Maria L. Toro (University of Pittsburgh)
ABSTRACT
This paper describes the design, prototype development and evaluation of a device that is aimed to make card games more accessible for people with visual impairments. It is a hand-held device where the user inserts a card and receives immediate auditory feedback about the card through a set of earphones. The device was designed following a user-centered design methodology; a first prototype was build to evaluate the feasibility of the concept. The first prototype was tested with 5 game players of which one was blind-folded. User feedback indicates that the device is useful and meets the needs of a visual impaired player. Suggestions for improvement included that the appearance and time it takes to insert the card needs to be improved.
BACKGROUND AND SIGNIFICANCE
The human need for recreation is very important in the life of a person with a disability, whether it is a physical, sensory, or cognitive impairment that affects the ability to function in everyday activities and environments (1). Specifically, for persons with an acquired disability, recreation can represent a return to a valued form of daily life activity (1). Furthermore, leisure plays an important role in confidence, psychological and physical health, and contributes to positive social interaction (2). Therefore, the more life is filled with activities that are interesting to a person, the higher their quality of life [(2), (3)].
Different approaches and solutions have been proposed to make games accessible for people with visual impairments. Most common in current literature are solutions that are based on computer software applications. For instance, Puret et al (4), developed a multimedia game for children with visual impairments. It can be used without the help of a sighted person and uses tactile and sound feedback during the game so the child is aware of what is happening in the game. Grammenos et al (5) developed a computer-based game with auditory input, which is accessible to people with different types of abilities, including those with visual impairments. Lieberman et al (6) described that for people with a hearing and visual impairment, card and board games are not popular recreational activities because of a lack of sensory feedback for the player. On the other hand, the American Foundation for the Blind (7), suggests using cards with Braille, large print cards or the use of card holders so that sighted players cannot see the visually impaired player’s cards, in order to make card and board games accessible to this population. Timanus (8) also suggests using Braille cards.
Character recognition devices are available on the market for people who are visually impaired. These devices allow the user to scan or take a picture of written information and convert it into audio playback or magnify the information. However these devices do not recognize images or colors and are therefore not useful to provide users with reliable card information when playing card games because it only tells the card letter or number but not the suit of the cards (hearts, clubs, diamonds).
The National Foundation for the Blind (9) states that merely 10% of people who are legally blind can read Braille. Therefore, 90% of the people who are blind do not have access to card games because the only accessible solution that is commercially available is Braille cards. According to the Online Gambling newspaper (10), there are around 954 Casinos in the US, of which Nevada has around 314 casinos according to the World Casino Directory (11).
STUDY OBJECTIVE
This project aims to design and develop a solution in the form of a universal card reader that can be used by a visually impaired player. The card reader will read one card from a regular deck of cards and translates this card into an audio signal to aid the player during the card game.
METHODS/APPROACH
A user centered design methodology was followed for this project. This is an iterative process which ultimate goal is to develop usable systems, achieved through involvement of potential users of a system in the design process (12).
Needs Assessment
Seven people with visual impairments were interviewed with the aim of identifying and understanding their needs when they play tabletop games such as card games. Table 1 lists the interview questions.
TABLE 1. Questions asked to potential end users about how they access card games
| Questions |
| 1) What needs do you have in playing games? |
| 2) Are there any games that you wish you could play but can’t? Why don’t you play them? |
| 3) What are your favorite games? How do play them? Is there anything that you could improve in them? |
| 4) Have you ever played cards? If not, why can’t you play? Would you like to have sound feedback about the play during the game? |
| 5) Do you know what games are played with Braille cards? |
| 6) Would you be willing to use a device in your ear that gives constant feedback about what is happening on the table during the game? |
All of the interviewees were trained to use Braille. So in general, interviewees responded that there are Braille cards which supply the need with the cards on hand, but stated the need for additional feedback on those cards that are being played on the table (by other players). Interviewees responded that any card game can be played if the cards have Braille. Ways in which people play cards currently, is to have a sighted player speak out loud which cards are being placed on the table by other players, or each player identifies the card(s) they are putting on the table. Five of the seven interviewees said that speaking out loud and identifying each card that is being place on the table is uncomfortable for other players and for themselves. Additionally, one of the interviewees replied that hearing the identification of a card only once is difficult for elder players because of the difficulty in remembering what was played or said. Finally, if cards have Braille, visually impaired players can always feel the cards to see which ones are placed on the table. However, touching cards that are already on the table could be perceived as obtrusive to players of the game or could not be permitted (e.g. in Casino’s).
Quality Function Deployment
After interviewing people who may benefit from this project, a Quality Function Deployment Matrix was used to translate user needs, market research and technical data into an appropriate set of engineering targets (technical requirements) to be met by the product solution (13). Then a brainstorm session was done to generate different ideas for each identified consumer need in the QFD matrix. After sketching each idea on paper, pros and cons were identified followed by a Pugh chart analysis to select the best idea. The best idea was evaluated upon feasibility by an actual card game trial with a group of 6 people including one user with low vision. This solution featured a handheld device with image recognition capabilities that converts each card suit and number into a spoken output signal that is directed to the visually impaired player through a set of earphones.
