A low pre processing system for low quality QR code

A low pre-processing system for
restoring low quality QR code
images
Presenting by-
Shameed C M
Roll No:49
S7 CS
Reg No:12123852
College of Engineering Attingal

Contents
 Introduction
 What is a QR code ?
 History of QR code
 Bar code VS QR code
 Inside QR codes
 What can a QR code do?
 Application of QR code
 Pre-processing
 QR code decoding
 Conclusion
 References

Introduction
 2D barcodes are the mainstream of recognition technologies used
for mobile marketing.
 Easily conversion of physical printed data to digital data
 2D barcodes are called two-dimensional because their information
is encoded in two directions, both horizontally and vertically.

WHAT IS A QR CODE ?
QR code: Quick Response
code
Is a 2-D bar code
Used to encode
information(URLs, automated
SmS, text data) in two
dimensional space.

HISTORY
Created by a Japanese
company- Toyota subsidiary
Denso-wave in 1994.
Used to track parts during vehicle
manufacture

Barcodes V/S QR Codes
 Stores data in two
dimensions unlike the 1-D
barcodes.
 High storage capacity in
less space.
 Can be scanned by any
smart phone with the
app installed.

Inside QR codes
 Finder Patterns: Finds if the given code
is a QR Code
 Timing Pattern: Tracks the timing of the
incoming code
 Separators: Separates data from the
patterns
 Alignment Pattern: Tells about initiation
& Separation of data.
 Dark Module: A black module placed
beside the bottom left finder pattern.

Cont…
 Many Alignment patterns
 Long timing pattern

Versions of QR code
Version 1 (21×21)
Version 2 (25×25)
Version 3 (29×29)

Versions of QR code cont…
Version 4 (33×33)
Version 10 (57×57)
Version 25 (117×117 enlarged to
640x640)

Versions of QR code cont…
Version 40 (177×177)

What can a QR code do?
 View a web page
 Dial telephone number
 Send a template text message
 Send an e-mail
 View a message or special offer
 Download contact details (vcard)
 View a google maps location
 View a social media profile
 Take a snap and view web page

 View a web page
 Send an e-mail
 Capture and call

 View a web page
 Send an e-mail
 Take a snap and send a template
message

 View a web page
 Send an e-mail
 snap and send a default mail

 View a web page
 Send an e-mail
 Take a snap
and see the
message

 View a web page
 Send an e-mail
 Capture and save contact

 View a web page
 Send an e-mail
 Capture and recognize the exact place

 View a web page
 Send an e-mail
 Snap and see the media profile

Applications
 Conference/event displays
 Print advertisements
 Business cards
 Brochures , posters, & flyers
 Post cards & mail

Pre-processing
 Capturing image
 Contrast enhancement
 QR code extraction
 Deblurring the code
 Code alignment : corner and pattern detection and
transformation
 Scanning the code

Pre-processing
 Capturing image
 Deblurring the QR code
 Pattern detection
 The images are captured by a 3 Megapixel-sensor
 continuously refreshed on the screen to provide the user
with some feedback during the image capture process

Pre-processing
 Capturing image
To balance environment light effect
Histogram equalization
 It is a method in image processing of contrast adjustment
using the image's histogram.
 Employs a monotonic, non-linear mapping which re-assigns
the intensity values of pixels in the input image such that the
output image contains a uniform distribution of intensities

Pre-processing
 Capturing image
Cropping out external surroundings
 Image binarisation
 Edge detection
 Morphological operations
 Quality index

QR code extraction
1. Image binarisation
Binarization image by multi-level
thresholding method.
Captured image
The simplest way to use image binarisation is to
choose a threshold value, and classify all pixels
with values above this threshold as white, and all
other pixels as black.
There are only two levels of intensity in the binary
image. We consider the pixels that have low
intensity is an image pixel where high intensity as
background pixel.

QR code extraction cont…
2. Edge detection
I. Smoothing
II. Finding gradients
III. Non maximum suppression
IV. Double thresholding
V. Edge tracking by hysteresis

2. Edge detection
I. Smoothing
Blurring the image to avoid noise.

2. Edge detection
I. Smoothing
The edges should be marked where the gradients
of the image has large magnitudes.
The algorithm basically finds edges where the
greyscale intensity of the image changes the most.
These areas are found by determining gradients of
the image.

