Embed Size (px)
Citation preview
Jeffrey P. BighamRichard Ladner, Ryan Kaminsky, Gordon Hempton, Oscar Danielsson
University of WashingtonComputer Science & Engineering
Browsing while blind Screen readers
Images cannot be read
W3C accessibility standards “Provide a text equivalent for every non-text
element”
What if no alternative text? Nothing Filename (060315_banner_253x100.gif) Link address (http://www.cs.washington.edu)
nav_svcs.gif
Outline
Web Studies
Providing Labels
WebInSight System
Evaluation
Future Work
Web Studies: All Images != Significant images need alternative
text Informative alt, title, and longdesc HTML attributes
Insignificant images need empty alt text
Automatic Determination?
<img src=“graph.gif” alt=“sales graph” title=“sales graph” longdesc=“sales_descrip.txt”>
<img src="images/spacer.gif" width="1" height="1">
More than one color AND both dimensions > 10 pixels An associated action (clickable, etc.)
<img src="images/spacer.gif" width="1" height="1“ alt=“”>
Web Studies Previous studies
img tags with defined alt attribute: 27.9%[1], 47.7%[2], and 49.4%[2]
Significant images have a defined alt attribute?
76.9%[3]
Gaps Some Ignore Image Significance Some Ignore Image Importance
[1] T. C. Craven. “Some features of alt text associated with images in web pages.” (Information Research, Volume 11, 2006).
[2] Luis von Ahn et al. “Improving accessibility of the web with a computer game.” (CHI 2006)[3] Helen Petrie et al. “Describing images on the web: a survey of current practice and prospects for
the future.” (HCII 2005)
University of Washington CSE Department Traffic
Web Studies
Significant images without alternative text.
Significant images withalternative text.
~1 week 11,989,898 images. 40.8% significant 63.2% alt text
Study ResultsGroup
Significant
Pages > 90%
Pages Images
High-traffic 39.6% 21.8% 500 32913
Computer Science
52.5% 27.0% 158 4233
Universities 61.5% 51.5% 100 3910
U.S. Federal Agencies
74.8% 55.9% 137 5902
U.S. States 82.5% 52.9% 51 2707
Percentage of significant images provided alternative text, pages with over 90%of significant images provided alternative text, number of web sites in group,and number of images examined.
Outline
Web Studies
Providing Labels
WebInSight System
Evaluation
Future Work
<a href=“p234.htm”><img src=“p523.gif”></a><a href=“p234.htm”><img src=“p523.gif” alt=“People of UW”></a>
Providing Labels: Context Labeling Many important images are links
Linked page often describes image What happens if you click
<html><head><title>People of UW</title><body><h1>People</h1>…</body></html>
[4] Jain et al. “Automatic text location in images and video frames.” (ICPR 1998)
Providing Labels: OCR Labeling
Improvement through Color Clustering[4]
ColorNew Image
Text Produced
,, ., ,,,n
Register now!
(Optical Character Recognition)
Improves recognition 25% relative to base OCR!
Providing Labels: Human Labeling
Humans are best Recent games compel accurate labeling WebInSight database has over 10,000 images Could do this on demand
[5] Ahn et al. “Labeling images with a computer game.” (CHI 2004)[6] Ahn et al. “Improving the accessibility of the web with a computer game.” (CHI 2006)
[5] [6]
Outline
Web Studies
Providing Labels
WebInSight System
Evaluation
Future Work
WebInSight System
Tasks Coordinate multiple labeling sources Insert alternative text into web pages Add code to insert alternative text later
Features Browsing speed preserved Alternative text available when
formulated Immediate availability next time
The Internet
The Internet
Proxy
Context Labeling
OCR Labeling
Human Labeling
Database
Blind User
Outline
Web Studies
Providing Labels
WebInSight System
Evaluation
Future Work
Evaluation
Measuring System Performance WebInSight tested on web pages from web
studies Used Context and OCR Labelers Labeled 43.2% of unlabeled, significant images Sampled 2500 for manual evaluation 94.1% were correct
Proper Precision/Recall Trade-off
Evaluation: Demo
Conclusion
Lack of alternative text is pervasive
WebInSight calculates alternative text
WebInSight inserts alternative text
High precision and moderate recall
Future Work
User Studies What do users want? How can we provide
it? Maintain experience.
Users Content Producers
User Studies Designer motivation.
Tools for Web Design People can always be
better Adapt user techniques
Common Themes
Technology Improved labeling Bring closer to user Move beyond images
More challenges Content Structure Dynamic Content Web applications
WebInSight
http://webinsight.cs.washington.edu
Thanks to: Luis von Ahn, Scott Rose, Steve Gribble and NSF.
