Modified Discrete Radon Transforms and Their Application to Rotation-Invariant Image Analysis

An autonomous parameter extraction algorithm for frequency modulated continuous wave (FMCW) radar signals using Wigner-Ville Distribution (WVD)-Hough transform was investigated in [1] and extraction of polyphase radar modulation parameters using a Wigner-Ville distribution-Radon transform was investigated in [2]. The algorithm in [1] produced very dependable results with as low as -6 dB SNR levels, however some degradation has been observed below -6 dB SNR. The proposed approach in this study uses the WVD as a timefrequency (T-F) detection technique and combined HoughRadon transform (HRT) to identify the parameters of the modulation.We showed that our algorithm can extract FMCW radar modulation parameters at low SNR levels, such as -9 dB, efficiently.

show abstract

“…Using the same definition of a line in Eq. (5), the Radon transform of a 2-D image ( , ) f x y is [7],…”

Section: Radon Transformmentioning

confidence: 99%

A parameter extraction technique for FMCW radar signals using Wigner-Hough-Radon transform

Gulum

Erdogan

Yıldırım

et al. 2012

2012 IEEE Radar Conference

View full text Add to dashboard Cite

show abstract

“…Finally, Tébessa I [13] method uses a direct segmentation strategy based on the oriented sliding window [29]. Recognition is carried out by projecting the vertical histogram (single digit image) or by using the Radon transform [30] to obtain the orientation angle of the sliding window, combined with SVM [25]. The second method proposed by the same group, Tébessa II [13], uses the same description as above, but with multiple orientation angles of the sliding window [29].…”

Section: A Methods From Icfhr 2014 Competitionmentioning

confidence: 99%

HDSR-Flor: A Robust End-to-End System to Solve the Handwritten Digit String Recognition Problem in Real Complex Scenarios

et al. 2020

View full text Add to dashboard Cite

Automatic handwriting recognition systems are of interest for academic research fields and for commercial applications. Recent advances in deep learning techniques have shown dramatic improvement in relation to classic computer vision problems, especially in Handwritten Text Recognition (HTR). However, several approaches try to solve the problem of deep learning applied to Handwritten Digit String Recognition (HDSR), where it has to deal with the low number of trainable data, while learning to ignore any writing symbol around the digits (noise). In this context, we present a new optical model architecture (Gated-CNN-BGRU), based on HTR workflow, applied to HDSR. The International Conference on Frontiers of Handwriting Recognition (ICFHR) 2014 competition on HDSR were used as baselines to evaluate the effectiveness of our proposal, whose metrics, datasets and recognition methods were adopted for fair comparison. Furthermore, we also use a private dataset (Brazilian Bank Check-Courtesy Amount Recognition), and 11 different approaches from the state-of-the-art in HDSR, as well as 2 optical models from the state-of-the-art in HTR. Finally, the proposed optical model demonstrated robustness even with low data volume (126 trainable data, for example), surpassing the results of existing methods with an average precision of 96.50%, which is equivalent to an average percentage of improvement of 3.74 points compared to the state-of-the-art in HDSR. In addition, the result stands out in the competition's CVL HDS set, where the proposed optical model achieved a precision of 93.54%, while the best result so far had been from Beijing group (from the competition itself), with 85.29%.

show abstract

“…where δ represents the unit-impulse function. The function f(x; y) represents the intensity of the image at the specified coordinates point, (x; y) [1,6].…”

Section: Methodsmentioning

confidence: 99%

“…Estimating the spatial coordinates and trajectory based on the analysis of the images taken from the scene is a common problem in many applications such as: in robotics -obstacles avoidance and autonomous routing - [2,3], in pattern recognition, in the detection of aircraft by radar, [1] as well as in many other areas. In the literature, constant attention has been paid to the mentioned issues which led to the development of sophisticated solving techniques based on numerous analytical or heuristic principles, [5,6,7].…”

Section: Introductionmentioning

confidence: 99%

An Implementation of the Radon Transformation for Features Estimation of Image-Objects with Symmetry

Danila¹

2018

RECENT

View full text Add to dashboard Cite

In this paper we develop a method for estimating the direction of symmetry and the mass-centre coordinates of the image objects with symmetry. The proposed method is based on the estimation of the axes of symmetry of the image-object by means of its enclosing equivalent ellipse. The coordinates of the mass-centre and the directions of the main axes of the equivalent ellipse are estimated with the Radon transformation method. Within the paper we propose the implementation of a forgetting-factor-type 2D filter that improves the accuracy of the estimate, especially for the blurred and noisy images.

show abstract

Modified Discrete Radon Transforms and Their Application to Rotation-Invariant Image Analysis

Cited by 24 publications

References 7 publications

A parameter extraction technique for FMCW radar signals using Wigner-Hough-Radon transform

A parameter extraction technique for FMCW radar signals using Wigner-Hough-Radon transform

HDSR-Flor: A Robust End-to-End System to Solve the Handwritten Digit String Recognition Problem in Real Complex Scenarios

An Implementation of the Radon Transformation for Features Estimation of Image-Objects with Symmetry

Contact Info

Product

Resources

About