IMPROVING PERFORMANCE OF THE DECODER FOR TWO-

DIMENSIONAL BARCODE SYMBOLOGY PDF417

Hee Il Hahn and Jung Goo Jung

Department of Information and Communications Eng., Hankuk University of Foreign Studies, Korea

Keywords: Two-dimensional barcode, Segmentation, Error correction code, Warping

Abstract: In this paper we introduce a method to extract the bar-space patterns directly from the gray-level two-

dimensional

barcode images, which employs the location and the distance between extreme points of

profiles scanned from the barcode image. This algorithm proves to be very robust from the high

convolutional distortion environments such as defocussing and warping, even under badly illuminating

condition. The proposed algorithm shows excellent performance and is implemented in real-time.

1 INTRODUCTION

Linear barcodes have been used globally in the

various fields such as supermarkets and other stores

for several decades. Usually, they do not have

detailed information but just carry a key to database,

because they can hold only several bytes of

information. The need to increase the amount of data

in a symbol brought the introduction of a new form

of barcodes with much higher density called the

two-dimensional (2-D) barcodes. They have been

introduced since 1990’s. While conventional or one-

dimensional barcodes usually function as keys to

databases, the new or two-dimensional barcodes

meet a need to encode significantly more data than

the conventional codes and would act as a portable

data file because the information could be retrieved

without access to a database. They have additional

features such as error correction and the ability to

encode in various languages like English, Korean

and Chinese, etc., besides their increased capacity.

Thus, the traditional concept of barcode as a key to a

database is changing towards a “portable data file”

in which all the relevant information accompanies

the item without access to a database.

There are two types of 2-D symbologies - stacked

and

matrix-type symbologies. The stacked barcodes,

to which Code49, PDF417, etc. belong, have the

structure of rectangular block comprising numbers

of rows, each of which is like 1-D symbology. The

matrix-type barcodes are essentially a form of binary

encoding in which a black or white cell can

represent either binary 1 or 0. These cells are

arranged in an array structured on a grid of

rectangular block. DataMatrix, Maxicode, and QR

code, etc. are representative of Matrix-type

symbology (Pavlidis, 1992).

In this paper, we focuss only on PDF417 (AIM

A, 1994), known as the most widely used 2-D

stacked symbology. PDF417 is a multi-row,

variable-length symbology offering high data

capacity and error-correction capability. A PDF417

symbol is capable of encoding more than 1,100

bytes, 1,800 ASCII characters, or 2,700 digits,

depending on the selected data compaction mode.

Every PDF417 symbol is composed of a stack of

rows, from a minimum of 3 to a maximum of 90

rows. Each PDF417 row contains start and stop

patterns, left and right row indicators. Fig. 1 shows

the high-density scanned barcode image of PDF417

code.

We explain about our barcode decoder and

opose a novel method of extracting codewords

directly from gray-level barcode image by analyzing

their profiles instead of binarizing the image.

2 LOCALIZING THE DATA

REGION

The data region is localized to extract the bar-space

patterns and to detect the codewords, in the

following way. Firstly, the start pattern or stop

pattern, or both are searched by scanning

horizontally and vertically. The start pattern is

81111113 and the stop pattern is 711311121, both

beginning with bar. Fig. 2 shows the segments

130

Hahn H. and Jung J. (2004).

IMPROVING PERFORMANCE OF THE DECODER FOR TWO-DIMENSIONAL BARCODE SYMBOLOGY PDF417.

In Proceedings of the First International Conference on Informatics in Control, Automation and Robotics, pages 130-134

DOI: 10.5220/0001139101300134

 SciTePress

corresponding to the start patterns, which are

extracted from the clear barcode image and the

highly blurred, high-density one respectively,

together with their profiles. Although the start

patterns shown in Fig. 2-(a) can be easily identified,

it might not be easy to detect them in case of Fig. 2-

(b) because the width of one module is less than 1.5

pixels and the dynamic range of the profile at the

region corresponding to narrow bar or space is

severely reduced due to convolution with the point

spread function of the camera. After the start pattern

and/or stop pattern are detected, line equations

corresponding to them are estimated through line

fitting to localize left and/or right row indicators.

