神经网络的聚类算法

NSCT非下采样Contourlet变换，实现对图像的多尺度多方向分解，是一种真正的二维图像表示方法(NSCT nonsubsampled Contourlet transform, to achieve multi-scale multi-orientation image decomposition, is a true two-dimensional image representation)

高斯最小移频键控功能以及相关眼图的实现代码(The performance of GMSK )

通过对集成误差公式的理论分析，提出了一种能主动引导个体网络进行差异性学习的集成网络学习算法。该方法通过对集成误差的分解，使个体网络的训练准则函数中包含个体网络误差相关度的因素，并通过协同训练，引导个体网络进行差异性学习。该方法在基于油气分析的变压器故障诊断的实验结果表明，该方法的故障诊断准确率优于传统的三比值法与BP神经网络，其性能也比经典的集成方法Bagging和Boosting方法更稳定可靠。(A learning algorithm is proposed in this paper by analyzing the error function of neural network ensembles, in which individual neural networks are actively guided to learn diversity. By decomposing the ensemble error function, error correlation terms are included in the learning criterion function of individual networks. And all the individual networks in the ensemble are leaded to learn diversity through cooperative training. The method is applied in fault diagnosis of power transformer based on Dissolved Gas Analysis. Experiment results show that, the algorithm has higher accuracy than IEC method and BP network. And the performance is more stable than conventional ensemble method, i.e., Bagging and Boosting. )

RICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE

说明：  目标跟踪的仿真程序，多目标多目标数据融合(Target tracking simulation program, multi-target multi-target data fusion)

fit nan in a figure or plot

用于故障诊断研究的vss程序代码，感兴趣的可以看看(The VSS program code for fault diagnosis research can be viewed as interesting)

超像素分割作为图像处理的预处理部分，能够保留图像的很多基本信息，只是针对颜色和位置进行分割。用基于k-means的SLIC方法与DBSCAN进行超像素分割，matlab实现，demo进入主程序可以运行。

fft去噪，及效果图。供大家参考！欢迎有需要的朋友下载参考(fft bianhuan)

本文对基于非线性控制方法的倒立摆系统控制进行了系统的研究(In this paper, control methods based on nonlinear inverted pendulum control system research)