PROJECT PROPOSAL

E-COMMERCE WEBSITE FOR VISUALLY IMPAIRED/BLIND PEOPLE

In recent years online shopping has grown significantly worldwide. As the technology advanced, new techniques
such as (HTML 5 and Flash-based content, JavaScript etc.) are used in e-commerce websites to visually present
information. However, these new techniques have accessibility problems for people with disabilities (visually
impaired/blind) when accessed using assistive technology. Therefore, it is also important to adopt web accessibility
such as the Web Content Accessibility Guidelines (WCAG 2.0) in business-to-consumer (B2C) e-commerce
websites to increase the consumer’s satisfaction of all ages and with disabilities. This study analyses 30 Australian
B2C websites in accordance to WCAG 2.0 using an automated web service. The result shows that B2C websites in
Australia are not paying attention to web accessibility for people with disabilities. However, e-commerce will
succeed in meeting WCAG 2.0 by making B2C e-commerce websites accessible to consumer of all ages and with
disabilities. Recommendations are proposed in order to improve web accessibility for people with sensory (hearing
and vision), motor (limited use of hands) and cognition (language and learning) disabilities in B2C e-commerce
websites.

1. INTRODUCTION

E-Commerce, also known as e-Business, or electronic business, is simply the sale and purchase of services
and goods over an electronic medium, like the Internet. It also involves electronically transferring data and
funds between two or more parties.
There are around 161 million people who are visually impaired people and around 36 million are blind.
These people are taking help from family member or friends to each and every for their needs. Consider a
blind individual who lives alone, but wants to make a purchase from an ecommerce site. He doesn’t have
anybody there by his side, in the moment, to either guide him on what to do on the site, or make the purchase
for him. It is a difficult for blind people to choose clothes with different colors or they find difficult to online
shopping. A system is developed to assist blind people that automatically recognizing clothing patterns and
colors. The system processes the voice output, and asks for the input by pressing the button of their desire
choices. This will help them to buy their desire things online. Every kind of instruction will be given them in
mode of audio they need to hear and follow the instructions.
FAIZ .M. Hasanuzzaman developed a system to automatically recognize banknote of any currency to assist
visually impaired people. This is also a camera based computer vision technology. This system has features
like high accuracy, robustness, high efficiency, ease of use. This system is robust to conditions like
occlusion, rotation, scaling, cluttered background, illumination change, wrinkled bills, and also eliminating
 false recognition and can guide the user to properly and correctly focus at the bill to be recognized using
speed up robust features (SURF). Dimitrios Dakopoulos and Nikolous developed a vision substitution
system for travel aid for blind people. Out of the three main categories of navigation systems (Electronic
Travel Aids, Electronic Orientation systems, Position Locator Aids) here the focus is on Electronic Travel
Aids. Yuan et al. (2007), proposed a system to help the blind people to match clothes from a pair of clothing
images. This system can favor a user with the information about identify the clothing patterns and colors
match. But, this system is still not able to automatically identify clothing patterns.
Multi-fractal spectrum (MFS) proposed by Xu et al. (2007). Density functions and orientation templates are
used to group the combined fractal dimensions of pixel sets. In order to make representations of texture more
strong to 3-D image transformations and changes in the illumination, many of the recent techniques provides
information on extracting local image features. Clustering the extracted local features generates the text on
dictionary. However; properties of an image in different aspects are captured by multiple features. If
different features are highly interrelating, their combination will enhance the feature representation.
Lazebnik et al. introduced a texture representation method depends on affine in variant descriptors and
detectors (RIFT and SPIN). Zhang et al. also linked the scale invariant feature transform (SIFT) and SPIN
for texture classication. An exhaustive comparative study of existing accessible shopping systems for blind
and visually impaired people was carried out by Kulyukin and Kutiyanawala in 2010. Shop Talk (2007) is a
wearable solution that requires the user to carry a barcode scanner and a UMPC in a backpack. Verbal route
instructions are issued through a headphone connected to the UMPC at the blind person’s backpack.
Although the supermarket does not need to install and maintain any hardware, the system requires access to
the supermarket’s inventory control system and binding of product barcodes into supermarket locations so
that guiding can be accomplished. The advantage of Blind Shopping is that it only demands a lightweight
Smartphone equipped with a camera to read QR Codes attached to shelf sections and the very blind person’s
white cane enhanced with an off-the-shelf RFID reader. The Tinetra (Nicholson et al., 2007) project offers
the possibility of detecting products via a barcode or RFID reader, and then it obtains related information via
GPRS from the server. How- ever, it does not include a guiding system as Blind Shopping. Similarly to us
they use Baracoda’s Pencil2 to scan barcodes and IDBlue to scan RFID tags.
2. LITERATURE REVIEW
E-Commerce websites are online portals that facilitate online transactions of goods and services through
means of the transfer of information and funds over the Internet. In the early days, e-commerce was done
partially through emails and phone calls. Now, with a single website, anything and everything that a
transaction needs can be executed online.

2.1 Types of E-Commerce websites

Different ecommerce websites are labeled or referred to differently, based on the function they fulfill.
● Business-to-Business (B2B): Electronic transactions of goods and services between companies. Example:
A business sells SAS products to other businesses.
● Business-to-Consumer (B2C): Electronic transactions of goods and services between companies and
consumers. Example: You buy a new t-shirt from an online store.
● Consumer-to-Consumer (C2C): Electronic transactions of goods and services between consumers,
mostly through a third party. Example: You sell your old Smartphone on eBay or Olx to another
consumer.
● Consumer-to-Business (C2B): Electronic transactions of goods and services where individuals offer
products or services to companies. Example: A Social media influencer offers exposure to their online
audience in exchange for a fee.
2.2 Advantages

● The proposed system will save significant amount of time and effort invested by the university

● This system is user friendly.

