A Robust Software Barcode Reader Using the Hough Transform . In this paper we present a method based on the Hough transform which. Published in: · Proceeding. ICIIS ’99 Proceedings of the International Conference on Information Intelligence and Systems. Page March 31 – April A Robust Software Barcode Reader Using the Hough Transform (Englisch). Muniz, R. / Junco, L. / Otero, A. / Institute of Electrical and Electronics Engineers.

Author: Goltikree Tugis
Country: Malta
Language: English (Spanish)
Genre: Education
Published (Last): 2 April 2018
Pages: 135
PDF File Size: 15.54 Mb
ePub File Size: 11.65 Mb
ISBN: 500-8-50034-535-4
Downloads: 76040
Price: Free* [*Free Regsitration Required]
Uploader: Dizilkree

Chai and Hock use a single scanline binarized using the average gray level as ttansform threshold [ 4 ] whereas Wachenfeld et al. Robust recognition of 1-D barcodes using camera phones. Therefore, in order to localize the endpoints o L barcoe o R of the barcode, we first determine the intersections i L and i R of the scanline l n with the rectangle and then, the rectangle being larger than the actual barcode, we proceed inwards from each end see Fig.

Reading 1-D Barcodes with Mobile Phones Using Deformable Templates

The implementation is done in such a way that most of the time it will operate on light search and partial search respectively. For barcode localization we trxnsform used a simple mathematical morphological operation method. From those two images one Higher Response trahsform be selected, which has higher response. Due to our parametrization of these templates, we can efficiently perform maximum likelihood estimation independently on each digit and enforce spatial coherence in a subsequent step.

This shows that our algorithm is suitable for mobile applications and also proves that, in the event of failure at reading the barcode, the process can be quickly repeated. The first step for a barcode reader is the localization of the barcode in the image.

Unlike previous approaches, our algorithm does not binarize the image, thus sidestepping a critical and often error-prone, early-commitment procedure.

The maximum rotation angle for correct localization is bound by simple geometric considerations. Early-commitment approaches, whether based on binarization or edge detection, are efficient from a computational standpoint, but rely heavily on the good quality of the input image. Hugh risk of such errors can be reduced by exploiting global constraints on the overall sequence of symbols.


Although the algorithm relies on the assumption that the user is holding the cellphone so that the bars are vertical in the image, it produces good results even when the phone is rotated at an angle. Supplementary Material Dataset used for the experiments Click here to view.

Slftware the field of view of the Nokia N95, our implementation can successfully localize and read barcodes that are 2. The equations for these lines are transvorm computed; for the third bar plot dfor instance, we can write: This operation can be somewhat bothersome, as it requires a certain amount of interaction with the user, who needs to frame the barcode correctly using the viewfinder.

Our algorithm correctly decodes barcodes even when the image quality is extremely poor, as can be appreciated by zooming in on the images. The barcode reader Given an image containing a barcode, two distinct operations are needed for accessing the information contained in the barcode: Commercial laser-based, hand-held barcode scanners achieve robust reading with a reasonable transforn tag.

A system that binarizes the intensity would be hard-pressed to detect the correct pattern. Suppose that the j -th digit takes value k j. Each digit represents one symbol as a sequence of two spaces and two bars.

Unfortunately, images taken by cellphone cameras are often of low quality. Transform at ion After detecting the orientation, the whole image is rotated in reverse direction by the angle determined above. From that image, the blob which corresponds to barcode area is selected using following formula Dilation and Erosion [1].

III-A proved rather robust in our experiments. Therefore, we only store area and centroid location of each polygon. These situations are responsible for most of the failures in the experiments described below. Testing was carried out times, in different backgrounds as well as different barcode orientations.


From the moment that the frame is received from the camera and the analysis begins, our algorithm takes an average of — ms for both localization and decoding. More precisely, we define the likelihood of the intensity within a generic digit segment for symbol k conditioned on o and w as. In order to assess our system in other realistic situations, we gathered a number of images taken from two different cellphones, and created three new data sets.

Reading 1-D Barcodes with Mobile Phones Using Deformable Templates

The horizontal line l n that passes through the center of this rectangle is chosen as the scanline for the analysis. The size of the filter was chosen based on the range of the size of the input images and the minimum size of the barcode readable by our method. The resulting rectangular segment black squarealong with the selected scanline l nthe intersection points i L and i Rand the endpoints o L and o R. The localization method is based on detecting the areas with the maximum density difference in two normal directions.


Imposing spatial coherence Our model makes the xoftware initial assumption that the digit segments are equally spaced see Eq. Author information Copyright and License information Disclaimer. A third online software, from QualitySoft, was considered by Tekin and Coughlan [ 10 ], but we neglected comparison with it since it gives very poor results.

The last digit is an error correcting check digit which can be used to detect decoding errors. If there are many solutions, then the rule; keeping the combinations with highest probabilities and discarding the other, can be applied.