Accurate and fast detection of the moving targets from a moving camera are an important yet challenging problem, especially when the computational resources are limited. In this paper, we propose an effective, efficient, and robust method to accurately detect and segment multiple independently moving foreground targets from a video sequence taken by a monocular moving camera [e.g., onboard an unmanned aerial vehicle (UAV)]. Our proposed method advances the existing methods in a number of ways, where: 1) camera motion is estimated through tracking background keypoints using pyramidal Lucas-Kanade at every detection interval, for efficiency; 2) foreground segmentation is applied by integrating a local motion history function with spatio-temporal differencing over a sliding window for detecting multiple moving targets, while the perspective homography is used at image registration for effectiveness; and 3) the detection interval is adjusted dynamically based on a rule-of-thumb technique and considering camera setup parameters for robustness. The proposed method has been tested on a variety of scenarios using a UAV camera, as well as publically available data sets. Based on the reported results and through comparison with the existing methods, the accuracy of the proposed method in detecting multiple moving targets as well as its capability for real-time implementation has been successfully demonstrated. Our method is also robustly applicable to ground-level cameras for the ITS applications, as confirmed by the experimental results. More specifically, the proposed method shows promising performance compared with the literature in terms of quantitative metrics, while the run-time measures are significantly improved for real-time implementation.

In spite of the success of the stream of variation (SoV) approach to modeling variation propagation in multistation machining processes (MMPs), the absence of machining-induced variations could be an important factor that limits its application in accurate variation prediction. Such machining-induced variations are caused by geometric-thermal effects, cutting-tool wear, etc. In this paper, a generic framework for machining-induced variation representation based on differential motion vectors is presented. Based on this representation framework, machining-induced variations can be explicitly incorporated in the SoV model. An experimentation is designed and implemented to estimate the model coefficients related to spindle thermal-induced variations and cutting-tool wear-induced variations. The proposed model is compared with the conventional SoV model resulting in an average improvement on quality prediction of 67%. This result verifies the advantage of the proposed extended SoV model. The application of the new model can significantly extend the capability of SoV-model-based methodologies in solving more complex quality improvement problems for MMPs, such as process diagnosis and process tolerance allocation, etc.

An agent-based hardware-in-the-loop simulation framework is proposed to model the UAV/UGV surveillance and crowd control system. To this end, a planning and control system architecture is discussed first, which includes various modules such as sensory data collection, crowd detection, tracking, motion planning, control command generation, and control strategy evaluation. The modules that are highly related with agent-based modeling (focus of this paper) are then discussed, which includes the UAV/UGV motion planning considering multi-objectives, crowd motion modeling via social force model, and enhancement of simulation environment via GIS 3D coordinates conversion. In the experiment, Repast Simphony is used as the agent-based modeling tool, which transmits sensory data and control commands with QGroundControl as hardware interface that further conducts radio communications with ArduCopter as a real UAV. Preliminary results show that finer grid scale and larger vehicle detection range generate a better crowd coverage percentage. Finally, conclusions and future works are discussed.