❖ Limitation

● Data need to be entered properly otherwise, outcome may won’t be accurate.

❖ Application

● This system can be used by the multiple peoples to get the counseling sessions online.

3. PROBLEM STATEMENT
New technology and techniques are implemented in Ecommerce websites to visually present information
due to significant growth in shopping worldwide. But these techniques have accessibility problem for
people with disabilities. Hence, it is also important to develop ecommerce websites which will provide
customer satisfaction of all ages with disabilities. There are around 161 million people who are visually
impaired and around 36 million are blind. It is difficult for blind people to choose clothes with different
colors hence find difficulty to do online shopping. An ecommerce website is developed to assist blind
people that automatically recognize clothing patterns and colors. This system processes the voice output,
and asks for the input by pressing the button of their desire choices which will help them to buy their
desired products online. All kind of instructions will be provided to them in the form of audio, they need to
hear and follow the instructions.
4. OBJECTIVES/OUTCOMES OF PROJECT
The proposed system is “Ecommerce based online Shopping for Visually Impaired/blind People using Speech
Recognition”. The main aim of this project is to enable the handicapped specially visually impaired peoples to
access the ecommerce based platform which are most widely used for online purchasing of items or products
now a days. But it is quite challenging task for visually impaired people to use this ecommerce based
framework for online shopping. Hence in order to allow handicapped people specially visually impaired
peoples to use this ecommerce online shopping site for purchasing of products without any limitations we are
developing an ecommerce based application using speech recognition to eliminates or reduces the various
practical difficulties that the visually impaired peoples come across.

4.1 Modules:
The system comprises of 2 major modules with their sub-modules as follows:
4.1.1 Admin
● Login: Admin can log in by using credentials.
● Manage Products: Admin can manage products by adding, updating or removing it.
● View Users: Admin can view the users
● View Orders: Admin can also view the order placed by users

4.1.2 User

● Registration: User need to register to obtain credentials.

The waterfall Model is a linear sequential flow. In which progress is seen as flowing steadily downwards (like
a waterfall) through the phases of software implementation. This means that any phase in the development
process begins only if the previous phase is complete. The waterfall approach does not define the process to go
back to the previous phase to handle changes in requirement. The waterfall approach is the earliest approach
that was used for software development.
.Net

❖ Hardware Requirement:

⮚ i3 Processor Based Computer or higher

❖ Software Requirement:

⮚ Windows 7 or higher
⮚ Shopify
⮚ Sublime Text
⮚ Notepad++
⮚ Google Chrome Browser
6. PROJECT TIMELINE

7. CONCLUSION
There are many different segments available for online stores, which are effected by constant evolutions. To
meet the new challenges and requirements of the market, the business owners need to think and make better
decisions. We are developing a system for visual impaired/blind people so that they can do online shopping.
 The system accept speech convert it into text format and display the desire result. This will help blind people
to lead an independent life. In this course project, typical deep learning algorithms, including deep neural
networks (DNN), and deep belief networks (DBN) have been implemented for speech recognition, Global,
local feature are used for identification and multilevel clustering is used to increase the accuracy of system.
Also the performance evaluation of the assistive can validate the resulting performance of the system.
8. BIBLIOGRAPHY
[1] C. Basdogan, M. A. Srinivasan: Haptic rendering in virtual environments. In K. M. Stanney (editor):
Handbook of Virtual Environments: Design, Implementation, and Applications. 117–134. Lawrence
Erlbaum Associates, Mahwah, New Jersey, 2002.
[2] J. P. Bigham, C. M. Prince: Webany where: A Screen Reader On-the-go. In E. Pontelli, S. Trewin
(editors): ASSETS’07. Proceedings of the Ninth International ACM SIGACCESS Conference on Computers
and Accessibility, Tempe, Arizona, USA, October 15- 17, 2007. 225–226, ACM Press, New York, NY,
USA, 2007.
[3] J. P. Bigham, C. M. Prince, R. E. Ladner: Webanywhere: A Screen Reader On-the-go. In D. Sloan, Y.
Yesilada (editors): International Disciplinary Conference on Web Accessibility Beijing,China–April 21–22,
2008. 73–82, ACM Press, New York, NY, USA, 2008.
[4] M. Chiousemoglou, H. Jurgensen: Setting the table for the blind. In ¨ Proceedings of the Fourth
International Conference on Pervasive Technologies Related to Assistive Environments, PETRA 2011, 25–
27 May, 2011, Crete, Greece. 8 pp., ACM Press, New York, 2011. Published on CD.
[5] J. Dolson. Designing Accessible Navigation, November 2013. http://www.
joedolson.com/articles/2013/07/designing-accessible-navigation/.
[6] Top 10 Best Flash Websites of 2013. eBizMBA, Inc., October 2013. http://www.
ebizmba.com/articles/best-flash-sites.
[7] Top 10 Best HTML5 Websites of 2013. eBizMBA, Inc., October 2013. http://www.
ebizmba.com/articles/best-html5-websites.