Object detection by clustering-based nonparametric kernel density estimation

2013 ◽  
Author(s):  
D. Hu ◽  
J. Hu
Keyword(s):  
Object Detection ◽  
Density Estimation ◽  
Kernel Density Estimation ◽  
Kernel Density ◽  
Nonparametric Kernel Density ◽  
Nonparametric Kernel

Application of Nonparametric Kernel Density Estimation in Hongkong Stock Market

2011 ◽  
Vol 55-57 ◽  
pp. 209-214
Author(s):  
Yu Ling Wang ◽  
Jing Wang
Keyword(s):  
Distribution Function ◽  
Stock Returns ◽  
Density Estimation ◽  
Kernel Density Estimation ◽  
Kernel Density ◽  
New Method ◽  
Fat Tails ◽  
Nonparametric Kernel Density ◽  
Nonparametric Kernel ◽  
Better Than

A new method, non-parametric kernel density, is used to research the distribution function of HangSeng index returns. The new method can not only depict the character of peak and fat tails of stock returns, but also capture the market risk better than normal distribution. Further more, more accurate conclusions are concluded.

scholarly journals A fast algorithm for background tracking in video surveillance, using nonparametric kernel density estimation

2005 ◽  
Vol 18 (1) ◽  
pp. 127-144 ◽  
Author(s):  
Codrut Ianasi ◽  
Vasile Gui ◽  
Corneliu Toma ◽  
Dan Pescaru
Keyword(s):  
Real Time ◽  
Density Estimation ◽  
Video Surveillance ◽  
Kernel Density Estimation ◽  
Computing Time ◽  
Kernel Density ◽  
Surveillance Systems ◽  
Event Analysis ◽  
Nonparametric Kernel Density ◽  
Nonparametric Kernel

Moving object detection and tracking in video surveillance systems is commonly based on background estimation and subtraction. For satisfactory performance in real world applications, robust estimators, tolerating the presence of outliers in the data, are needed. Nonparametric kernel density estimation has been successfully used in modeling the background statistics due to its capability to perform well without making any assumption about the form of the underlying distributions. However, in real-time applications, the O(N2) complexity of the method can be a bottleneck preventing the object tracking and event analysis modules from having the computing time needed. In this paper, we propose a new background subtraction technique, using multiresolution and recursive density estimation with mean shift based mode tracking. An algorithm with complexity independent on N is developed for fast, real-time implementation. Comparative results with known methods are included, in order to attest the effectiveness and quality of the proposed approach.

scholarly journals Adaptive Nonparametric Kernel Density Estimation Approach for Joint Probability Density Function Modeling of Multiple Wind Farms

Energies ◽  
2019 ◽  
Vol 12 (7) ◽  
pp. 1356 ◽  
Author(s):  
Nan Yang ◽  
Yu Huang ◽  
Dengxu Hou ◽  
Songkai Liu ◽  
Di Ye ◽  
...  
Keyword(s):  
Power Systems ◽  
Density Estimation ◽  
Kernel Density Estimation ◽  
Joint Probability ◽  
Kernel Density ◽  
Wind Farms ◽  
Nonparametric Kernel Density ◽  
And Control ◽  
Nonparametric Kernel ◽  
Joint Probability Density

The uncertainty of wind power brings many challenges to the operation and control of power systems, especially for the joint operation of multiple wind farms. Therefore, the study of the joint probability density function (JPDF) of multiple wind farms plays a significant role in the operation and control of power systems with multiple wind farms. This research was innovative in two ways. One, an adaptive bandwidth improvement strategy was proposed. It replaced the traditional fixed bandwidth of multivariate nonparametric kernel density estimation (MNKDE) with an adaptive bandwidth. Two, based on the above strategy, an adaptive multi-variable non-parametric kernel density estimation (AMNKDE) approach was proposed and applied to the JPDF modeling for multiple wind farms. The specific steps of AMNKDE were as follows: First, the model of AMNKDE was constructed using the optimal bandwidth. Second, an optimal model of bandwidth based on Euclidean distance and maximum distance was constructed, and the comprehensive minimum of these distances was used as a measure of optimal bandwidth. Finally, the ordinal optimization (OO) algorithm was used to solve this model. The scenario results indicated that the overall fitness error of the AMNKDE method was 8.81% and 11.6% lower than that of the traditional MNKDE method and the Copula-based parameter estimation method, respectively. After replacing the modeling object the overall fitness error of the comprehensive Copula method increased by as much as 1.94 times that of AMNKDE. In summary, the proposed approach not only possesses higher accuracy and better applicability but also solved the local adaptability problem of the traditional MNKDE.

Sign in / Sign up

Export Citation Format

Share Document