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: | Mautilar Kazigal |

Country: | Kuwait |

Language: | English (Spanish) |

Genre: | Travel |

Published (Last): | 1 January 2010 |

Pages: | 113 |

PDF File Size: | 12.41 Mb |

ePub File Size: | 14.96 Mb |

ISBN: | 474-7-42157-470-4 |

Downloads: | 24548 |

Price: | Free* [*Free Regsitration Required] |

Uploader: | Vumi |

Barcodes are used in wide variety of areas such as for industrial identification, for publications, for pharmaceutical products and for postal codes. The only requirement is that the beginning and the end of the barcode pattern in the scanline are detected with a certain accuracy.

A sample of images correctly decoded by our algorithm is shown in Fig. In our experiments, this assumption was always found to be correct. In addition, we propose an optimization procedure to enforce spatial coherence of the individual digits found by deformable template matching.

### Reading 1-D Barcodes with Mobile Phones Using Deformable Templates

Currently he is a Ph. Most existing applications require the user to move the camera towards the barcode so as to maximize its apparent width in the image. In order to assess our system in other realistic situations, we gathered a barcdoe of images taken from two different cellphones, and created three new data sets. This uncertainty derives from the finite tolerance on the estimation of o L and o R.

Few cellphones have a flash and, therefore, ttransform blur and noise can be expected with low ambient light. We implemented other algorithms from the literature [ 11 ], [ 14 ]; however, these methods produced results comparable or inferior to our simple method, at a substantially higher computational cost.

Other approaches assume that the center of the image falls within the barcode area [ 7 ], [ 12 ], thus greatly simplifying the problem: The peak detection method is far more superior to edge detection methods as it has a higher degree of immunity to noise. The red lines represent the original digit segments, obtained from Eq. A system that binarizes the intensity would be hard-pressed to detect the correct pattern.

Robust recognition of 1-D barcodes using camera phones. Next the region of interest ROI is extracted after applying necessary transformation using Hough line detection 2 method. Transformed Image Figure We sofrware presented a new algorithm for barcode decoding localization and reading that can deal with images that are blurred, noisy, and with low resolution.

Each of the consecutive, non-overlapping segments which encode the symbols are called digits segments or simply digits. III describes in detail the proposed algorithm. The normalized distances can be calculated by dividing the distances between consecutive peaks and valleys by the minimum bar width.

The second step requires that the scanline intersect all the bars, from the left to the right guard bars; since we extract an horizontal line, this becomes harder as the angle increases too much.

Although this algorithm relies on the assumption that the bars of the barcode are approximately vertical in the image, our studies show that the map I e n can be segmented even when this assumption is not satisfied. Information is encoded in the width of the stripes; sequences of a fixed number of stripes encode a character.

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. We propose a simple and fast algorithm for localization that assumes that the bars are approximately vertical. Unfortunately, images taken by cellphone cameras are often of low quality. A parameterized model is a shifted and scaled deformed version of the original model: The matching of the stripe-to-stripe combinations in the consecutive domains and the value of the probability can be considered as the rules for checking the consistency.

The illumination problem could be overcome by properly illuminating controlling of the light level the barcode image.

Indeed, storing the partition of the dosortware plane and intersecting it at runtime with a rectangle whose sides are given by the tolerances described in Sec. Most of these barcode readers use a laser beam to scan the barcode and give the resulting value.

In some barcode types a checksum digit is included in order to validate teh reading. Virtually any existing algorithm for barcode reading performs some sort of binarization of the input brightness data. From those two images one Higher Response can be selected, which has higher response. Matlab implementation Localization The algorithm described in Sec. Also this process uses Canny 0o edge detection mechanism to get the edge image from the given 45 o 90 o original image.

Suppose robhst the j -th digit takes value k j.

If we break up the sum in Eq. 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. Considering the shift of the normalized distances due to the noise, robuet predefined probability can be assigned to the stripe-to-stripe combination.

The lateral guard bars are sided by a space of width equal to at least 9 times the base width quiet zonealthough this requirement is sometimes violated in real-world instances. Therefore, in order to localize the endpoints o L and 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. He then worked on a novel bioimaging technique at the Smith-Kettlewell Eye Research Institute from to More precisely, we compute the sum in Eq.

More precisely, we define the likelihood of the intensity within a generic digit segment for symbol k conditioned on o and w as. With this modification, the number of tranwform is reduced to 17 in the pixel scanline width case, and to in the pixel case. Author manuscript; available in PMC Jun If, due to noise or blur, a few pixels have intermediate gray values as in Fig.

## Reading 1-D Barcodes with Mobile Phones Using Deformable Templates

The execution time is only a function of the size of the image. The equations for these lines are easily computed; for the third bar plot dfor instance, we can write: Computing this integral may seem like a daunting task, especially if it needs to be performed on an embedded system such as a cellphone.

Their method compares favorably with previous approaches although it was only implemented on laser-based barcode scanners. The log-likelihood term D can be expressed as.