Local Derivative Pattern Versus Local Binary Pattern: Face Recognition With High-Order Local Pattern Descriptor

Abstract

This paper proposes a novel high-order local pattern descriptor, local derivative pattern (LDP), for face recognition. LDP is a general framework to encode directional pattern features based on local derivative variations. The n(th)-order LDP is proposed to encode the (n-1)(th) -order local derivative direction variations, which can capture more detailed information than the first-order local pattern used in local binary pattern (LBP). Different from LBP encoding the relationship between the central point and its neighbors, the LDP templates extract high-order local information by encoding various distinctive spatial relationships contained in a given local region. Both gray-level images and Gabor feature images are used to evaluate the comparative performances of LDP and LBP. Extensive experimental results on FERET, CAS-PEAL, CMU-PIE, Extended Yale B, and FRGC databases show that the high-order LDP consistently performs much better than LBP for both face identification and face verification under various conditions.

Evolutionary algorithm based classifier parameter tuning for automatic diabetic retinopathy grading: A hybrid feature extraction approach

“…It has been successfully applied to a wide range of different applications from texture segmentation [34] to face recognition [5]. The LBP feature vector can be determined as explained below.…”

A Multi-Modal Face Recognition Method Using Complete Local Derivative Patterns and Depth Maps

“…Thus, image analysis by the Gabor functions is similar to perception in the human visual system. It was proved in [15,33,34] that Gabor-based and LBP-based features are complementary to each other. LBP features can extract the local texture details, whereas Gabor features can extract texture information on a broader range of scales.…”

“…It has been proved that high-order descriptor can provide more detailed and more discriminative information. A series of high-order local descriptors have been proposed, such as Local Derivative Pattern (LDP) [15], Patterns of Oriented Edge Magnitudes (POEM) [15,16], etc. In the LDP method,( n − 1) th -order derivative images along four fixed directions are obtained first.…”

An overview of content-based image retrieval techniques (CBIR)

“…After calculating the triplet code, each pattern triplets, LBP is decomposed into two codes (Fig 7) [6].  use of Local Derivative Patterns (LDP): Zhang [14], suggested LDP that works on directional patterns based on local derived changes, and be able to describe more detailed encoded information"s which primarily binary patterns (LBP) has no such power. LDP generating a pixel image using the four directions, separately labeled, and 32-bit binary sequence produced by attaching four 8-bit binary sequences.…”