning time (about one-half hour). The coding of both length and height can be thought of as a special case of the addition of a secondary cue to ease identification, which is described in the discussion of discrimination versus absolute judgment in Chapter 2. More work is needed to identify the optimum sizes, but this needs to involve the BEP to ensure that the cost efficiencies of different sizes, including plate layout considerations, are also taken into account. To determine the optimum set of sizes for six denominations, the approaches used by the many countries that include this denomination cue in their currency should be evaluated. In addition, the threshold differences in dimension required between two adjacent denominations need to be examined to add a technical basis for the choice of sizes. For steps to safely exceed the Weber fraction for length, they will probably need to be at least 3 percent of the base length.
OCR for page 43
--> The European Blind Union recommends a minimum difference of 5 mm between denominations (EBU, 1994). Such differences could lead to extra-large notes. Most countries use a second cue to improve performance. This can often take the form of a template that is provided at the launch of the new notes design to provide improved discrimination. Alternatively, the notes can vary in two dimensions; this is the norm and is also recommended by the European Blind Union (EBU, 1994). With this option, the examiner has two edges to relate judgment against. Often the hand replaces the template as a known guide, using, for example, the distance from the thumb joint up the first finger. In the 1983 BEP report, a table contains the dimensions of currencies issued by 38 countries (BEP, 1983).1 From the information listed in the BEP table, 36 countries varied both dimensions of the banknote, and the Netherlands and Israel varied banknote length only. Most of the dimensional changes are not linear through the series of banknote denominations; that is, the step size from one denomination to the next larger denomination is not constant through the entire denomination sequence. Step sizes in both length and width generally fell between 4 mm and 16 mm, with some as small as 1 mm or as large as 28 mm. According to the table, the Netherlands and Israel increased banknote length by 6 mm and 7 mm per step, respectively. Design and plate layout considerations generally dictate the choice of step sizes between successive denominations. There have been general comments that variations in sizes will affect the use of ATMs (automatic teller machines) and other cash-handling machines. Although some alterations would be necessary, the committee was made aware that these machines are also used in Europe where banknote sizes vary with denomination. Recently both England and Germany have issued new series of banknotes that have different sizes than the previous notes. Despite a complicated multiplicity of sizes, the providers of ATM services were able to keep the machines working. U.S.-manufactured ATMs are used throughout the world in countries with size-denominated banknotes. A previous report (BEP, 1983) indicated that size changes were the most effective way to denominate banknotes for the visually disabled but that such changes would be costly. At the time that report was issued, there was no currency redesign anticipated, and their cost estimates included the full costs of redesigning the currency. The committee recommends that the calculation be revisited, taking into account the fact that a currency redesign is underway and based on present banknote manufacturing equipment employed and the current distribution of denominational production. For example, over 40 percent of production consists of $1 bills (Church, 1994). These would become the smallest notes under a varying size policy. Smaller notes would mean that more notes could be produced on each sheet, thus leading to savings in paper, ink, and production time. Against this, some capital equipment would require replacement, and methods of handling the finished product by BEP and the Federal Reserve banks would need changes. In addition to the BEP and Federal Reserve bank costs, there would be conversion costs for vending and other cash-handling machines outside the BEP and Federal Reserve banks, as well as other "downstream" costs to merchants and individuals. 1 Banknote designs change frequently around the world, so the numbers in the BEP report are not current. A summary is presented as an indication of the types of dimensional variations used.
