Exposing parallelism of discrete radon transform

Gómez-Cárdenes, Óscar; Oliva-García, Ricardo; Rodríguez-Abreu, Gabriel A.; Marichal-Hernández, José G.

doi:10.1145/3369555.3369559

Cited by 1 publication

(3 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A method to better expose the parallelism of the DRT [39], which achieves, in one pass, the eminently horizontal lines of the form y = ±(x • slope + displacement), and in another pass, those of the form x = ±(y • slope + displacement), thus changing from four quadrants to two groups of 90 degrees (around the horizontal axis and the vertical axis). Additionally, this reformulation eliminates the need for prior zero padding, which the conventional DRT requires.…”

Section: Discrete Radon Transformmentioning

confidence: 99%

“…See Figure 3. The method described by Oliva-García et al [33], consisting of the above method, works because the partial m stage of the modified DRT as described by Gómez-Cárdenes et al [39] computes the sums at 2 m+1 − 1 possible angles of line segments of length 2 m , starting at each position (2…”

Section: From a Line Detector To A Local Bar Detectormentioning

confidence: 99%

“…The unmodified Radon transform [39] ends in stage n = log 2 N, with N being the side size of a square image, exhibiting computational complexity O(N 2 log 2 N) or, more precisely, examining the range of the for loops; the number of operations (two data reads, one addition and one write per iteration) for the horizontal partial DRT is…”

Section: Computational Complexity Without Overlappingmentioning

confidence: 99%

See 2 more Smart Citations

An Encoder–Decoder Architecture within a Classical Signal-Processing Framework for Real-Time Barcode Segmentation

Gómez-Cárdenes

Marichal-Hernández

Son

et al. 2023

Sensors

Self Cite

View full text Add to dashboard Cite

In this work, two methods are proposed for solving the problem of one-dimensional barcode segmentation in images, with an emphasis on augmented reality (AR) applications. These methods take the partial discrete Radon transform as a building block. The first proposed method uses overlapping tiles for obtaining good angle precision while maintaining good spatial precision. The second one uses an encoder–decoder structure inspired by state-of-the-art convolutional neural networks for segmentation while maintaining a classical processing framework, thus not requiring training. It is shown that the second method’s processing time is lower than the video acquisition time with a 1024 × 1024 input on a CPU, which had not been previously achieved. The accuracy it obtained on datasets widely used by the scientific community was almost on par with that obtained using the most-recent state-of-the-art methods using deep learning. Beyond the challenges of those datasets, the method proposed is particularly well suited to image sequences taken with short exposure and exhibiting motion blur and lens blur, which are expected in a real-world AR scenario. Two implementations of the proposed methods are made available to the scientific community: one for easy prototyping and one optimised for parallel implementation, which can be run on desktop and mobile phone CPUs.

show abstract

Section: Discrete Radon Transformmentioning

confidence: 99%

Section: From a Line Detector To A Local Bar Detectormentioning

confidence: 99%

Section: Computational Complexity Without Overlappingmentioning

confidence: 99%

See 1 more Smart Citation