DESIGN
A prototype was built with a readily available webcam (Logitech QuickCam® Orbit 2 mega pixels), a custom-made housing with a card slot and space for the webcam, a laptop computer with programming software Matlab (v. 7.9.0) and a set of earphones for voice output. The housing was designed using Solid Works® 2009 and a physical model was built through Selective Laser Sintering (SLS). The Matlab program was used to capture and recognize the images taken from the webcam. The prototype cost included SLS prototyping at $50.4, a webcam at $129.99 and ear-phones at $20.
First the webcam captures a video stream of a section of the card, then a snapshot is taken of the video stream and this image is processed. A template matching procedure in Matlab is performed where the acquired image is compared with each of the 52 card templates using a Matlab correlation function. The comparison between cards that has the greater correlation is selected as the correct card and a card code is retrieved. Then, the card code is entered into a voice output function that reads a .wav file that corresponds with the card code (i.e: “Ace of Clubs”, “Eight of Diamonds”). Figure 1 shows the signal processing block diagram.
FIGURE 1. System block diagram
EVALUATION
The prototype card reader was evaluated by two blind-folded users and four sighted players. The game “up and down the river” was played with five cards for each player. Each user inserted five cards in the card slot and an administrator pressed the “run” button in the Matlab program to read the dealt cards. Other players at the table called their card while they placed it on the table. The visually impaired user inserted each dealt card in the reader when they were dealt and also before playing a card to check if it was indeed the card that he wanted to place on the table. Both the visually impaired user and the sighted players were asked to answer questions regarding the performance of the device regarding its usability, speed and recommendations.
Findings included that the device did not allow or quick use because both blind-folded users found it difficult to detect the slot and insert the card into it. The blind-folded players also had a difficulty knowing when it was their turn to play a card. Additional to this, it was suggested that the device had a memory aid to remember those cards that have been read already. However, another suggestion was that if the use of the device was faster, a memory aid wouldn’t be necessary as the cards could be read continuously throughout the game. It was also recommended to have the voice output of the card be repeated at least twice in order for the player to be more confident about the type of card that was scanned. Sighted users also commented that they forgot to “call” the card they were playing. All of the interviewees agreed that the size of the prototype needed to be reduced.
Regarding the software, the template matching was done with only one template per card. To improve the accuracy of the reader and software matching process and to allow for slight shifting of inserted cards relative to the position of the camera, additional templates can be recorded (shifted and rotated cards).
Figure 2 shows an improved prototype design based on the users’ feedback and recommendations. This device will use an ultra-compact camera module that cost around $50 (TOSHIBA America electronics components, 2006) to reduce its size.
FIGURE 2. Improved design based on user feedback. Includes an easier to feel slot, a small size, improved in portability.
ACKNOWLEDGEMENTS
This project was done as part of HRS/BioEng 2703 at the Department of Rehabilitation Science and Technology. Prototyping was sponsored by the Quality of Life Technology Engineering Research Center, National Science Foundation (EEC-0540865). The author also expresses thanks to Dr. Linda van Roosmalen, Dr. David Brienza and the students and volunteers who gave input during the project.
REFERENCES
1. Buning, M.E., Rice, I.M., Cooper, R.A., Fitzgerald, S.G. (2007). Adaptive Sports and Recreational Technology. In Cooper, R.A., Ohnabe, H., Hobson, D.A. An Introduction to Rehabilitation Engineering (pp. 385-412). Location: Taylor & Francis, New York.
2. Aitchison, C. (2003). Disability Leisure: developing data, definitions and discourses. Disability and Society, 18(7), 955-969.
3. Keilhofner, G. (1985). A Model of Human Occupation: Theory and Application. Location: William & Wilkins, Baltimore.
4. Puret, A., Archambault, D., Monmarché, N., & Slimane, M. (2006). A simple game generator for creating audio/tactile games. Technology and Disability, 18, 227-236.
5. Grammenos, D., Savidis, A., & Stephanidis, C. UA-Chess: A Universally Accessible Board Game. Available online at: http://dmt.fh-joanneum.at/kd3/objects/application_pdf/1650_UA_accessible_chess_game.pdf
6. Liebernam, L., Stuart, M. (2002). Self-determined Recreational and Leisure Choices of Individuals with Deaf-Blindness. Journal of Visual Impairment and Blindness, 96(10), 724-735.
7. Game Night. Retrieved from American Foundation for the Blind website: http://www.afb.org/seniorsite.asp?SectionID=67&DocumentID=3284
8. Timanus, E. (2002). Modifying Games for the Blind. Retrieved from The Games Journal website:
http://www.thegamesjournal.com/articles/GamesForTheBlind.shtml
9. National Federation of the Blind. (2009). The Braille Literacy Crisis in America. Retrieved from National Federation of the Blind web site: http://www.nfb.org/images/nfb/documents/pdf/Braille_Literacy_Report_web.pdf
10. Online Gamblind News. (2005). http://www.ogpaper.com/casinos/UnitedStates.html
11. World Casino Directory. (2008). http://www.worldcasinodirectory.com/nevada-casinos/casino-list5
12. KARAT, J. 1996, User Centered Design: Quality or Quackery?, in the ACM/SIGCHI magazine, Interactions July-August 1996.
13. Crow, K. (2004). Quality Function Deployment: What, Why, and How. Retrieved from DRM associates website http://www.npd-solutions.com/whyqfd.html.
Maria L. Toro,
Department of Rehabilitation Science and Technology,
University of Pittsburgh, Pittsburgh, PA