OCR for page 44
--> Although there is a substantial infrastructure in the United States that relies on single-sized banknotes, the committee believes that the needs of the visually impaired would be well-served by making the banknotes size-denominated. Because of the strong recommendations from members of the target communities who have experience using this feature and because of the amount of experience worldwide in the manufacture and use of this technology, the committee recommends that the BEP work to implement size-denominated banknotes to ease use of U.S. currency by all. Edge Modification Another variation of substrate dimension changes is to notch, cut comers, or modify edges of the note, as depicted in figures 4-2a and 4-2b. Clipping comers has been considered by other countries but is not currently in use. Similar to the case of dimensional changes above, lower denomination bills would have more comers cut. For example, the $100 bill could have no comers cut off; the $50 bill, one comer (perhaps, the lower left); the $20 bill, two comers (the lower left and right); the $10 bill, a different two comers (the lower left and upper left); the $5 bill, three comers (lower left and right and upper left); and the $1 note could have all four comers cut. In this scheme, the cut comer approach would provide an orientation cue, except in the case of the one or one hundred dollar notes. In the committee's opinion, cutting the comers of the note would not be expected to increase the possibility of failure by delamination of the paper layers and, in fact, may lessen this tendency. There would be capital cost associated with this feature, since means to cut the comers or make notches would be required. Because folded comers could be confused with cut comers by both people and machines, this approach is not considered suitable as a primary denomination cue. Edge modifications to the note could be a notch with square or rounded comers cut into the banknote edge (see Figure 4-2b). A sharp notch may have a tendency to enhance delamination or tearing, whereas a more gradual pattern should not. Such edge patterns, in addition to helping with denomination, could also provide information on the orientation of the note for a blind person and help to stack notes of the same denomination in piles. Denominating the notes could be accomplished using a pattern on one edge only or on multiple edges. The patterned edges could be thought of as aids much like the edges of coins. For example, edge patterns could help distinguish between nearest neighbor denominations or could complement note size features (mentioned above) for sizes that are close. As will be discussed later, comer clipping and edge modifications could both be used in combination with size changes. For instance, clipping the comer of alternate denominations would improve accuracy in denomination and aid in orientation of the bill. As in the case of cut comers, there will be some capital cost associated with the edge modification feature. For cut comers or other edge modifications, the potential for enhanced degradation of the note, while believed by the committee to be small, must be evaluated. Because the concerns in handling and degradation with wear, the modification of banknote edges does not appear to the committee to be a reasonable approach at this time.
OCR for page 45
--> Holes Holes punched or drilled into the note offer another opportunity for both orientation and denomination. Holes should be simple and inexpensive to implement, and the denomination information could be carried by the location of the holes, the number of holes (see Figure 4-2c), or both. In addition, this approach would provide definite orientation information if the "active" quadrant was always the same. For example, if the holes were confined to the lower left quadrant of the front of the note for all denominations, it would be easy to orient the bill properly. The idea of holes was presented to the committee (Oschwald and Hendricks, 1994) and has received limited field testing with encouraging results. During these experiments, a template with raised bumps that would be felt through the holes was used to enable reliable detection. There is some debate as to the efficiency of detecting holes by feel alone. The committee was informed that depressed areas are more difficult to detect from a background than raised areas, so that holes significantly larger than braille dots would be required for accurate identification. Holes in the interior of the banknote could also be "read" by folding the note along the line of the holes and feeling the notches along the folded edge. This method combines the benefits of an edge modification, where the notches are easily distinguished, with the improved robustness of having the holes in the banknote interior, where they are less likely to start tears or delamination problems or to jam the transport mechanisms of currency-handling machines. However, continued folding along the same line will cause banknote degradation. Production implementation would require assessment, since a drilling apparatus will be necessary. Circular holes are straightforward to drill compared with shaped holes. The cost of drilling should also be cheaper than edge modifications or comer cutting. However, the impact of increasing the number of processing steps and wastage of material will need to be assessed together with durability. The committee believes that circular holes would be more durable than shaped holes. It also felt that larger holes, which would be more easily located and identified, would be more vulnerable to damage than small holes, thus a balance of perception requirements versus durability for hole size would need investigating. One drawback to the scheme of holes is that punching a hole in a document is often carried out in commercial transactions to denote that an item has no monetary worth. Because the concept of using hole location or number of holes to indicate banknote denomination is very simple and would be easy to use, the committee believes that it would be worthwhile for the BEP to assess such issues as the impact of holes on banknote durability and the psychological matters regarding issuing and using banknotes containing holes. Visual-Only Features Visual-only features would benefit everyone except people who are blind. These types of features include large numerals on a nonprinted background (to obtain high contrast), printed geometrical shapes unique to the denomination of the note, coarse printed patterns that would correspond to note value, and different colors for different note values. None of these features would provide for authentication of the note.
