Wyniki wyszukiwania - BazTech

1

Local Characterisation and Detection of Woven Fabric Texture Based on a Sparse Dictionary

Wu Ying, Wang Ren, Lou Lin, Wang Lali, Wang Jun

Fibres & Textiles in Eastern Europe

|

2022

|

Vol. 30, no. 3

33--40

EN

To achieve enhanced accuracy of fabric representation and defect detection, an innovative approach using a sparse dictionary with small patches was used for fabric texture characterisation. The effectiveness of the algorithm proposed was tested through comprehensive characterisation by studying eight weave patterns: plain, twill, weft satin, warp satin, basket, honeycomb, compound twill, and diamond twill and detecting fabric defects. Firstly, the main parameters such as dictionary size, patch size, and cardinality T were optimised, and then 40 defect-free fabric samples were characterised by the algorithm proposed. Subsequently, the Impact of the weave pattern was investigated based on the representation result and texture structure. Finally, defective fabrics were detected. The algorithm proposed is an alternative simple and scalable method to characterise fabric texture and detect textile defects in a single step without extracting features or prior information.

2

Multichannel seismic impedance inversion driven by logging–seismic data

Li Xinyu, Wang Yaojun, Liu Yu, Cai Hanpeng

Acta Geophysica

|

2021

|

Vol. 69, no. 6

2261--2274

EN

The prior information constrained impedance inversion is an important tool to improve the inversion effect. With the traditional constrained prior information extracted from logging data by the analytic formula, it is difficult to accurately describe the information of a complex reservoir. In addition, the traditional inversion method is trace-by-trace, which ignores the lateral information contained in seismic data. This paper presents a multichannel seismic impedance inversion method combining logging and seismic. In this method, the dictionary learning method is used to extract the vertical prior information of the formation from the logging data. At the same time, we can learn the dip information from seismic data cube. Under the framework of multichannel inversion, regularization and sparse representation technology are used to simultaneously add the vertical and the transverse distribution prior information into the inversion process. Block coordinate descent method is used to solve the multichannel inversion problem, making the seismic inversion efficient. This method excavates the spatial prior information in a data-driven way and is used for constrained inversion, avoiding the false prior cognition caused by manual interpretation. Through the model and field data testing, it is verified that this method is effective.

3

Sparse representation of a non-stationary signal in compressive sensing technique

Pałczyńska Beata, Świsulski Dariusz

Przegląd Elektrotechniczny

|

2019

|

R. 95, nr 11

66--68

EN

The paper presents the application of the compressive sensing technique to reconstruct a non-stationary signal based on compressed samples in the time-frequency domain. A greedy algorithm with different dictionaries to seek sparse atomic decomposition of the signal was applied. The results of the simulation confirm that the use of compressive sensing allows reconstruction of the non-stationary signal from a reduced number of randomly acquired samples, with slight loss of reconstruction quality.

PL

Przedstawiono zastosowanie techniki oszczędnego próbkowania do rekonstrukcji sygnału niestacjonarnego na podstawie skompresowanych próbek w dziedzinie czas-częstotliwość. Zastosowano nadmiarowy algorytm z różnymi słownikami aby znaleźć rzadką reprezentację sygnału. Wyniki symulacji potwierdzają, że zastosowanie oszczędnego próbkowania pozwala na rekonstrukcję sygnału niestacjonarnego z małej liczby losowo pobranych próbek, z niewielką utratą jakości rekonstrukcji.

4

Incoherent Dictionary Learning for Sparse Representation in Network Anomaly Detection

Andrysiak T., Saganowski Ł.

Schedae Informaticae

|

2015

|

Vol. 24

63--71

EN

In this article we present the use of sparse representation of a signal and incoherent dictionary learning method for the purpose of network traffic analysis. In learning process we use 1D INK-SVD algorithm to detect proper dictionary structure. Anomaly detection is realized by parameter estimation of the analyzed signal and its comparative analysis to network traffic profiles. Efficiency of our method is examined with the use of extended set of test traces from real network traffic. Received experimental results confirm effectiveness of the presented method.

5

Semantic Sparse Representation of Disease Patterns

Przelaskowski A.

International Journal of Electronics and Telecommunications

|

2010

|

Vol. 56, No. 3

273-280

EN

Sparse data representation is discussed in a context of useful fundamentals led to semantic content description and extraction of information. Disease patterns as semantic information extracted from medical images were underlined because of discussed application of computer-aided diagnosis. Compressive sensing rules were adjusted to the requirements of diagnostic pattern recognition. Proposed methodology of sparse disease patterns considers accuracy of sparse representation to estimate target content for detailed analysis. Semantics of sparse representation were modeled by morphological content analysis. Subtle or hidden components were extracted and displayed to increase information completeness. Usefulness of sparsity was verified for computer-aided diagnosis of stroke based on brain CT scans. Implemented method was based on selective and sparse representation of subtle hypodensity to improve diagnosis. Visual expression of disease signatures was fixed to radiologist requirements, domain knowledge and experimental analysis issues. Diagnosis assistance suitability was proven by experimental subjective rating and automatic recognition.