Paper Details

Classification of videos has become a challenging task in the multimedia field. Classification of advertisement(ADD) videos from the general programs (NADD) provides an efficient approach to manage and utilize theadvertisement video data. Detection of advertisement video plays a major role for advertisement of content management,advertisement for targeted customers, querying, retrieving, inserting, and skipping the advertisement to view the desiredchannels. Detection of advertisement frames creates an unique application in the multimedia systems. In this paper, theextraction of features that enable identification of advertisement (ADD) videos and non advertisement (NADD) videosdirectly from the TV streams are discussed. The features are extracted using Colour Histogram Features namely, RGBhistogram feature and HSV Histogram feature. The best performing features are identified and selected by decision tree(J48) algorithm and these selected features are used for classification by the K-Star algorithm. The experimental resultsdemonstrate the performance evaluation of K-Star algorithm, the importance of dimensionality reduction and thecomparative study of the K-Star classifier with RGB and HSV histogram features. The K-Star algorithm is performed ina better way and achieved 95.92% for RGB histogram feature and 95.32% for HSV histogram feature of classificationaccuracy is reported for further study

Dr.B.Rebecca Jeyavadhanam, P.Shoba, “Colour Histogram Features for Detection of Advertisement Frames using K-Star Algorithm”, International Journal of Advance Engineering and Research Development (IJAERD), Vol. 6, Issue 7, pp. 161-169, July 2019.