PRE2017 3 Groep2

From Control Systems Technology Group
Revision as of 14:06, 18 February 2018 by S157223 (talk | contribs)
Jump to navigation Jump to search

0LAUK0: Robots Everywhere Group 2

Group members

Project definition

A huge amount of information is communicated to people through text. Many of that is not obtainable for fully blind or near-blind people. It is possible to make a large portion of this information accessible by designing a portable device to assist the blind. This device would be able to scan text and present it in braille code. The goal is to make it easy for the blind to read books, hand written texts and text contained within images. A hypothetical concept for this device will be designed for fully blind people.

Approach

At first literature will be studied to learn about the current technological aids for the blind and for the current state of OCR (optical character recognition) and refresh-able braille displays. Then contact with a institute, involved in research for helping the blind, will be sought to gain feedback on the device. This feedback will then be carefully incorporated into the device to ensure its practical use.

Results from literature research

After conducting a vast literature research, a brief summary of all articles and patents deemed relevant has been made. These summaries are listed below.

A BRAILLE O.C.R. FOR BLIND PEOPLE[1]
In this paper the principles for an optical character recognizer, OCR, for written braille code are developed. Several image processing techniques are incorporated such as adaptive thresholding and skew angle detection. The paper was sponsored by the National Organization of Spanish Blind People.

OCR-OPTICAL CHARACTER RECOGNITION[2]
In this paper a simple, efficient and less costly OCR for regular text is presented. The OCR uses a database to recognize the characters to lower the computational cost of the program. The paper discusses various stages of the program including preprocessing and post processing.

Portable Braille Computer Device[3]
This patent describes a portable device that can scan text using a scanner, interpret the scanned text using an OCR program and present it using a braille display. The patent also claims to provide a complete information management system within the device. Such as a personal agenda and notes.

Wearable assistive devices for the blind[4]
This paper reviews important steps made in the field of wearable assistive devices for the blind. From these important steps universal design concepts for these wearable devices, and also other systems for the blind, are extracted. The future potential of various prototypes are also discussed.

A Blind Person’s Interactions with Technology[5]
This paper considers a blind person's interaction with various technologies. The technologies are discussed both practically and socially, since both are important when using it in one’s daily life. Findings about the interaction with technology of a person blind from birth are presented.

An interactive and multi-functional refreshable Braille devicefor the visually impaired[6]
The authors of this paper have developed a book reader with 96 braille cells, and also includes to option to convert text to audio. The articles describes in depth the design of the device, including all the buttons and their functions. It uses a scanner which sends information to the device via USB or Bluetooth, which then converts the information to braille. Tests done by the authors with visually impaired people that can or cannot read braille indicate a high usability of this device. Overall, this device is very similar to our initial conception.

Methods for presenting braille characters on a mobile device with a touchscreen and tactile feedback.[7]
In order to circumvent the high cost of a mechanical braille reader (usually in the range of 5000 to 15.000 USD, according to the authors), different methods for implementing braille reading in a mobile device were researched. The authors identified three different ways of presenting braille characters, and showed the feasibility of all methods through experiments in which participants were asked to identify a single character. Both the accuracy and reading time varied, and the authors stated that further research would be required to find out how to present words or sentences.

Determining the optimum font size for braille on capsule paper.[8]
This paper presented research on what is the optimal size of a braille front. The results showed that there was an optimum size, which gave the highest reading speed and the lowest error. The optimal size showed little variation for different age groups. The most important conclusion to be drawn is that enlarging the font size doesn’t necessarily improve either reading accuracy or reading time.

Camera Reading for Blind People.[9]
This article describes the foundation for a technique that should enable blind people to photograph a piece of text using a mobile device, and subsequently have that device read out the text. The paper present an overview and state of art (from 2014) of Optical Character Recognition (OCR) and Text-To-Speech (TTS) software.

A Review on Optical Character Recognition Techniques.[10]
This paper describes more in depth the process of OCR. The steps through which an OCR protocol goes are subdivided into digitization, pre-processing, segmentation and feature extraction.

