Abstract The increasing pervasiveness of location-acquisition technologies has embedded collection of huge amount of trajectories for almost any kind of moving objects. Discovering useful patterns from their movement behaviors can convey valuable knowledge to a variety of critical applications. In this light we propose a concept, called gathering, which is a trajectory pattern modeling various group incidents such as celebrations, parades, protests,... The increasing pervasiveness of location-acquisition technologies has embedded collection of huge amount of trajectories for almost any kind of moving objects. Discovering useful patterns from their movement behaviors can convey valuable knowledge to a variety of critical applications. In this light we propose a concept, called gathering, which is a trajectory pattern modeling various group incidents such as celebrations, parades, protests, traffic jams and so on. In this work, we first develop a set of techniques to tackle the challenge of efficient discovery of gathering patterns on archived trajectory dataset. For finding gathering firstly we have to find snapshot cluster, crowd and then super crowd. After getting super crowd, gathering can be identified. Here we proposed an efficient algorithm for finding super crowd if cluster database is given and algorithm for identify gathering after identifying the super crowd. Afterwards, since trajectory databases are inherently dynamic in many real-world scenarios such as traffic monitoring, fleet management and battlefield surveillance, we further propose discovery solution by applying a series of optimization schemes to handle the incremental data. Read More Read Less

Abstract Information exists in multiple media forms, i.e. in the form of text, audio, image, and video objects called multiple media information. Textual, acoustic, and visual modalities of information associated with multiple media information. Existing search approaches broadly categorized as media specific and multiple media based. Mostly the multiple media search approaches not focused towards aggregated search, blended integration of the search results, and... Information exists in multiple media forms, i.e. in the form of text, audio, image, and video objects called multiple media information. Textual, acoustic, and visual modalities of information associated with multiple media information. Existing search approaches broadly categorized as media specific and multiple media based. Mostly the multiple media search approaches not focused towards aggregated search, blended integration of the search results, and user navigation/ browsing of the search space via multiple modalities of information. In this research, we will propose a novel framework to implement Next Generation of multiple media search systems. The framework supports the implementation of search systems that provides aggregated search, blended integration of the search results, and navigation within the database of media objects by exploiting multiple modalities of information associated with media objects. The proposed search framework instantiated over real scenarios. Read More Read Less

Abstract Depth information plays an important role in many computer vision applications including moving object detection. In this paper, we introduce a fast moving object detection method using Kinect v2. Comparing to Kinect v1, the Kinect v2 provides a higher resolution for RGB images and adopts a ToF (Time-of-Flight) sensing mechanism for depth measurement. The upgraded depth sensing method yields depth images with less... Depth information plays an important role in many computer vision applications including moving object detection. In this paper, we introduce a fast moving object detection method using Kinect v2. Comparing to Kinect v1, the Kinect v2 provides a higher resolution for RGB images and adopts a ToF (Time-of-Flight) sensing mechanism for depth measurement. The upgraded depth sensing method yields depth images with less noises and holes, which allows us to simplify the pre-processing of eliminating noises and filling holes. Also, the depth information of the Kinect v2 is fruitful enough to detect the moving object without resorting to RGB image. These properties of the Kinect v2 lead us to devise a fast moving object detection method. Read More Read Less