Visual object recognition has become a very popular and important research topic in recent years because of its wide range of applications such as image/video indexing and retrieval, security access control, video monitoring, etc. Despite a lot of efforts and progress that have been made during the past years, it remains an open problem and is still considered as one of the most challenging problems in computer vision community, mainly due to inter-class similarities and intra-class variations like occlusion, background clutter, changes in viewpoint, pose, scale and illumination. Thus the first important step is to generate good visual description, which should be both discriminative and computationally efficient, while possessing some properties of robustness against the previously mentioned variations. In this context, the objective of this thesis is to propose some innovative contributions for object recognition task,
in particular concerning several new visual features/descriptors to
effectively and efficiently represent the visual content of objects
for recognition. The proposed features/descriptors intend to capture
an object's information from different aspects. More precisely, we
propose multi-scale color local binary pattern (LBP) features to
enhance the discriminative power and the photometric invariance
property of the original LBP. We propose the orthogonal combination of
local binary patterns (OC-LBP) for dimensionality reduction of LBP and
use it for local image region description. We introduce the DAISY
descriptor for the task of visual object recognition to efficiently
capture the gradient information. We propose a novel local image
descriptor called histograms of the second order gradients (HSOG) to
capture the second order gradient information which are seldom
investigated in the literature but proven useful for object
recognition. The proposed approaches have been validated through
comprehensive experiments conducted on several popular datasets.