Content based image retrieval using textural features based on pyramid-structure wavelet transform

Xavier, Lidiya; Mary, I. Thusnavis Bella; Raj, W Newton David

doi:10.1109/icectech.2011.5941861

Cited by 16 publications

(4 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Haar wavelet transform is the fastest among all wavelets since its coefficients are either 1 or -1. Thus, they are the less complex, simplest and most widely used wavelets, while Daubechies ones are characterized by fractal structures [25], [26]. The 1st-order and 2nd-order statistics of coefficient sub-bands, such as the mean and standard deviation of wavelet coefficients, are the most commonly used features for texture classification and segmentation problems.…”

Section: Texture Featuresmentioning

confidence: 99%

Performance Evaluation and Benchmarking of Six Texture-Based Feature Sets for Segmenting Historical Documents

Mehri

Mhiri

Héroux

et al. 2014

2014 22nd International Conference on Pattern Recognition

View full text Add to dashboard Cite

Recently, texture-based features have been used for digitized historical document image segmentation. It has been proven that these methods work effectively with no a priori knowledge. Moreover, it has been shown that they are robust when they are applied on degraded documents under different noise levels and types. In this paper an approach of evaluating texture-based feature sets for segmenting historical documents is presented in order to compare them. We aim at determining which texture features could be more adequate for segmenting graphical regions from textual ones on the one hand and for discriminating text in a variety of situations of different fonts and scales on the other hand. For this purpose, six well-known and widely used texturebased feature sets (autocorrelation function, Grey Level Cooccurrence Matrix, Gabor filters, 3-level Haar wavelet transform, 3-level wavelet transform using 3-tap Daubechies filter and 3-level wavelet transform using 4-tap Daubechies filter) are evaluated and compared on a large corpus of historical documents. An additional insight into the computation time and complexity of each texture-based feature set is given. Qualitative and numerical experiments are also given to demonstrate each texture-based feature set performance.

show abstract

Section: Texture Featuresmentioning

confidence: 99%

Performance Evaluation and Benchmarking of Six Texture-Based Feature Sets for Segmenting Historical Documents

Mehri

Mhiri

Héroux

et al. 2014

2014 22nd International Conference on Pattern Recognition

View full text Add to dashboard Cite

show abstract

“…There are many methods of texture extraction we can mentioned co-occurrence matrix (Pianpian et al, 2008), wavelet-based (Xavierm et al, 2011), Gabor Features (Kim & JooSo, 2018) and Local Binary Pattern (LBP) and its variants. From the time when Ojala presents the method to these days, many studies prove that it is considered among the most valuable method.…”

Section: Introductionmentioning

confidence: 99%

Local Binary Pattern Regrouping for Rotation Invariant Texture Classification

Asma

Nini

2022

Journal of Information Technology Research

View full text Add to dashboard Cite

This paper represent a deep study of the Local Binary Pattern (LBP) method and its variants of patterns regrouping , which is largely used in texture classification as well in other domain. The analysis of LBP’s two hundred fifty-six patterns has led us to propose a new organization of uniform and no uniform patterns into twenty-eight groups; each group assembled a number of patterns varied according to specific terms. The principal idea is to preserve the low complexity of LBP and simultaneously increase the method robustness against quality degradation caused by image operations like rotation, grey level changes, illumination and mirror effects. The experiments are done with the two texture databases Outex and Brodatz; the tests are proving the robustness of Local Binary Pattern Regrouping (LBPG) under circumstances.

show abstract

“…These features are compared to the extracted features of the query image. A CBIR typically converts images in feature vector representations and uses them to match similar images [1].…”

Section: Introductionmentioning

confidence: 99%

CBIR Evaluation using Different Distances and DWT

Yadav¹,

Suryawanshi²

2014

IJCA

View full text Add to dashboard Cite

In this paper, evaluation is made on the result of CBIR system based on haar wavelet transform with different distances for similarity matching to calculate the deviation from query image. So, different distances used are Chessboard distance, Cityblock distance and Euclidean distance. In this paper discrete wavelet transform is used to decompose the image. The image is decomposed till sixth level and last level approximate component is saved as feature vector. Comparison is made between different distances to see the best suited distance for CBIR. The wavelet used is "Haar". Haar has compact support and it is the simplest orthogonal wavelet. It is very fast also.

show abstract

Content based image retrieval using textural features based on pyramid-structure wavelet transform

Cited by 16 publications

References 6 publications

Performance Evaluation and Benchmarking of Six Texture-Based Feature Sets for Segmenting Historical Documents

Performance Evaluation and Benchmarking of Six Texture-Based Feature Sets for Segmenting Historical Documents

Local Binary Pattern Regrouping for Rotation Invariant Texture Classification

CBIR Evaluation using Different Distances and DWT

Contact Info

Product

Resources

About