Output details
11 - Computer Science and Informatics
University of Westminster
Multidimensional image selection and classification system based on visual feature extraction and scaling
<23> Originality: The paper is concerned with the identification of objective methods of pictorial image classification that correlate with subjective classification. It describes the way inherent scene features, can be combined to provide an answer to image selection and classification, which is compatible with human observations. A novel multidimensional system is built based upon such global scene features.
Significance: The paper proposes a solution to an important aspect in designing pictorial image databases.
Rigour: The extraction of relevant feature metric values was achieved by implementing existing and proposed novel algorithms.