Secondly, the header information such as the

number of rows and columns, error correction level,

row number, error correction level and the width of

module, etc. are obtained by extracting and decoding

the left and/or right row indicators (Hahn, 2002).

Figure 1: The scanned PDF-417 barcode image

Finally, four vertices of the data region are detected

by scanning in the directions perpendicular to the

start pattern line and the stop pattern line. To ensure

the vertices obtained are correct, row numbers are

checked by decoding the left or right row indicators

around the vertices. If the row numbers

corresponding to the detected four vertices are not

the first row or the last row, the true four vertices are

predicted by considering the proportion of the row

number to the number of row, as depicted in Fig. 3.

Since barcode image is usually scanned using a

digital camera, it should be warped due to the

nonlinearity of lens and the viewing angle of the

camera. In this paper, Affine transform is adopted to

warp the data region of the scanned image

(Gonzalez, 1993). Fig. 4 shows the warped result of

the data region inside the barcode in Fig. 1.

3 DECODING CODEWORDS

FROM BAR-SPACE PATTERNS

After the data region is localized and warped as

mentioned above, bar-space patterns are extracted

from the data region and are decoded to obtain the

corresponding codewords. The problem is how we

get bar-space pattern from the barcode data region.

Usually, image thresholding is employed for this

goal. Many researchers developed the image

thresholding algorithms, most of which reported to

date are based on edge information or histogram

analysis using information-

(a)

(b)

Figure 2: Profiles and segments corresponding to the start

patterns of the scanned barcode, which are excerpted from

(a) clear image and (b) highly blurred, high-density image.

theoretic approaches (Parker, 1991). The selection of

optimal thresholds has remained a challenge over

decades. However, they can not be applied to decode

the barcode images, because the widths of bars or

spaces can be 2 pixels or even less in case of high-

density barcode images and even slight variation of

threshold can cause severe errors (Eugene, 1991).

Figure 3: Locating true vertices by checking row numbers.

Fig. 5 shows the segmented bar-space patterns and

their corresponding profiles obtained from the

focussed clean image and the outfocussed image.

Although the widths of bars and spaces of Fig. 5-(a)

IMPROVING PERFORMANCE OF THE DECODER FOR TWO-DIMENSIONAL BARCODE SYMBOLOGY PDF417

131

can be easily measured, it might not be a simple

matter to measure them in case of Fig. 5-(b) and the

obvious thing is single global threshold can not even

discriminate them. Fig. 6 shows the intensity profile

of the codeword ( 824 : 1, 5, 1, 2, 1, 1, 1, 5 )

comprised of four bars and four spaces, together

with their widths and classification results.

As shown in Fig. 6-(a), it is impossible to select

the single optimal threshold for detecting the widths

of four bars and four spaces because the pixel values

change dynamically according to the widths of bars

and spaces.

Figure 4: Warping the data region of Fig. 2 using Affine

transform.

The widths and peaks of narrow bars or spaces

corresponding to 1 or 2 module values get smaller

compared to the wide ones even under the same

illumination, due to convolution with the point

spread function. The proposed algorithm employs

the high curvature points and local extreme points to

extract four bars and four spaces from the warped

barcode image. The points of high curvature on the

waveform represent the midpoints of the bar or

space, whereas the local extreme points are

estimates of their centers.

Figure 5: Segmented bar-space patterns and their

corresponding profiles obtained from (a) the focussed

image, (b) the outfocussed image.

At first, our algorithm localizes four local minimum

points and four local maximum points by scanning

the profile of each bar-space pattern, as shown in Fig.

6-(b). The local minimum point whose value is large

compared to the other minimum points or the

maximum point smaller than the adjacent ones are

compensated to increase their dynamic range as

depicted in Fig. 6-(c). Then, the break regions are

detected between the compensated extreme points,

where the break region means the region whose

profile is horizontal. The regions marked as

rectangular box in Fig. 6-(d) represent the break

regions. Finally, the break regions are partitioned

according to the following rules. Usually, the edges

between bars and spaces lie in the break regions. In

Fig. 7, we define

∆

= x2 – x1 and = x3 – x2.

b∆

∆

is greater than

∆

, the more part of the break

region belongs to the space region, vice versa. The

ratios are obtained by experimenting with several

hundreds of bar-space patterns, extracted from the

various barcode images scanned under varying

conditions. Thus, the widths of bars and spaces are

represented as real values rather than integer ones to

measure the bar-space patterns as correctly as

possible.