OCR for page 46
--> Numerals The use of large, high-contrast numerals on a simple, uniform background can provide an effective means to enable visually disabled people to more easily denominate banknotes. ''Large'' here refers to numerals greater than 40 percent of the full height of the current note (see Figure 4-2d). This size would be well within the limit of people with acuity of 20/160 or better for viewing bills at a distance of one meter, which is roughly the distance from the eye to the checkout counter of a grocery store, enabling easy and rapid identification. (See the discussion in Chapter 2 dealing with acuity levels and currency identification.) If the numeral were 60 percent of the current note height, it would serve people with acuity as low as 20/240 in the same situation. Moreover, the large, easy-to-read numerals would be recognizable by people with extremely low acuities (less than 20/1000) if bills were held at a normal reading distance of 40 cm (16 inches). Numerals of this size would be substantially larger than those on current U.S. banknotes. Large numerals are not needed on all four comers, but it would be preferable that one large numeral be included on each side of the banknote. However, a large numeral on one side only would provide an additional orientation cue. Large numerals may offer a location for additional anticounterfeiting features, such as finely detailed numeral-edge engraving. The overall high contrast with the background must be maintained, however. Large numerals may also distract from finely engraved security features, such as the portrait work, so care must be taken to integrate the numerals into the design in such a way that distraction is minimized. Along with the larger numerals must be a plain, uncluttered background to provide high contrast for easy readability. Large numerals on a plain background are common in foreign currency and were recommended by organizations supporting visually disabled people (EBU, 1994; Farber, 1994; Bennett, 1994). Current U.S. notes do not, in general, offer much contrast. Photometric measurements performed by a committee member show that contrast ranges from 30 percent (worn) to 60 percent (new; Legge, 1994). A National Research Council committee recommended in 1980 a minimum contrast of 85 percent for symbols used in acuity testing (NRC, 1980); this should be used as a starting point for design of a high contrast numeral. The size of the clear background needed relative to the size of the numeral is uncertain at this point and would likely require additional study. A reasonable starting assumption, however, is that the surrounding clear field should be at least the width of the lines forming the character. The numeral could be either white on a black background or black on a white background. People with light scatter in the ocular media (e.g., cataract) often see better with white characters on a dark background, so an argument can be made for the merits of white on black. However, dark numerals on a light background, as can be found on the new Canadian $20 note, may stand out as a "target" against a light background better than a light numeral on a dark background. Large, high-contrast numerals are simple to implement; the only concern is where to place large numerals in the crowded design. The committee believes that a large portion of the visually disabled would be served by the implementation of this feature.
OCR for page 47
--> Color Color can serve as a denomination aid if each note value is assigned a different predominant color. For example, a $5 bill could be predominantly blue. A person with low vision who could not make out the numerals (large or otherwise) may be able to discern color differences to determine the denomination. The use of color could also be a convenient aid for the normally sighted. This feature would require learning the colors associated with a given note value. It is customary in currency design for each denomination to have a single, predominant color. That color is usually on both sides of the banknote, so that notes can be distinguished by viewing either banknote face. However, if a denominating color is only on one side of the note, the color feature could also assist in front to back orientation. Having the denominating color on the front of the banknote is preferable, since people tend to orient banknotes with the front face toward the user. If policy is to require the fronts of the banknotes to be printed in black, then a second choice would be to apply a distinctive predominant color to the banknote backs. Although an argument can be made that a complicated color design with a mix of colors on one face may be a counterfeit deterrent, it is not considered desirable from a denomination standpoint. The color observed by the eye is dependent on the spectral content of the illumination source, so colors selected for banknotes would have to remain distinct under varied lighting conditions. For example, certain blues and greens can be hard to distinguish from each other in low light or by people with low vision; banknotes should not be designed to rely on discrimination between such colors. There is enough information in the area of color research to allow experts in the field to identify six colors that could distinguish the six denominations with little research or development needed (Arditi, 1994). The committee is aware that other aesthetic and design considerations may place some restrictions on the choice of a set of colors. In addition, color choices should be made that will not result in confusion for the 10 percent of the population (mostly male) with inherited color deficits. Existing printing technology at the BEP offers two ways to apply color to the banknote: intaglio printing and letterpress printing. (Lithographic printing is used in other countries to apply color to banknotes, but it is not used at the BEP.) For printing an entire face in one color, the intaglio press is used. The intaglio equipment at the BEP is capable of printing three separate colors in each print run. As a banknote is printed, the press uses either green ink (for the banknote back) or black ink (for the front of the banknote). Between denomination runs, the intaglio press is cleaned, and the plates are changed to the appropriate denomination plates. At the same time as the operator cleans the machine and changes the plates, the ink can be changed with little additional work. Since so many other countries use different-colored intaglio inks, there is little limitation to the number of color choices available off-the-shelf. The availability of many colors of intaglio ink and the capabilities of the intaglio presses make it possible to achieve a different predominant color for each denomination with little increase over current production costs. An alternative use of color to distinguish among denominations would be to apply color in a smaller area in a distinctive shape or coarse pattern (patterns are discussed in the following section). A small, distinctively placed patch of color would serve as an aid to orientation. The
