Concave-convex local binary features for automatic target recognition in infrared imagery

Abstract

This paper presents a novel feature extraction algorithm based on the local binary features for automatic target recognition (ATR) in infrared imagery. Since the inception of the local binary pattern (LBP) and local ternary pattern (LTP) features, many extensions have been proposed to improve their robustness and performance in a variety of applications. However, most attentions were paid to improve local feature extraction with little consideration on the incorporation of global or regional information. In this work, we propose a new concave-convex partition (CCP) strategy to improve LBP and LTP by dividing local features into two distinct groups, i.e., concave and convex, according to the contrast between local and global intensities. Then two separate histograms built from the two categories are concatenated together to form a new LBP/LTP code that is expected to better reflect both global and local information. Experimental results on standard texture images demonstrate the improved discriminability of the proposed features and those on infrared imagery further show that the proposed features can achieve competitive ATR results compared with state-of-the-art methods.

Automatic Land Cover Reconstruction From Historical Aerial Images: An Evaluation of Features Extraction and Classification Algorithms

“…10) Soft Concave-Convex Orthogonal Combination of Robust Local Ternary Patterns [62]: The SCCOCRLTP filter is based on the RLTP filter. It proposes to increase the number of discriminative patterns while reducing their memory footprint through the concepts of orthogonal combination [65] and concave-convex discrimination [55]. The idea behind the orthogonal combination is that a concatenation of K histograms obtained from K orthogonal filters on a (P, R) neighborhood should represent the same information than an unique histogram obtained from a complete filter but in a more compact manner (i.e.…”

Local line directional neighborhood pattern for texture classification

“…Wang et al [22] introduced local binary circumferential and radial derivative pattern to capture the global texture features. Sun et al [23] proposed a technique involving concave and convex strategy to improve the robustness of local feature extraction in an image. Song et al [24] introduced adjacent evaluation local binary patterns (AELBPs) that build an adjacent evaluation window around a neighbor for texture classification.…”

Improved Local Ternary Patterns for Automatic Target Recognition in Infrared Imagery

“…In this paper, we focus on local binary pattern (LBP), a simple yet effective approach, for infrared ATR. It also has achieved promising results in several ATR applications in recent years, such as maritime target detection and recognition in [ 19 ], infrared building recognition in [ 20 ], ISAR-based ATR in [ 21 ] and infrared ATR in our previous work [ 22 ].…”

“…In recent years, the LBP operator has been extensively exploited in many applications, such as texture analysis and classification, face recognition, motion analysis, ATR and medical image analysis [ 24 ]. Since Ojala' s original work [ 23 ], the LBP methodology has been developed with a large number of extensions in different fields, such as the extensions from the viewpoint of improving the neighborhood topology [ 25 – 30 ], the extensions from the viewpoint of reducing the impact of noise [ 31 – 34 ], the extensions from the perspective of reducing the feature dimensionality [ 25 , 35 , 36 ], the extensions from the viewpoint of improving the encoding methods [ 22 , 37 – 42 ] and the extensions from the perspective of obtaining rotation invariant property [ 25 , 43 – 46 ].…”