Local quantized extrema quinary pattern: a new descriptor for biomedical image indexing and retrieval

Abstract

ABSTRACTIn this paper a new feature descriptor ‘local quantised extrema quinary pattern (LQEQryP)’ is proposed for biomedical image indexing and retrieval. The binary and non-binary codings such as local binary patterns (LBP), local ternary patterns (LTP) and local quinary patterns (LQP) encode the gray scale relationship between the centre pixel and its surrounding neighbours in two dimensional (2D) local region of an image, whereas the proposed method encodes the spatial relation between any pair of neighbours in a local region along the given directions (i.e. 0°, 45°, 90° and 135°) for a given centre pixel in an image. The novelty of the proposed method is it uses quinary pattern features from horizontal-vertical-diagonal-anti-diagonal (HVDA7) structure of directional local extrema values of an image to encode more spatial structure information which lead to better retrieval. LQEQryP also provides a significant increase in discriminative power by allowing larger local pattern neighbourhoods. The experiments have been carried out for proving the worth of proposed algorithm on three different types of benchmark biomedical databases; (i) computed tomography (CT) scanned lung image databases named as LIDC-IDRI-CT and VIA/I–ELCAP-CT, (ii) brain magnetic resonance imaging (MRI) database named as OASIS-MRI. The results demonstrate the superiority of the proposed method in terms of average retrieval precision (ARP) and average retrieval rate (ARR) over state-of-the-art feature extraction techniques such as LBP, LTP and LQEP, etc.