OCR for page 48
--> concept of distinctive denomination colors could possibly be combined with the large numeral by using the letterpress to apply a single large numeral in color. The intaglio design would have to contain one large, clean area for application of any patch of color. The letterpress, which applies the Treasury seal and the banknote serial number (in green) and the Federal Reserve Bank seal (in black), is used only on the front side of the banknote; it would be a major change to require two-side letterpress printing. In the current U.S. banknote design, the front side of the note is crowded, with the numerals and two seals occupying the majority of the space. There is little space for an additional colored mark anywhere on the banknote front. The Treasury and Federal Reserve Bank seals represent possible areas to color distinctively by denomination. Traditionally these seals are green and black, respectively. There may be restrictions on the colors that can be used for either of the seals. The use of a predominant color for each denomination is useful to a large segment of the visually disabled population, including those whose visual acuity is below that required to read the denominational numerals, and there appear to be no technological impediments to implementation. The committee believes that this feature should be included in future banknotes. Shapes and Coarse Patterns Two additional visual features that have promise are the use of geometric shapes or coarse patterns to denote denomination. Geometric shapes (circles, squares, triangles, etc.) would preferably be large and printed on a clear or uncluttered background, just as the numerals would be (see Figure 4-2e). These sorts of shapes are currently in use on British currency (Jarvis, 1994). It would be necessary for users to learn the relationship between shape and note value. Since there is no logical connection between geometric shape and current note values, there would be some learning time necessary. Coarse patterns are a variation of the above feature, except in this case, the position of this character (or characters) rather than the shape of the character determines note value. There may be a better intuitive connection here between the pattern of the characters and note value, and again, it would be very important to have large, high-contrast characters on an uncluttered background. Figure 4-2f depicts one example of the use of coarse patterns to aid in note denomination. One subtlety to the coarse-pattern idea is that the printed pattern itself may be detailed (e.g., the portrait on the bill or a large unprinted watermark area) but may appear "coarse" to the person with low vision. The areas could be located at different, distinctive positions for the six denominations. Although people with low vision may not be able to identify the faces in portraits (or even see that they are faces), they may be able to use the locations of the patterns on the banknotes to gain information. Any coarse features implemented must be large enough to be recognizable from a reasonable distance. While recognition of a large shape should not require as large a symbol as numeral recognition (discussed in Chapter 2), it is still reasonable to assume that patches the size of the Treasury and Federal Reserve Bank seals (with diameters approximately the same as the height of the current denomination numerals) would be difficult to identify for visually disabled people when viewed at a distance usual for cash transactions. As the committee recommends that
OCR for page 49
--> numeral size should be larger than one-half the current banknote height to be legible for people with visual disabilities, it also recommends that any denominational shapes or coarse patterns used should be about the same size as the large numerals. The addition of color to these shapes or coarse patterns may improve their visibility and allow for smaller symbols to be used, since color recognition is generally easier than numeral or shape recognition. The committee considers the shape feature to be secondary to large, high-contrast numerals with a uniform background or to denomination by color and believes that at best it would be a redundant clue for denomination. The coarse-pattern feature would be useful if the portraits or similar large, shaped patterns were distinctively located on the banknote. Tactile-Only Features Tactile features are those which rely on the sense of touch to provide information as to note denomination. At present the optimum height above the background and areal size of tactile features to give the best "readability" are not known. Braille dots are approximately 400 µm high; intaglio print can give a relief of 60-70 µm. Although research has been carded out on spacing of dots, limited results are available for actual relief perception and for the effect on perception of the interactions of feature area, shape height, etc. The French Post Office produced a stamp with embossed braille symbols 72 µm high. This was not easily readable by a committee member who is proficient in reading braille. It should be noted, however, that the task of identifying six denominations with a tactile mark is much less complex than the full character recognition required for braille markings. The difficulty in reading the stamp markings does not preclude the usefulness of tactile markings of this height for denominating banknotes. It does indicate that the tactile markings should be kept as simple as possible, with differentiation not depending on subtle differences between denomination symbols. Several foreign countries have used different types of printed or embossed tactile marks on their banknotes with limited success (Koeze, 1990; Cruz V., 1994; Péterfi, 1994; Soekarna, 1994; Sirkis, 1994; Benali and Iraqi, 1994; Gaiteiro, 1994; Santipong, 1994; Johansen and Bjørgo, 1994; Trachsel and Meroni, 1994; Blomberg, 1994). Several of these sources mentioned that on crisp, new notes the printed tactile marks are quite readable, but marks quickly lose their distinctness with banknote wear, a fact that was confirmed by Koeze (1990). Nevertheless, tactile features potentially offer a variety of ways to assist visually disabled people in denominating banknotes if the durability problem is overcome. Unlike the tactile features discussed above, there are other tactile features that involve the addition of information during paper manufacture or, alternatively, later in a printing or embossing operation. Features that could be incorporated during sheet manufacture include a tactile watermark and tactile threads or planchettes that would change the local thickness and thus could be detected by touch. The first feature is used in Japanese currency (Minoshima, 1994). A cross-section and a three-dimensional measurement of the watermark from a Japanese banknote are shown in Figure 4-3, indicating a depth of approximately 50 µm for the "doughnut-shaped" watermark (BEP, 1983). In any of the tactile marking cases, the number, frequency, or distribution of the perturbations could denote denomination.