Computerized microfluidic cell culture using elastomeric channels and Braille displays.[11]
This article describes a new computer-controlled microfluidics system used on a refreshable Braille display. It is a grid of 320 vertical pins to depict the braille. Due to a new method used to power the integrated pumps and valves, controlling the pins, in the silicone rubber the system is a lot more efficient. More pumps and valves can be put in a smaller space improving the display immensely.

Optical Character Recognition for Handwritten Cursive English characters[12]
This paper talks about OCR for handwritten text. To improve OCR for handwritten text a new method is developed with noise cancelling. A median filter is applied to filter out unwanted scribbles to improve the results. To get the results the following steps are used: image acquisition, image preprocessing, segmentation, feature extraction, recognition.

Real-time scene text localization and recognition[13]
An article about text recognition. It discusses a method for real-time text recognition on images for example. It creates an Extremal Region (ER), a box, around the text region to include the entire text without missing anything, while not including areas where no text is present. The system is tested by two methods. The ICDAR 2011 and the Street View Text dataset.

A system for converting print into Braille[14]
The paper goes over different methods of braille that are in use. It adds certain language rules to make converting text to braille easier. It does however also state that the rules should be developed further.

REFRESHABLE BRAILLE DISPLAY SYSTEM[15]
This is a patent of a refreshable braille display system. It is a display that can extend and retract dots to form words or short sentences. It describes a microelectromechanical device.

Voice Assisted Text Reading System for Visually Impaired Persons Using TTS Method[16]
In this article the TTS method is elaborated. A finger mounted camera is used to capture the text image from the printed text and the captured image is analyzed using OCR. A predefined dataset is loaded in order to match the observed text with the captured image. Once it is matched the text is synthesized for producing speech output. MATLAB is used for performance analysis.

Eight-dot Braille[17]
Elaboration on the use of 8 dot braille. It can be used to give more information per cell and it opens a way to produce a wider variety of symbols. It might be used to determine the users location.

FingerReader: A Wearable Device to Support Text-Reading on the Go[18]
The article describes a finger worn device with a camera which allows the user to skim through a page. This camera records the text and sends the images to a computer chip to convert the text into speech. It uses fibration to correct the users movement.

Character recognition — A review[19]
This article elaborates on some of the OCR techniques. Due to the age it might not be the most relevant information.

Design and Implementation of OCR to identify English Characters and Numbers[20]
This article describes the matrix matching method of OCR. According to the article there are two distinct methods for OCR: Matrix Matching in which the system compares the scanned characters with the library character matrices. This system works best when the characters to be scanned and the library characters have very little or no variation in style. Feature Extraction: Feature Extraction generally deals with the features of characters like shape, closed areas, diagonal lines, line interaction and curves. It is more effective and flexible methods as it has a wide scope to identify the same character with different shapes and dimensions. It also gives a short description of how a Matrix Matching OCR is build up.

Display of virtual braille dots by lateral skin deformation: feasibility study[21]
This article describes an attempt to make a display for blind people in braille. It explains how lateral skin deformation works and how it can be applied to a keyboard.It gives a good guideline on how certain things work and how they can be used to benefit reading braille.


