[4]Jain et al. “Automatic text location in images and video frames.”
(ICPR 1998)
Providing
Labels: OCR Labeling
Register
now!
,,., ,,,n
Text
Produced
New
Image
Color
Improvement
through Color Clustering[4]
(Optical
Character Recognition)
Improves recognition 25%
relative to base OCR!
The second method is OCR
Labeling.Many images on the web
contain text, but screen readers cannot directly use it in its graphic
form.Optical Character Recognition
works well when applied to black text on white backgrounds scanned in at a
known resolution, but can have problems with web images.To label images using OCR, we first
preprocess the image to convert it to a resolution and format that the OCR
prefers.Next, following a process
similar to that developed by Jain et al. we convert it to several black &
white versions by highlighting major color groups.The OCR can more easily handle these black
and white images.
The example here shows that
for this image the OCR originally produces no text.Our method produces six new black &
white images.The OCR produces no
output for 4 of the 6 new images, garbage text for one, and correct text for
the last.We verify the output of the
OCR using a web-based spelling checker.