Wyniki wyszukiwania - BazTech

1

Radon‐wavelet based novel image descriptor for mammogram mass classification

Obaidullah Sk Md, Ahmed Sajib, Gonçalves Teresa, Rato Luís

Journal of Automation Mobile Robotics and Intelligent Systems

|

2020

|

Vol. 14, No. 2

74--80

EN

Mammography based breast cancer screening is very popular because of its lower costing and readily availability. For automated classification of mammogram images as benign or malignant machine learning techniques are involved. In this paper, a novel image descriptor which is based on the idea of Radon and Wavelet transform is proposed. This method is quite efficient as it performs well without any clinical information. Performance of the method is evaluated using six different classifiers namely: Bayesian network (BN), Linear discriminant analysis (LDA), Logistic, Support vector machine (SVM), Multilayer perceptron (MLP) and Random Forest (RF) to choose the best performer. Considering the present experimental framework, we found, in terms of area under the ROC curve (AUC), the proposed image descriptor outperforms, upto some extent, previous reported experiments using histogram based hand‐crafted methods, namely Histogram of Oriented Gradient (HOG) and Histogram of Gradient Divergence (HGD) and also Convolution Neural Network (CNN). Our experimental results show the highest AUC value of 0.986, when using only the carniocaudal (CC) view compared to when using only the mediolateral oblique (MLO) (0.738) or combining both views (0.838). These results thus proves the effectiveness of CC view over MLO for better mammogram mass classification.

2

Problematyka rozpoznawania obrazu opłaty pocztowej w systemach logistycznych poczty

Miciak M., Wiatr R., Talaśka T., Idźkowski M., Rydlewski A.

Logistyka

|

2014

|

nr 6

14461--14468, CD 6

PL

Ważnym problemem, którego rozwiązanie jest niezbędne w celu pełnej automatyzacji przepływu strumienia pocztowego jest weryfikacja opłaty pocztowej. Problem sprowadza się do identyfikacji i rozpoznawania odpowiednich obrazów związanych opłatą. Proces ten jest skomplikowany ze względu na liczne obiekty, które również występują na polu adresowym przesyłek. Stosowane obecnie metody opierają się na analizie odblasku farby fluorescencyjnej i perforacji znaczka naklejonego na liście mogą być zastąpione metodą identyfikacji, w której podstawowym kryterium porównawczym jest informacja o kolorze. W artykule przedstawiono metodę rozpoznawania i weryfikacji opłat pocztowych, w której zaproponowano wykorzystanie informacji o kolorze. Głównym celem artykułu jest zastosowanie transformaty Radona do uzyskania inwariantnych cech obrazu opłaty pocztowej. Ponadto wyszczególniono podstawowe elementy systemu oraz możliwość wykorzystania go do weryfikacji opłat pocztowych. Skupiono uwagę na zagadnieniach związanych z wstępnymi operacjami przetwarzania obrazu. Przedstawiono również przedstawione wyniki analizy proponowanego systemu pod kątem skuteczności jego działania w zakresie identyfikacji opłaty pocztowej.

EN

An important problem whose solution is necessary in order to fully automate the mail flow is verification of postal fee. The problem is focused in to the identification and recognition of appropriate images related with postal fee. This process is complicated because of the many objects that also occur in the address field of mailpieces. Currently used methods are based on the analysis of reflection fluorescent paint and perforation. Those methods may be replaced by the identification, in which the basic criterion is colour. The article presents a method of identifying and verifying the postal fee, which proposed the use of the colour information. The main objective of this paper is the use of the Radon transform to obtain invariant postal image features. In addition, we presented detailed the basic elements of the system and the ability to use it to verify the postage fee. We focused on issues related to the preliminary image processing. It also presents the results of the analysis of the proposed system in terms of the effectiveness of identification postal fee.

3

Analysis of advanced techniques of image processing based on automatic detection system and road sings recognition

Smorawa D., Kubanek M.

Journal of Applied Mathematics and Computational Mechanics

|

2014

|

Vol. 13, nr 1

103--113

EN

This paper describes implementations of the Hough Circle Transform and the Radon Line Transform. The presented transforms were used to build a system for identifying road signs from selected images and recordings. This can be very important and especially useful for monitoring and prevention in driver assistance systems. The design of our system assumed that the objects should be found automatically from images or video sequence. Furthermore, the detection was based on the shapes of common traffic signs which correspond to used transforms. Additionally, the Huffman method was applied to encode the data before being compared. Consequently, the traffic signs were matched directly with the Euclidean distance.

4

Radon-Zernike moments method applied in postal application

Miciak M., Wiatr R., Andrysiak T., Maszewski M.

Zeszyty Naukowe. Telekomunikacja i Elektronika / Uniwersytet Technologiczno-Przyrodniczy w Bydgoszczy

|

2013

|

z. 17

17--26

EN

In this paper a new method of a handwritten characters recognition is introduced. The proposed algorithm is applied to classification of post mails on the basis of postal code information. In connection with this work the research was conducted with numeric characters used in real post code of mail pieces. Moreover, the article contains image processing, for instance, filtration of Radon transformation of the character. The main objective of this article is to use the Radon transform parameter space to obtain a set of moment features on basis of which postal code will be recognized.