Research

Solution

Discussion

Conclusion

References

  1. Hermida, X. F., Rodríguez, A. C., & Rodríguez, F. M. (2008). A BRAILLE O.C.R. FOR BLIND PEOPLE
  2. Verma, A., Arora, S., & Verma, P. (2016). Ocr-Optical Character Recognition. 7th International Conference on Recent Innovations in Science, Engineering and Management, 230–240.
  3. Kahn, S. (2003). PORTABLE BRAILLE COMPUTER DEVICE. doi: 10.1126/science.Liquids
  4. Velazquez, R. (2010). Wearable assistive devices for the blind. Lecture Notes in Electrical Engineering, 75 LNEE, 331–349. doi: 10.1007/978-3-642-15687-8-17
  5. Tenenberg, J. (n.d.). A Blind Person’s Interactions with Technology. COMMUNICATIONS OF THE ACM, 52 (8). doi: 10.1145/1536616.1536636
  6. Fatih Basciftci, A. E. (2016). An interactive and multi-functional refreshable Braille device for the visually impaired. Displays, 41 , 9.
  7. Rantala, J., Raisamo, R., Lylykangas, J., Surakka, V., Raisamo, J., Salminen, K., Hippula, A. (2009). Methods for presenting braille characters on a mobile device with a touchscreen and tactile feedback. IEEE Transactions on Haptics, 2 (1), 28–39. doi: 10.1109/TOH.2009.3
  8. Watanabe, T. (2014). Determining the optimum font size for braille on capsule paper. IEICE Transactions on Information and Systems, E97-D(8), 2191–2194. doi:10.1587/transinf.E97.D.2191
  9. Neto, R., & Fonseca, N. (2014). Camera Reading for Blind People. Procedia Technology, 16 , 1200–1209. Retrieved from http://linkinghub.elsevier.com/retrieve/pii/S2212017314003624 doi: 10.1016/j.protcy.2014.10.135
  10. Modi, H., Scholar, P. G., & Parikh, M. C. (2017). A Review on Optical Character Recognition Techniques. International Journal of Computer Applications, 160 (6), 975–8887. Retrieved from http://www.ijcaonline.org/archives/volume160/number6/modi-2017-ijca-913061.pdf
  11. Gu, W., Zhu, X., Futai, N., Cho, B. S., & Takayama, S. (2004). Computerized microfluidic cell culture using elastomeric channels and Braille displays. Proceedings of the National Academy of Sciences, 101 (45), 15861–15866. Retrieved from http://www.pnas.org/cgi/doi/10.1073/pnas.0404353101 doi: 10.1073/pnas.0404353101
  12. Aparna, A., & Muthumani, P. I. (2014). Optical Character Recognition for Handwritten Cursive English characters. International Journal of Computer Science and Information Technologies, 5 (1), 847–848.
  13. Neumann, L., & Matas, J. (2012). Real-time scene text localization and recognition. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 3538–3545. doi: 10.1109/CVPR.2012.6248097
  14. Blenkhorn, P. (1997). A system for converting print into Braille. IEEE Transactions on Rehabilitation Engineering, 5 (2), 121–129. doi: 10.1109/86.593266
  15. Schmidt, R. (2002). REFRESHABLE BRAILLE DISPLAY SYSTEM. Patent no: US 6,354,839 B1
  16. Sanjana, B., & RejinaParvin, J. (2016). Voice Assisted Text Reading System for Visually Impaired Persons Using TTS Method. , 6 (3), 1–5. doi: 10.9790/4200-0603031523
  17. Dixon, B. J. (2007). Eight-dot Braille. (September).
  18. Roy Shilkrot, J. H., Connie K. Liu, P. M., & Nanayakkara, S. (2014). FingerReader: A Wearable Device to Support Text-Reading on the Go. In Proceedings of CHI ’14 Extended Abstracts on Human Factors in Computing Systems(Vi). doi: 10.1145/2559206.2581220
  19. Govindan, V. K., & Shivaprasad, A. P. (1990). Character recognition — A review. Pattern Recognition, 23 (7), 671–683. Retrieved from http://www.sciencedirect.com/science/article/pii/003132039090091X doi: 10.1016/0031-3203(90)90091-X
  20. Adhvaryu, R., Parikh, R., & Vora, K. (2018). Design and Implementation of OCR to identify English Characters and Numbers. , 4 (2), 57–62.
  21. Lévesque, V., Pasquero, J., Hayward, V., & Legault, M. (2005a). Display of virtual braille dots by lateral skin deformation: feasibility study. ACM Transactions on Applied Perception, 2 (2), 132–149. Retrieved from http://portal.acm.org/citation.cfm?doid=1060581.1060587 doi: 10.1145/1060581.1060587

Coaching Questions Group 2