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: Kagagal Fauran
Country: Guyana
Language: English (Spanish)
Genre: Politics
Published (Last): 28 June 2005
Pages: 260
PDF File Size: 3.20 Mb
ePub File Size: 4.33 Mb
ISBN: 329-6-71763-440-2
Downloads: 61020
Price: Free* [*Free Regsitration Required]
Uploader: Shakaramar

A possible strategy could be to search for the deformation parameters that maximize the likelihood, i. This database, which was created by Tekin hougu Coughlan, is accessible at www.

Reading 1-D Barcodes with Mobile Phones Using Deformable Templates

A parameterized model is a shifted and scaled deformed version of the original model:. Many cellphone cameras on the market are equipped with low-grade lenses, generally lacking focusing capability, which often produce blurred images.

In Canny method it uses a filter which is narrow o as possible to provide suppression of high frequency noise and to provide good localization of the edges. Note how blur and low resolution affect the intensity profile. 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.

The images are all of high resolution, and each image was manually cropped around the barcode.

In order to allow other researchers to compare their results against ours, we provide a publicly accessible barcode image data set bafcode 2 ]. For example, when the scanline width is equal to pixels, then 25 cells are generated.

These crop-outs are the output of the localization algorithm described in Sec. We then define a likelihood function to measure how well a deformed shifted and scaled template explains the transcorm intensity. This led to the development of SmartCamBCR, the first barcode reader to use a web cam a small camera used for sending images over the Internet as gransform capturing device and to use image processing and artificial intelligence techniques to obtain the barcode value.

The second one, from DTK 3was shown to produce impressive results. In our implementation we assume a localization tolerance in either end point equal to twice the width of the narrowest bar. 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.


One important contribution of transfom work is to derive an algorithm to compute this integral exactly and in affordable computing time. First barcode region given by the localizer is transformed into Hough Det ect Parallel Lines plane. Due to the presence of the noise, tgansform are some infrequent narrow peaks and valleys appeared in the waveform. Note that block filtering can be implemented efficiently so that only few operations per pixel are required.

Table 2 shows a part of the probabilities with the different normalized transfodm ranges. Automatic real-time barcode localization bough complex scenes. This device has autofocus capability, although not all images collected were properly in-focus whether because the focus was on the wrong object or because the focusing algorithm failedand some were affected by motion blur. Few cellphones transformm a flash and, therefore, motion blur and noise can be expected with low ambient light.

Comparative results showing the number of barcodes correctly detected in the different data sets considered on the left the data sets from Tekin and Coughlan [ 10 ], on the right our data sets.

Flow Chart of Transformer Figure 4: Barcode Localization Localization process is based on mathematical morphology. In this case, the integral is proportional to the area of the polygonal cell, which can be easily computed and stored offline. Associated Data Supplementary Materials Dataset used for the experiments. For each cell Usihg jk tthe negative log-likelihood D t needs to be computed, which requires two additions and two multiplications per sample.

For instance, they can enable users to access reviews or price comparisons for a sofftware from a picture of its barcode while still in the store. Dilation, in general, causes objects to dilate Edge Image or grow in size; while erosion causes objects to shrink. For example, consider the segment between pixels 32 and 38 representing a sequence of two black and two white bars.

Rather than computing the connected components of the thresholded map, we simply select the pixel n 0 that maximizes I s nunder the assumption that the correct blob, i.

In addition, we propose an optimization procedure to enforce spatial coherence of the individual digits found by deformable template matching. Unfortunately, we could find only one barcode image database for comparative assessment 1.

Indeed, all existing image-based barcode readers have limited performance when it comes to images taken in difficult light conditions, or when the camera is not close enough to the barcode. An example of barcode localization with our algorithm. This, in turn, is due to the very nature of the scanline which is itself piecewise constant: We use deformable templates to represent each digit of the barcode, averaging over the set of all expected deformations.


The consistency checking bsrcode, a constraint network, employs artificial intelligence searching methods. If there are many solutions, then the rule; keeping the combinations with highest probabilities and discarding the other, can be houvh.

Additionally, their design often requires the codes to be fairly close to the scanner.

Reading 1-D Barcodes with Mobile Phones Using Deformable Templates

The illumination problem could be overcome by properly illuminating controlling of the light level the barcode gough. If necessary, other Transform at ion transformations such as scaling are applied to the image. Open in a separate window. In the right half of the code, the same sequence of widths is used to encode a symbol, however the role of spaces and bars is inverted.

Footnotes 1 Wachenfeld et al. Implementation and tests We implemented and tested our algorihm both in Matlab and in Symbian OS, a platform used by several cellphone brands. In the Hough plane, each point in the original plane is represented by a line. From those two images one Higher Response can be selected, which has higher response.

Indeed, storing the partition of the dodw plane and intersecting it at runtime with a rectangle whose sides are given by the tolerances described in Sec. Finally, we binarize I s n with a single threshold, selected using the method proposed by Otsu [ 8 ]. Indeed, a number of cellphone apps have appeared that provide access via barcode reading to the full characteristics of and user reviews for a product found at a store.

The consistency checking with the constraint network rectifies the effect of the noise by checking the relationships of the widths of the nearby bars and taking into consideration of the decoding rules of EAN Also this process uses Canny 0o edge detection mechanism to get tgansform edge usimg from the given 45 o 90 o original image.