OCR for page 50
--> Figure 4-3 Illustration of variation in substrate thickness achieved by watermark in Japanese currency (BEP, 1983). One major problem with most of these approaches is that stack height of the notes will be altered. This will always be the case with the addition of a local disturbance that increases sheet thickness. If such features always appear at the same location on the note, the bills may not stack evenly, causing a variety of problems for the BEP during manufacture and for the BEP, Federal Reserve banks, and other banks during subsequent handling. Many of the other countries using these types of features do not have the automated handling and stacking equipment used by the BEP and the Federal Reserve banks in the United States and so may not experience problems due to such markings. One solution to the stacking problem is to not fix this feature to the same location on each note. However, this is generally an undesirable approach, since it adds complexity as the user does not know exactly where to look for the denomination clue. Another approach is to make the tactile information a negative displacement relative to the average thickness of the sheet. The tactile watermark uses this approach. Unfortunately, negative-displacement tactile information (holes or depressions) is much harder to "read" than positive displacement (bumps or points). To avoid the above problems, the banknotes could include a texture that encompasses the entire surface area of a note. This "sandpaper" approach could be developed during the papermaking process or by embossing the paper after printing. An embossed surface that offered
OCR for page 51
--> different patterns for different denominations could preserve stack "squareness," although stack height might differ from current height. However, if embossed patterns that would nest could be found, it might be possible to create stack heights that match the current ones. This possibility would make embossing an attractive option. Changing sheet roughness during manufacture can be done, but there may not be enough of a range of tactile sensations to allow denomination of bills. Furthermore, the intaglio printing process would tend to flatten or smooth out the irregularities, as was found during the development of the Japanese watermark (Minoshima, 1994). A related approach would be to make the current substrate much smoother. This could accentuate the intaglio printing, making its tactile sensation more obvious or noticeable. All of the tactile-only features discussed above will degrade with wear. Raised features, whether obtained by addition of specific materials or by embossing, will wear away, making the feature less useful. The tactile watermark is less prone to this type of degradation than printed or embossed features. If the banknote base paper were a laminated structure, as is now the case for some security papers, it would be possible to punch holes in one of the outside lamination layers (illustrated in Figure 4-4). This could provide a negative tactile feature on one side while not compromising sheet strength or affecting stack height. Unfortunately, as noted in the discussion regarding holes, it is more difficult to "read" depressions (as compared with bumps). This means the depression caused by the hole in one ply of the laminated sheet, just like the tactile watermark, would probably not be a very effective aid for blind people. The committee concludes here is that today there is no sure method to provide reliable tactile information on banknotes. The nature of research needed in this area is discussed in Chapter 5. Figure 4-4 Holes in one thickness of a laminated structure.