4 EXPERIMENTAL RESULTS

The “edge to similar edge” estimation method is

employed to check whether the detected bar-space

pattern is correct. The detected bar-space patterns

can be converted to the encoded codewords by using

the lookup table specified in (AIM USA, 1994). The

codewords are checked whether there are any errors

through Reed-Solomon error correction coding

algorithm. Given the error-corrected codewords,

they are decoded in the manner as specified in (AIM

USA, 1994), to get the message.

In order to verify our algorithm , we benchmark-

tested our decoder with the test barcode images. Our

database is composed of 153 barcode images, which

were taken under various conditions. In other words,

they are rotated, outfocussed, warped or even

severely damaged by cutting off some region, as

shown in Fig. 8. Almost of them are taken under

badly illuminated conditions. At first, 2,000 pieces

of profiles corresponding to the bar-space patterns

are extracted from our database. Each profile is

tested to extract the bar-space pattern and decode the

corresponding codeword. Among them, 1,466

profiles are detected correctly. As an example, the

image of Fig. 8-(a) is obtained by hardcopying the

barcode image several times, whose upper part is

severely degraded. When it is applied to our decoder,

42 erasures and 32 errors are detected among total

329 codewords, which can be decoded correctly

through Reed-Solomon error correction algorithm.

Fig. 8-(b) is an image reassembled but aligned

incorrectly after tearing it into two parts. Fig. 8-(c) is

obtained from Fig. 8-(a) by cutting off the right part

of it and Fig. 8-(d) are taken outfocussed under too

bright illuminating condition. Our algorithm

decoded 138 images correctly among total 153

images. This result is expected to be good for

ICINCO 2004 - SIGNAL PROCESSING, SYSTEMS MODELING AND CONTROL

132

manufacturing purpose although there might be no

public images for benchmarking test

5 CONCLUSION

We have proposed algorithms to decode two-

dimensional barcode symbology PDF417 and

implemented a barcode reader in real-time using

ARM core. Our decoder employs a method to

extract the bar-space patterns directly from the

profiles of the gray-level barcode image. Our

algorithm shows performance improved further than

the method of extracting the bar-space patterns after

binarizing the barcode image, when we experiment

with the test images of variable resolution and error

correction levels. In order to improve the

performance further, it is needed to extract bar-space

patterns more accurately from the barcode image,

which might be defocused or taken under badly

illuminating conditions

(a)

(b )

(c )

(d )

Figure 6: The intensity profiles of the codeword comprised

of four bars and four spaces

Figure 7: Partition of break region according to the ratio

∆

b∆

(a) (b)

Figure 8: Sample images for measuring the performance

of our decoder.

REFERENCES

AIM USA, 1994. Uniform Symbology Specification

PDF417.

IMPROVING PERFORMANCE OF THE DECODER FOR TWO-DIMENSIONAL BARCODE SYMBOLOGY PDF417

133

Eugene Joseph and Theo Pavlidis, 1991. Waveform

Recognition with Application to Barcodes, Symbol

Technologies Inc. 116 Wilbur Place, Bohemia, NY.

R. C. Gonzalez, R.E.Woods, 1993. The book, Digital

Image Processing, Addison Wesley.

Hee Il Hahn, Joung Goo Joung, 2002. Implementation of

Algorithm to Decode Two-Dimensional Barcode PDF-

417, 6

International Conference on Signal

Processing ICSP’02.

J. R. Parker, 1991. Gray level thresholding in badly

illuminated images, IEEE Trans. on PAMI, vol.13,

No.8.

T. Pavlidis, J. Swartz, Y. P. Wang, 1992. Information

encoding with two-dimensional barcodes, IEEE

Computer.

ICINCO 2004 - SIGNAL PROCESSING, SYSTEMS MODELING AND CONTROL

134