· · · Institutional Repository

With Wireless Sensors widely used in various domains like home automation, industrial monitoring there is a market urge to deploy the wireless sensors in sporting applications. By deploying wireless sensors in sports, various dimensions of use-case scenarios become obvious which include monitoring sports players to help assess their fitness levels during training sessions and during play, enhance game strategy and provide TV broadcasters with lucrative statistics for the audience. As a first step to realize these use-cases, a platform to create such applications is needed to rapidly prototype devices as a proof of concept. However, to monitor professional sports players and to help make scientific analysis, a deluge of information is needed with less error margin. In this thesis, a wireless sensor platform is designed and developed, customized for creating prototypes of nodes for the sports players. Multiple gateways can be used along the boundary of the play-field to cover the entire playfield and with the mobile sensor nodes making hand-off between the gateways based on their proximity. A time-sharing mechanism is used by the nodes to gain access to the channel and is centralized at the gateway. The gateway provides authentication to which sensor node can transmit data in a round-robin manner. Experimental results show that the packet losses are around 1% with varying cases explaining that only one node communicate with the gateway at any point of time. One of the major drawbacks of such time-slotted protocols is the latency due to the failed nodes and in this protocol a mechanism is devised to mitigate this latency. The net data-rate is also enhanced by transmitting multiple packets in a time-compacted slot without linearly increasing the slot-width.

Object Tracking has been a very active area in the field of C omputer Vision. Over the years, a variety of approaches have been put forth to solve this problem and though many of them have demonstrate considerable success none of them have been completely successful. With more methods being written each day, the evaluation of such systems becomes a very important task. If an evaluation system exists that is able to point out specific flaws in the stage of development, it can lead to a very robust and improved algorithm. This work attempts to create such an evaluation framework. Given an algorithm that detects people and simultaneously tracks them, we evaluate its output by considering the complexity of the input scene. Some videos used for the evaluation are recorded using the Kinect sensor and a benchmark dataset from the PETS workshop is also used. To analyze the performance of the tracking system,the reasons due to which the algorithm might fail are investigated and quantified over the entire video sequence. A set of features called Scene C omplexity Measures are obtained for each input frame. The variability in the algorithm performance is modeled by these complexity measures using various regression models. From the regression statistics, we show that we can compare the performance of two different algorithms and also quantify the relative influence of the scene complexity measures on a given algorithm.