LBP based on multi wavelet sub-bands feature extraction used for face recognition

Abstract

The strategy of extracting discriminant features from a face image is immensely important to accurate face recognition. This paper proposes a feature extraction algorithm based on wavelets and local binary patterns (LBPs). The proposed method decomposes a face image into multiple sub-bands of frequencies using wavelet transform. Each sub-band in the wavelet domain is divided into non-overlapping subregions. Then LBP histograms based on the traditional 8-neighbour sampling points are extracted from the approximation sub-band, whilst 4-neighbour sampling points are used to extract LBPHs from detail sub-bands. Finally, all LBPHs are concatenated into a single feature histogram to effectively represent the face image. Euclidean distance is used to measure the similarity of different feature histograms and the final recognition is performed by the nearest-neighbour classifier.The above strategy was tested on two publicly available face databases (Yale and ORL) using different scenarios and different combination of sub-bands. Results show that the proposed method outperforms the traditional LBP based features.

KurdFace Morph Dataset Creation Using OpenCV

“…Different texture descriptors were used in many different applications, such as Local Binary Patterns (LBP) (Asma & Brahim, 2022), Local Phase Quantization (LPQ) features (Raghavendra et al, 2018), and the Binarized Statistical Image Feature (BSIF) (Adjabi et al, 2021). These texture descriptors are important for more applications, especially face recognition (Rashid et al, 2013), and face morph detection (Damer et al, 2018). The design of our morphing detector is based on the idea that the morphing process changes the variation in the micro_texture of images.…”

Topological data analysis to improve exemplar-based inpainting

“…In facial images,[5] are the first who applied the idea of uniform LBP.It is easy to show that within the (8,1) neighbourhood, there are only 58 uniform LBP-patterns, and the traditional uniform LBP histogram consists of 59 bins accounting for the 58 uniform patterns and one bin that corresponds to the summation of non-uniform patterns[6]. Researchers are taking histograms of LBP-image as a feature for the classification and recognition purposes[7] [8].…”