Local Ternary Cross Structure Pattern: A Color LBP Feature Extraction with Applications in CBIR

Abstract

With the advent of medical endoscopes, earth observation satellites and personal phones, content-based image retrieval (CBIR) has attracted considerable attention, triggered by its wide applications, e.g., medical image analytics, remote sensing, and person re-identification. However, constructing effective feature extraction is still recognized as a challenging problem. To tackle this problem, we first propose the five-level color quantizer (FLCQ) to acquire a color quantization map (CQM). Secondly, according to the anatomical structure of the human visual system, the color quantization map (CQM) is amalgamated with a local binary pattern (LBP) map to construct a local ternary cross structure pattern (LTCSP). Third, the LTCSP is further converted into the uniform local ternary cross structure pattern (LTCSPuni) and the rotation-invariant local ternary cross structure pattern (LTCSPri) in order to cut down the computational cost and improve the robustness, respectively. Finally, through quantitative and qualitative evaluations on face, objects, landmark, textural and natural scene datasets, the experimental results illustrate that the proposed descriptors are effective, robust and practical in terms of CBIR application. In addition, the computational complexity is further evaluated to produce an in-depth analysis.

Research on Hot Operation of a Petrochemical Plant Based on Compound Edge Operator

“…With the development of computer technology, artificial intelligence technology has become more and more mature (Díaz et al, 2021;Feng et al, 2019;Gordo et al, 2016;Huang et al, 2015). Deep learning consists of deep neural networks, which consist of an input layer, multiple hidden layers, and an output layer.…”

Low-Rank Hypergraph Hashing for Large-Scale Remote Sensing Image Retrieval

“…Retrieving similar RS images from large-scale RS datasets is very important and demanding [1][2][3]. Interestingly, content-based image retrieval (CBIR) [4][5][6] is widely involved in many real-world tasks, such as natural image retrieval and network searches. Nevertheless, large variations are usually contained in the RS images due to their large data volume, small object size and rich background [7,8], and thus how to extract valuable information and further adapt existing CBIR methods to remote sensing image retrieval (RSIR) is considered a key issue [9,10].…”