OCR for page 52
--> Machine-Readable Features This category includes features that can be used with devices that determine the denomination of a banknote. For a person who cannot benefit from visual security features, this is the only category of features that has the potential to authenticate the note, in addition to determining the denomination. Some of the features that would enhance the use of devices for denomination would be included in other categories, as they also give visual or tactile clues that can be read by an individual. Other features in this category are only useful with the aid of a device. The previous sections have addressed the potential ways that visually disabled people could discriminate between different denominations without using devices. The committee strongly believes that these routes should be pursued as primary strategies, since the features suggested can be used without instrumental aids. However, during the course of its discussions the committee concluded that the other three types of features (visual/tactile, visual-only, and tactile-only) did not provide a solution to the problem of banknote authentication. The committee felt that a device would provide the visually disabled community with a method of authentication. However, the committee also recognized that the principal device approach for blind people should be rapid denomination sorting, with the possibility of authentication where possible. Potential devices form one area for future research; some concepts are discussed in Chapter 5. The current section addresses devices and features that would enhance device usability and that the committee felt were currently achievable and could therefore be addressed in the short term. The devices can be broken down into two categories: passive devices and active devices. Passive Devices In the section on visual/tactile features, the committee indicated that size differentiation among denominations could be a useful method of discrimination. This method could be made more reliable through the use of a simple size-guide template. Such a device is made with a fixed alignment corner and varying lengths and heights of notes given by a stepped opposite edge. The device can be made in many materials and be cheap enough to issue free through various agencies. This type of device may act as an educational aid to blind people in the early stages of implementation to give them confidence in their judgment on size-based discrimination. When a person becomes familiar with the new currency, the device may no longer be necessary. Another feature discussed earlier in the report was holes. Various combinations and locations of holes could be used as a method of providing denomination discrimination. Again, a simple template would be useful. The template could contain notches that relate to the hole positioning and note values. Such a device was demonstrated to the committee (Oschwald and Hendricks, 1994). Other examples of passive devices would be magnifying glasses, flashlights, etc., which would not be limited in usefulness to currency denomination.
OCR for page 53
--> Active Devices In this category, the committee considered detection systems that require some basis of powered electronic analysis. Several banknote features that would enhance the use of these devices were considered. Several of the features recommended in a previous National Research Council report (NRC, 1993) to discourage counterfeiting would enhance the ability of devices to denominate banknotes. Specifically, security threads with either width or location indicating denomination were recommended for incorporation in the near term and would allow device designers to design their equipment based on a known code. For intermediate-and long-term implementation, the optically active coated fibers and particles, as well as metallic or specular woven thread features, might be useful to designers of banknote denomination devices. In addition to the features recommended in the previous report (NRC, 1993), the committee felt that a magnetic thread that contained a coding unique to each denomination could be inserted in the paper. This type of technology is known and has been incorporated in banknotes elsewhere, for example, in England and Italy. A possibility exists to incorporate planchettes or fibers that contain machine-identifiable characteristics as opposed to the current red and blue fibers. By incorporating these planchettes or fibers in different volume distributions or banded locations, it should be possible to derive a code relating to the denomination. The principle of detection could be magnetic or radar/microwave. It is known that machine-identifiable fibers can be placed in banknote papers. Since the use of uniform product code (UPC) or other coding is prevalent throughout the world, the committee considered the placement of an appropriate optical pattern code on the dollar bills. These codes would be specific to the denomination but not to the individual banknote. Many devices exist to read such codes, and they could form the basis for production of a small hand-held unit for blind people. The initial goal would be denomination sorting, as the committee recognizes that the code could be easily reproduced in a counterfeit manner. However, linkage with some other feature in the note could provide the basis for authentication. Dutch notes have this type of feature. A feature that appears the same on both ends of the banknote would facilitate the design of devices that can denominate independent of note orientation. Examples of symmetrical features are ordinary barcodes that can be scanned in either direction, or a feature, such as a magnetic strip, that was located along the banknote length on the central axis. The committee found that a