共查询到10条相似文献,搜索用时 46 毫秒
1.
由于海洋动力学和目标起伏引起发射信号包络的时变衰减,相应于频域就是频率扩展,这种信道称为频率扩展信道,又名快速衰落信道(FFD)。在FFD信道中,理想信道的最佳检测器——匹配滤波器的性能将有所下降,Baggenstoss已经证明FFD信道的最佳检测器是分段副本相关器(SRC)。近年来出现的小波变换具有和宽带相关处理相似的计算结构,所以文章拟从连续小波变换的理论和性质出发,推导FFD信道小波变换域最佳检测器的形式,并针对一个仿真算例,进一步用Monte-Carlo实验计算出该检测器的性能曲线,结果表明该检测器的性能优于小波域的副本相关器。 相似文献
2.
基于阈值判断的自适应中值滤波算法 总被引:1,自引:0,他引:1
针对标准的中值滤波算法在去除噪声与保留图像细节方面难以取舍的缺陷,在自适应中值滤波算法的基础上提出了一种改进的基于噪声点检测的自适应中值滤波算法.该算法在进行噪声点检测时采用了一种阈值判断法,充分利用了当前像素点与邻域像素点的灰度值之间的关系.结果表明,在噪声浓度较高时仍然可以区分噪声点与边缘点,滤波的同时有效地保护了图像的细节. 相似文献
3.
4.
Abstract A sinusoidal detector with adaptive thresholding CFAR (constant‐false‐alarm‐rate) processor operating in noise of changing statistics is analyzed for some very general situations. The detector contains FFT and a cell‐averaging processor, which can be easily implemented for real time applications, such as in pulse doppler radar. For signals in stationary noise in particular, the second‐order statistics of the FFT‐based periodogram of the windowed data are derived. In the AWGN case, this detector possesses the desired CFAR characteristics. For Gaussian colored noise, the detector performance is derived analytically with certain reasonable approximations. It is shown from some numerical examples that in this case the variation of the false alarm probability with respect to the unknown sinusoidal frequency and noise spectrum is not significant, which reveals the property of robustness. Moreover, the detection probability is satisfactory for most cases. 相似文献
5.
复杂背景下多目标提取的高灵敏度方法 总被引:1,自引:0,他引:1
对电视经纬仪拍摄到的飞行小目标图像进行了归一化高斯模板相关。传统的模板相关方法不能将真假目标集合充分分离,导致真目标的漏检或引入虚假目标。基于模板相关的高斯函数拟合方法(Gaussian Function Fitting Method, GFFM),对模板相关所得到的目标集(含有真目标和假目标)中的每一个元素进行高斯函数拟合,并引入了一个更为灵敏的检验量—高斯函数拟合误差,可以将真假目标集合明显区分开,减小阈值确定的难度。实验表明:当相关系数阈值rth=0.8时,传统模板相关方法漏检率20%,虚警率40%;而GFFM方法则检出了所有真目标,且无虚假目标。 相似文献
6.
7.
8.
D. S. McGregor M. D. Hammig Y. -H. Yang H. K. Gersch R. T. Klann 《Nuclear instruments & methods in physics research. Section A, Accelerators, spectrometers, detectors and associated equipment》2003,500(1-3):272-308
Semiconductor-based thermal neutron detectors provide a compact technology for neutron detection and imaging. Such devices can be produced by externally coating semiconductor-charged-particle detectors with neutron reactive films that convert free neutrons into charged-particle reaction products. Commonly used films for such devices utilize the 10B(n,)7Li reaction or the 6Li(n,)3H reaction, which are attractive due to the relatively high energies imparted to the reaction products. Unfortunately, thin film or “foil” type thermal neutron detectors suffer from self-absorption effects that ultimately limit neutron detection efficiency. Design considerations that maximize the efficiency and performance of such devices are discussed. Theoretical and experimental results from front coated, back coated, and “sandwich” designs are presented. 相似文献
9.
一种尺度自适应Canny边缘检测方法 总被引:5,自引:1,他引:5
作为最优边缘检测算子,Canny算子在线性滤波及选取阈值方面具有一定的局限性.在其基础上,采用尺度可变的自适应高斯滤波器进行滤波,通过计算滤波器各滤波窗口内像素点灰度值的均值,以及该均值与当前像素点灰度值的差值,确定滤波窗口的平滑度,并将其作为高斯函数尺度的大小,在滤除噪声的同时确保了边缘信号不被平滑掉;同时采用基于简单统计法及梯度阚值法的双阈值分割方法检测经过非极大值抑制的梯度幅值图像的边缘.实验结果表明,该方法能够有效滤除噪声,且假边缘及丢失边缘的数量都比Canny边缘检测的少. 相似文献
10.
LoG算子作为一种零交叉检测器对局部灰度变化敏感,不易受非均匀光场的干扰,可检测出能较准确反映局部灰度变化的二阶过零点。利用对LoG算子检出的零交叉点的插值可求出动态阈值曲面,并对图像进行二值化,能有效克服非均匀光场对图像分割结果所带来的影响。 相似文献