2. Edge detection
I. Smoothing
Only local maxima should be marked as edges.
The purpose of this step is to convert the “blurred”
edges in the image of the gradient magnitudes to
“sharp” edges.
Basically this is done by preserving all local
maxima in the gradient image and deleting
everything else.

2. Edge detection
I. Smoothing
The edge-pixels remaining after the non-
maximum suppression step are marked with their
strength pixel-by-pixel.
The edge detection algorithm uses double
thresholding.

2. Edge detection
I. Smoothing
Final edges are determined by suppressing all
edges that are not connected to a true (strong)
edge.
Thus strong edges will (almost) only be due to true
edges in the original image.
The weak edges can either be due to true edges
or noise/color variations.

3. Morphological operations
I. Morphological dilation
II. Morphological closing
Morphological operations are affecting the form, structure or shape of
an object.
used for getting a representation or description of the shape of
objects/regions.

II. Morphological closing
Dilation adds pixels to the boundaries of objects
in an image.
The number of pixels added to the objects in an
image depends on the size and shape of the
structuring element used to process the image.

II. Morphological closing Closing removes small holes(islands) and other
thin filaments of background pixels.
i.e.; it can fill holes in the regions while keeping
the initial region sizes.

4. Quality index
I. Error correction
II. Discard low quality images
Used in further modules as a parameter to check the quality.
QI has been designed so that it can be readily computed for every
input image once the QR code area has been extracted.

4. Quality index
I. Error correction
II. Discard low quality images
For error correction the Reed-Solomon code is
used.
It is a non binary cyclic-error correcting code.
It is has some steps
Takes block of data.
Adds “redundant” bits to form a code
word.
The code word can be decoded to
correct for lost or corrupted bits.

4. Quality index
I. Error correction
II. Discard low quality images The very low quality images are more complex
in the deblurring module, so the image will be
avoided

Contrast adjustment and code extraction: (a) captured image; (b) contrast- enhanced image;
(c) Binarized image; (d) external elements removal;(e) edge detection; (f) high edge density area
identification after dilation; (g) high edge density area identification after dilation and closing;
and (h) codeextraction

Pre-processing
 Capturing image
Although ideally QR codes are two-tone images.
Used some deblurring algorithms.

Pre-processing
 Capturing image
determine the proper compensating transform, the four
corners of the code have to be determined and the patterns
of the QR code detected.
The corner coordinates associated with the three finder
patterns are considered precise enough.
the fourth corner must be refined since it may be actually
defined by position of the finder patterns, which is
undetectable by the scanning, causing pixel loses.

Pre-processing
 Capturing image
Decoding the printed data to the digital data

Decoding QR code
The QR code to decoding

Decoding QR code cont…
Grouping to octets…

Conclusion
 In this paper, a practical image Pre-processing method was
proposed for QR code recognition.
 The influence by background noise and inhomogeneous light and
geometric distortion was minimized.
 In additional, it used alignment patterns to adaptively sample the
barcode in terms of regions, which greatly improved the recognition
rate.
 The number of calculation of this algorithm is reduced, its
recognition speed is higher.

References
 [1]IEEE Explore ,Munoz-mejias,Gonzalez-Diaz,Diaz-de-Maria:” A low
pre-processing system for restoring low quality QR code images”
 [2]Wikipedia/QR-code
 [3]Yunhua Gu,Weixiang Zhang,"QR code recognition Based on
Image Processing",International Conference on Information Science
and Technology ,March 26-28, 2011 Nanjing, Jiangsu, China
 [4]M.Sun,LS.Fu,XTYang,S.H.zhang,"Image Analysis Method for QR
Code's Automatic Recognition",Journal of University of Electronic
Science and Technology of China, Vo1.38,No.6,pp.122-
125,Nov.2009.

A low pre processing system for low quality QR code

A low pre processing system for low quality QR code

More Related Content

What's hot (20)

Similar to A low pre processing system for low quality QR code (20)

Recently uploaded (20)

A low pre processing system for low quality QR code