number of countries throughout the world use fluorescent, magnetic, or infrared reactive inks. These inks could be incorporated in a simple bar code arrangement to allow for basic machine interrogation. These types of codes are used on Scottish banknotes and Eastern Caribbean Central Bank currency. Magnetic bar coding is used on the newer series of U.S. dollar bills and could be used as the basis for device design. In the report based on the study conducted in 1983 by the BEP for U.S. Representative Edward Roybal, the BEP recommended continued development of electronic devices for denominating banknotes as the most effective way at that time to address the needs of the low vision and blind communities (BEP, 1983). The committee was aware that Canada is currently promoting the use of a device to assist blind people in recognizing different denominations. It is estimated that there are almost 8,000
OCR for page 54
--> blind Canadians who cannot benefit from the large numerals and colors on the Canadian banknotes (Bank of Canada, 1994). Canadian currency designs contain colored patches approximately 4.75 mm square that indicate denomination by location and are interpreted by a device to provide denomination information. Since 1990, the Bank of Canada has supplied 4,600 readers (devices) free-of-charge to users through the Canadian National Institute for the Blind. There is a continued demand of about 50-60 readers per month. The readers are manufactured by Brytech Corporation and cost the Bank of Canada approximately $300 (Canadian) each. In a limited poll of users in 1991, the Bank of Canada determined that the majority of users are elderly and that they typically use the readers a few times a week at home, as opposed to carrying the readers with them as they work or shop. The Bank of Canada intends to conduct a more inclusive survey on the uses and usefulness of their readers and will work to improve the devices to make smaller or faster readers. The committee feels that continued development of smaller and faster denomination devices will address the needs of people who are multiply handicapped or cannot otherwise use features available to denominate U.S. banknotes. Devices add a possibility of authentication for blind people, who cannot use the visual counterfeit-deterrent features. A way to help reduce the cost of designing and producing a denomination device is for the BEP to include one or more overt features in the new design that could be read by a device and interpreted for the user. Feature Combinations Combinations of features are likely to improve the ease with which banknotes can be denominated, as discussed in Chapter 2, and may also provide additional assistance in authentication (NRC, 1993). In addition, some combinations of features could improve the ability to distinguish front from back and to orient the face properly. Among the numerous possibilities might be dimensional changes combined with color, holes, or another tactile feature that would provide information regarding front-to-back and face orientation. Large numerals could be combined with color. Those combinations that also permit authentication and permit the inclusion of anticounterfeiting features should be favored. Of the 171 banknote types described in the table in Appendix D, less than 40 percent are denominated using a single cue on each denomination; 28 use color only, and 31 use size only. Over 100 of the issuing authorities use both size and color on all denominations in the series, with some adding a tactile mark, large numerals, or both to increase the number of denominational cues available to visually disabled people. Four of these issuing authorities provide banknotes with color and a tactile symbol, so that there are two denominational cues. The European Blind Union recommends that there be at least two different ways of discriminating among banknotes using two different senses (EBU, 1994). It is difficult to quantify the improvement in denomination ease with additional features, but complementary features add to the difficulty of simulation and might improve banknote security (NRC, 1993).
OCR for page 55
--> Conclusions and Recommendations From the long list of features that would assist in the denomination and authentication of banknotes, some were determined to be applicable immediately, some were dropped, and some would require investigation to determine if they are appropriate or if further research will make them applicable for currency use. Experience with some of the features is already available in other countries (Appendix D). Visual/Tactile Features The committee recommends the variation of length, height, or both dimensions as a key to denomination to address the needs of the broadest population. A secondary cue might provide orientation information. Aside from the use of devices, this feature is the only one that the committee recommends that addresses the needs of blind people. The concept of holes for banknote denomination is attractive, but it is likely further from implementation than size variation. Although it would not be difficult to manufacture banknotes with holes in them, there are a number of issues regarding durability, public acceptance, etc., that require more work before this feature can be recommended for implementation. Visual-Only Features The committee recommends the use of large, high-contrast numerals on a simple, uniform background in the next redesign of U.S. banknotes. Incorporation of this feature would address the needs of a large fraction of the population for quick, easy identification of a banknote's denomination. The committee also recommends the use of different predominant colors for the six denominations printed. This feature addresses a large population as well and could be realized with little increase in unit costs. A third possibility, the use of geometric shapes and coarse patterns, is seen by the committee as a redundant, additional visual clue that might benefit the sighted but is not considered to be essential if the numerals are large enough and have sufficient background contrast. Tactile-Only Features The committee was unable to identify any tactile-only feature that can be recommended without further research. Machine-Readable Features The committee recommends the inclusion of overt features in the new currency design that would encourage the development of devices. Of the features considered, machine-detectable