PL

W artykule przedstawiono nowe rozwiązanie zadania rozpoznawania znaków pisanych ręcznie dla zastosowań pocztowych. Zaproponowano algorytm klasyfikacji przesyłek pocztowych działający na podstawie informacji zawartej w zapisie kodu pocztowego. Główny nacisk położono na wykorzystanie transformaty Radona i momentów Zernike do uzyskania zbioru cech, na podstawie, których rozpoznawano kod pocztowy. Otrzymane wyniki eksperymentów pozwoliły wykazać skuteczno ść proponowanej metody.

5

On Numerical Reconstruction of a Function from Incomplete Data on Arc Means in Seismic Tomography

Denisiuk A.

Machine Graphics and Vision

|

2011

|

Vol. 20, No. 4

413-437

EN

Consider the problem of reconstruction of a small perturbation of the acoustic wave speed field from traveltime data with linear background slowness. Mathematically, the problem is equivalent to reconstruction of a function from the data of integrals along the circle arcs. The data is limited, in the sense that the base points belong to a compact set. We propose and numerically test a new approach, based on reduction of the problem to the inverse problem for the Radon transform. The data completion procedure is considered as well.

6

Data Matrix in the PCB production

Marchewka A., Zdrojewski J.

Image Processing & Communications

|

2011

|

Vol. 16, no 3-4

31-36

EN

This article presents an approach to Data Matrix code analysis used during preparation of the exposure machine. Tasks realized by the machine are described in a denned way. The choice of description and the verification of film choice correctness are performed auto-matically, based on data from camera images. This method allows automating the settings and transmission of specific data for the performed task.

7

A fast ACOM/EBSD system

Schwarzer R.

Archives of Metallurgy and Materials

|

2008

|

Vol. 53, iss. 1

5-10

EN

Automated backscatter Kikuchi diffraction in the scanning electron microscope allows the microstructure, crystallography and texture of polycrystalline solids to be characterized on the sub-grain size level. It is about to become a tool for process and quality control. Mandatory requirements for these applications are a high speed and measures to enable re-examination of the results at any time. Separate acquisition and store of sequences of diffraction patterns in a raw data format, followed by off-line calculation of the grain orientations, has distinct advantages over conventional on-line EBSD analysis. Aspects of hardware and software for realizing a fast ACOM/EBSD system are discussed.

PL

Automatyzowana rejestracja dyfrakcji Kikuchiego promieni zwrotnych w elektronowym mikroskopie skaninhowym pozwala na scharakteryzowanie mikrostruktury, krystalografię i teksturę polikrystalicznych ciał stalłch w skali rozmiaru subziaren. Technika to staje się narzędziem kontroli procesów i jakości. Obowiązującą normą dla tych zastosowań to duża szybkość oraz możliwość rewizji wyników w dowolnym czasie. Rozdzielenie akwizycji i przechowania sekwencji dyfraktograrnów w formie danych pierwotnych w połączeniu z obliczeniami off-line orientacji ziarn posiada wyraźną przewagę nad konwencjonalną EBSD analizą on-line. Dyskutowane są aspekty hardwarowe oraz softwarowe dla realizacji szybkiego systemu ACOM/EBSD.

8

Discrete Radon transform based invariant object recognition system

Turan J., Farkas P., Siskovicova D.

Image Processing & Communications

|

2002

|

Vol. 8, no 3

21-30

EN

An important problem in pattern analysis is the automatic recognition of an object in scene regardless of the position, size and orientation. In this paper an Invariant Feature Extractor based on the combination of the Radon transform, the Correlation and the fast translation invariant N transform (NT) is described. This Invariant Feature Extractor was used in Invariant Object Recognition System, implemented as a programme package on a powerful PC and tested in recognition experiments on three different classes of objects. The proposed Invariant Object Recognition System has three sub-systems : Digital Image Preprocessing Svstem, Invariant Feature Extractor and Classificator. As Classificator a neural network from the class of ARTMAP neural networks was used.

9

Horocyclic Radon transform on Damek-Ricci space

Abouelaz A., El Fourchi O.

Bulletin of the Polish Academy of Sciences. Mathematics

|

2001

|

Vol. 49, no 2

107-140

EN

We study the horocyclic Radon transform, defined in [5], of Damek-Ricci space. This transform R is obtained by integration over an orbitals family of NA space. We establish a Plancherel's formula for this transform. In particular, we characterize the range of horocyclic Radon transform of certain subspace of [L^2] ( NA, dx). The operators R and R*, where R* is the dual Radon transform of R, are inverted by a differential operator with constant coefficients (if dim NA is odd), or an integro-differential operator (if dim NA is even). The inversion formulas obtained are similar to the inversion formulas in symmetric space of noncompact type (see [7], [18], [19]). Also, we prove a radial Paley-Wiener-Schwartz's theorem for Damek-Ricci space.

10

Irregular pattern recognition using the Hough transform

Żorski W., Foxon B., Blackledge J., Turner M.

Machine Graphics and Vision

|

2000

|

Vol. 9, No. 3

609-632

EN

The paper presents an application of the Hough Transform to the tasks of learning and identifying irregular patterns in computer vision systems. The method presented is based on the Hough Transform with a parameter space defined by translation, rotation and scaling operations. An essential element of this method is the generalisation of the Hough Transform for grey-level images. The technique simplifies application of the Hough Transform to pattern recognition tasks as well as accelerates the calculations considerably. The technique could be used, for example, in a robotic system or for image analysis.