Rotation Invariant Image Description with Local Binary Pattern Histogram Fourier Features

By Timo Ahonen, Jiri Matas, Chu He, Matti Pietikäinen. 2009

470 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

SUN database: Large-scale scene recognition from abbey to zoo (2010, background): “We also try the rotation invariant extension version [2] of LBP to examine whether rotation invariance is suitable for scene recognition.”
Multi-class texture analysis in colorectal cancer histology (2016, methodology): “Local texture descriptors such as GLCM 9 or LBP 26,27 are second-order statistics which consider the joint variability of the grey levels of pairs or groups of pixels.”
SUN Database: Exploring a Large Collection of Scene Categories (2014, methodology): “Ahonen et al. (2009) also extended this approach to be a rotation invariant image descriptor, called Local Binary Pattern Histogram Fourier (LBP-HF). We try this descriptor to examine whether rotation invariance is suitable for scene recognition. For both LBP and LBP-HF, we use histogram intersectio”
Deep Multitask Learning for Railway Track Inspection (2015, background): “• LBP-HF with approximate Nearest Neighbor: The Local Binary Pattern Histogram Fourier descriptor introduced in [43] is invariant to global image rotations while”
A Training-free Classification Framework for Textures, Writers, and Materials (2012, methodology): “Alternatively, [14] uses modified SIFT and intensity domain SPIN images, [13] plain SIFT, [6, 20, 23] variants of LBP descriptors, [31] a grayscale image patch for its Joint descriptor and an MRF representation; [36] local fractals, and [18, 19] (sorted) random projections over small patches.”
Combined scattering for rotation invariant texture analysis (2012, methodology): “This angular distribution may be captured by computing a Fourier transform along the rotation parameter, whose modulus is rotation invariant, as in LBP-HF[5].”
An Efficient Algorithm for Fractal Analysis of Textures (2012, methodology): “The two remaining datasets, KTH-TIPS [6] and Textured Surfaces [7] are publicly available and widely employed to evaluate texture analysis methods [8], [9], [10].”
Texture Feature Extraction Methods: A Survey (2019, background): “Other rotation invariant versions have also been proposed [81]–[84].”
Pairwise Rotation Invariant Co-Occurrence Local Binary Pattern (2014, background): “Ahonen et al. [26] proposed a LBPHF_S_M, an extension of LBPHF, by combining the sign and magnitude information.”
Steel defect classification with Max-Pooling Convolutional Neural Networks (2012, background): “• Local Binary Pattern Histogram Fourier (LBP-HF) [15]: is a rotation invariant descriptor computed from discrete Fourier transforms of LBP.”
Weakly supervised histopathology cancer image segmentation and classification (2014, methodology): “Specifically, we use widely adopted features including L a b Color Histogram, Local Binary Pattern (Ojala et al., 2002; Ahonen et al., 2009), and SIFT (Lowe, 2004).”