OCR for page 56
--> fibers, UPC coding, or simple geometric shapes in standard ink that could be read optically are readily available today. These same shapes in fluorescent, magnetic, or infrared inks might offer the possibility of lower-cost or smaller devices. The committee recommends that the Treasury Department and the Federal Reserve Bank work with the device developers to determine where the most improvements can be made. Feature Combinations As discussed in the Weber criteria section of Chapter 2, there are indications that combinations of features will be superior to any one by itself. These combinations tend to improve accuracy and to broaden the application range. Multiple features also may improve the ability to orient or authenticate a banknote. References Abbott, G. 1994. Presentation to the Committee on Currency Features Usable by the Visually Impaired. March 1994. Arditi, A. 1994. Presentation to the Committee on Currency Features Usable by the Visually Impaired. March 30, 1994. Bank of Canada. 1994. Personal communication from Ms. M. Lefebvre-Manthorp of the Bank of Canada. July 8, 1994. Benali, A., and M. Iraqi. 1994. Personal communication to the Committee on Currency Features Usable by the Visually Impaired. May 1994. Bennett, D.G.M. 1994. Personal communication to the Committee on Currency Features Usable by the Visually Impaired. May 1994. Blomberg, T. 1994. Personal communication to the Committee on Currency Features Usable by the Visually Impaired. May 1994. Brabyn, J. A. 1994. Personal communication to the Committee on Currency Features Usable by the Visually Impaired. August 16, 1994. BEP (Bureau of Engraving and Printing). 1983. A Study of Mechanisms for the Denomination of U.S. Currency by the Blind or Visually Impaired. Bureau of Engraving and Printing report to Congressman Edward R. Roybal, Chairman of the Select Committee on Aging, U.S. House of Representatives. Church, S. 1994. Personal communication to the Committee on Currency Features Usable by the Visually Impaired. March 1994. Cruz V., R. 1994. Personal communication to the Committee on Currency Features Usable by the Visually Impaired. June 1994. Durlach N., L. Delhorne, A. Wong, W. Ko, W. Rabinowitz, and J. Hollerback. 1989. Manual discrimination and identification of length by the finger-span method. Perception and Psychophysics 46:29-38.
OCR for page 57
--> EBU (European Blind Union). 1994. A Report of the European Blind Union Expert Working Group on Currency: Recommendations for the Design of the ECU. Paris, France: EBU. Farher, W. 1994. Personal communication to the Committee on Currency Features Usable by the Visually Impaired. May 1994. Gaiteiro, J. 1994. Personal communication to the Committee on Currency Features Usable by the Visually Impaired. May 1994. Hill, B.J. 1994. Presentation to the Committee on Currency Features Usable by the Visually Impaired. March 1994. Jarvis, A. 1994. Personal communication to the Committee on Currency Features Usable by the Visually Impaired. May 1994. Johansen, J. E., and J. Bjørgo. Personal communication to the Committee on Currency Features Usable by the Visually Impaired. May 1994. King, S. 1994. Personal communication to the Committee on Currency Features Usable by the Visually Impaired. May 1994. Koeze, P. 1990. A Large-Scale Experiment on the Effectiveness of the Marks for the Blind on Netherlands Banknotes. Presented at the European Banknote Printers' Conference, Rome. Legge, G.E. 1994. Personal communication to the Committee on Currency Features Usable by the Visually Impaired. August 16, 1994. Miller, O. 1994. Presentation to the Committee on Currency Features Usable by the Visually Impaired. March 1994. Minoshima, Y. 1994. Personal communication to the Committee on Currency Features Usable by the Visually Impaired. May 1994. National Research Council Committee on Vision. 1980. Recommended standard procedures for the clinical measurement and specification of visual acuity. Advances in Ophthalmology 41:103-148. NRC (National Research Council). 1993. Counterfeit Deterrent Features for the Next-Generation Currency Design. National Materials Advisory Board, NRC. Washington, D.C.: National Academy Press. Oschwald, J. and C. Hendricks. 1994. Presentation to the Committee on Currency Features Usable by the Visually Impaired. March 30, 1994. Péterfi, S. 1994. Personal communication to the Committee on Currency Features Usable by the Visually Impaired. May 1994. Santipong, C. 1994. Personal communication to the Committee on Currency Features Usable by the Visually Impaired. May 1994. Sirkis, S. 1994. Personal communication to the Committee on Currency Features Usable by the Visually Impaired. May 1994. Soekarna, E. 1994. Personal communication to the Committee on Currency Features Usable by the Visually Impaired. June 1994. Trachsel, P., and M. Meroni. 1994. Personal communication to the Committee on Currency Features Usable by the Visually Impaired. May 1994. Wallach, B. 1994. Presentation to the Committee on Currency Features Usable by the Visually Impaired. March 1994.
OCR for page 58
This page in the original is blank.
Representative terms from entire chapter: