@article{scholars2905,
          volume = {45},
            note = {cited By 23},
          number = {9},
             doi = {10.1016/j.patcog.2012.02.036},
           title = {Novelty detection in wildlife scenes through semantic context modelling},
            year = {2012},
         journal = {Pattern Recognition},
           pages = {3439--3450},
          author = {Yong, S.-P. and Deng, J. D. and Purvis, M. K.},
            issn = {00313203},
             url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84861610344&doi=10.1016\%2fj.patcog.2012.02.036&partnerID=40&md5=1eb7de62bd01b5b6efb9b62f92018a20},
        keywords = {Binarizations; Co-occurrence-matrix; Dimension reduction; Feature extraction and classification; Image blocks; Image data; Image segments; Image sets; Novelty detection; Outliers detection; Scene categories; Scene classification; Scene statistics; Semantic context; Semantic context modelling, Algorithms; Feature extraction; Image segmentation; Principal component analysis; Semantic Web; Semantics, Animals},
        abstract = {Novelty detection is an important functionality that has found many applications in information retrieval and processing. In this paper we propose a novel framework that deals with novelty detection in multiple-scene image sets. Working with wildlife image data, the framework starts with image segmentation, followed by feature extraction and classification of the image blocks extracted from image segments. The labelled image blocks are then scanned through to generate a co-occurrence matrix of object labels, representing the semantic context within the scene. The semantic co-occurrence matrices then undergo binarization and principal component analysis for dimension reduction, forming the basis for constructing one-class models on scene categories. An algorithm for outliers detection that employs multiple one-class models is proposed. An advantage of our approach is that it can be used for novelty detection and scene classification at the same time. Our experiments show that the proposed approach algorithm gives favourable performance for the task of detecting novel wildlife scenes, and binarization of the semantic co-occurrence matrices helps increase the robustness to variations of scene statistics. {\^A}{\copyright} 2012 Elsevier Ltd. All rights reserved.}
}