Directional local quinary patterns: A new feature descriptor for image indexing and retrieval

Abstract

A novel evaluation algorithm using directional local quinary pattern (DLQP) for image retrieval (IR)was proposed here. The proposed method extracts the directional edge information between reference pixel and its surrounding neighbors by computing gray-level difference based on local quinary value (-2,-1, 0, 1, 2) instead of binary and ternary values in 0°, 45°, 90°, and 135° directions of an image which are absent in (Ojala, Pattern Recogn 29:51–59, 1996), (Takala, Block-based methods for image retrieval using local binary patterns, SCIA 2005, LNCS 3450:822–891, 2005), (Baochay, Pattern Recogn:2337–2344, 2010), (Marko, Pattern Recogn:42:425–436, 2009). The performance of the proposed method is tested on Corel 5000 (DB1) database.The performance yields better recognition ratewhen compared with standard local binary pattern (LBP), center-symmetric local binary pattern (CS-LBP), directional binary pattern (DBC), and other existing transform domain methods. © Springer India 2014.