Mahalanobis distance

About: Mahalanobis distance is a research topic. Over the lifetime, 4616 publications have been published within this topic receiving 95294 citations.

Robust Vehicle Detection Through Multidimensional Classification for on Board Video Based Systems

Abstract: This paper presents a new in-vehicle real-time vehicle detection strategy which hypothesizes the presence of vehicles in rectangular sub-regions based on the robust classification of features vectors result of a combination of multiple morphological vehicle features. One vector is extracted for each region of the image likely containing vehicles as a multidimensional likelihood measure with respect to a simplified vehicle model. A supervised training phase set the representative vectors of the classes vehicle and non-vehicle, so that the hypothesis is verified or not according to the Mahalanobis distance between the feature vector and the representative vectors. Excellent results have been obtained in several video sequences accurately detecting vehicles with very different aspect-ratio, color, size, etc, while minimizing the number of missing detections and false alarms.

3D Multi-Layered Normal Distribution Transform for Fast and Long Range Scan Matching

Abstract: Simultaneously Localization and Mapping (SLAM) problem requires a sophisticated scan matching algorithm, in which two consecutive point clouds belonging to highly correlated scene are registered by finding the rigid body transformation parameters when an initial relative pose estimate is available. A well-known scan matching method is the Iterative Closest Point (ICP) algorithm, and the basis of the algorithm is the minimization of an error function that takes point correspondences into account. Another 3D scan matching method called Normal Distribution Transform (NDT) has several advantages over ICP such as the surface representation capability, accuracy, and data storage. On the other hand, the performance of the NDT is directly related to the size of the cell, and there is no proved way of choosing an optimum cell size. In this paper, a novel method called Multi-Layered Normal Distribution Transform (ML-NDT) using various cell sizes in a structured manner is introduced. In this structure a number of layers are used, where each layer contains different but regular cell sizes. In the conventional NDT, the score function is chosen as Gaussian probability function which is minimized iteratively by Newton optimization method. However, the ML-NDT score function is described as the Mahalanobis distance function, and in addition to Newton optimization method, Levenberg–Marquardt algorithm is also adapted to the proposed method for this score function. The performance of the proposed method is compared to the original NDT, and the effects of the optimization methods are discussed. Moreover, an important issue in a scan matching algorithms is the subsampling strategy since the point cloud contains huge amount of data which has a non-uniform distribution. Therefore, the application of a sampling strategy is a must for fast and robust scan matching. In the performance analysis, two sampling strategies are investigated which are random sampling and grid based sampling. The method is successfully applied to experimentally obtained datasets, and the results show that ML-NDT with grid based sampling provides a fast and long range scan matching capability.

An Extended Version of the GustafsonKessel Algorithm for Evolving Data Stream Clustering

Abstract: This chapter contains sections titled: Introduction Cluster Analysis by GK Algorithm Evolving Clustering Based on GK Similarity Distance Methodological Considerations Recursive Computational Aspects Evolving GK-Like (eGKL) Clustering Algorithm Conclusion Acknowledgments References

An adaptive and efficient dimensionality reduction algorithm for high-dimensional indexing

Abstract: The notorious "dimensionality curse" is a well-known phenomenon for any multidimensional indexes attempting to scale up to high dimensions. One well known approach to overcoming degradation in performance with respect to increasing dimensions is to reduce the dimensionality of the original dataset before constructing the index. However, identifying the correlation among the dimensions and effectively reducing them is a challenging task. We present an adaptive multilevel mahalanobis-based dimensionality reduction (MMDR) technique for high-dimensional indexing. Our MMDR technique has three notable features compared to existing methods. First, it discovers elliptical clusters using only the low-dimensional subspaces. Second, data points in the different axis systems are indexed using a single B/sup +/-tree. Third, our technique is highly scalable in terms of data size and dimensionality. An extensive performance study using both real and synthetic datasets was conducted, and the results show that our technique not only achieves higher precision, but also enables queries to be processed efficiently.