Description

We propose a novel object descriptor, the highorder Local Derivative Pattern (LDP), for robust face recognition. In general, LBP can be conceptually considered as a nondirectional first-order local pattern, which is the binary result of the first-order derivative in images. The second-order LDP can capture the change of derivative directions among local neighbors, and encode the turning point in a given direction. Compared to LBP, the high-order LDP achieved superior performance. Moreover, we propose to extend LDP to feature images. LDP features are directly extracted from gray-level images or feature images without any training procedure. Like LBP, LDP is a micropattern representation which can also be modeled by histogram to preserve the information about the distribution of the LDP micropatterns.