Keynote Speaker 11
| Speaker Name | Dr. Elizabeth Sherly |
| Title of Talk | Classification of Moving Objects In Surveillance Videos Using Deep Neural Network |
Biography
Elizabeth Sherly is the former Director and Professor of Indian Institute of Information Technology and Management-Kerala (IIITM-K). Having 25 years of experience in teaching and research in Computer Science, holds Project Head of different central government projects, Editorial Boards of several journals and Conferences, Director Board member of various government organizations etc. Her research interest includes, Image Processing, NLP, Object Modelling and has a stint interest in cloud computing and Educational Technologies. Member of CSI, ACM and IEEE and has received several outstanding awards in innovative project implementation of Central Government.
Talk description
The use of video is becoming prevalent in many surveillance applications such as detection of pedestrians, identification of anomalous behaviour in a crowded area, activities of the terrorist, smuggling activities, armed robbery, illegal intrusions, monitoring of traffic and more. The talk concentrates on developing an intelligent video surveillance system capable of analyzing and interpreting the video data in terms of objects identification, classification and deducing the hours of video to most significant segments having salient events. Mixture of Gaussian is used for preprocessing, as it models multimodal distribution of background and the foreground pixels are then segmented into regions using the connected components algorithm and identify bounding rectangles around the regions of interest. A deep neural network based object classification is then applied to classify each region of interest as a vehicle, type of the vehicle, or a person.
