Super-resolution Restoration of Remote-sensing Images

A novel image restoration scheme, which is super-resolution image restoration algorithm Poisson-maximum-afterword-probability based on Markvo constraint （MPMAP） combined with evaluating image detail parameter D, has been proposed. The advantage of super-resolution algorithm MPMAP incorporated with parameter D lies in the fact that superresolution algorithm MPMAP model is discrete, which is in accordance with remote-sensing imaging model, and the algorithm MPMAP is proved applicable to linear and non-linear imaging models with a unique solution when noise is not severe. According to simulation experiments for practical images, super-resolution algorithm MPMAP can retain image details better than most of traditional restoration methods; at the same time, the proposed parameter D can help to identify real point spread function （PSF） value of degradation process. Processing result of practical remote-sensing images by MPMAP combined with parameter D are given, it illustrates that MPMAP restoration scheme combined PSF estimation has a better restoration result than that of Photoshop processing, based on the same original images. It is proved that the proposed scheme is helpful to offset the lack of resolution of the original remote-sensing images and has its extensive application foreground.     

LNN Blind Multi-user Detection Algorithm for Multi-path-fading CDMA Channels

A blind multi-user detection algorithm （BMUD） which is suitable for multi-path-fading Channels based on Lagrange neural network （LNN） is proposed. Based on the minimum output energy （MOE） criterion, the blind detection algorithm is formulated as a constrained optimization problem inherently and is then resolved efficiently using the neural network . Compared with the previous RLS（recursive least squares ）-MOE blind detection algorithm or for multi-path channel, the BMUD based on LNN has better performances： lower computational complexity, faster convergence speed and capability in the multi-path-fading channel. The bit error rate （BER） and signal-to-interference-and-noise ratio （SINR） performances of the detection algorithm in multi-path channel are close to that in single path channel.     

Pautomatic Sea Target Detection Based on Wavelet Transform

An effective automatic target detection algorithm based on wavelet transform, which takes advantage of the localization and the orientation of wavelet analysis, is proposed. The algorithm detects the target in the vertical component of the wavelet transformation of the image. After mutual energy combination and sea clutter suppression through spatial weighting and thresholding, the target is located through maximum energy determination and its size is indicated through similarity measurement function of two overlapping windows. Experiment results show that the target can be detected by the algorithm in a single image frame and the better efficiency can be obtained also under the complicated backgrounds of existing the disturbances of cloud layer and fish scale light.     

A Sub-pixel Image Processing Algorithm of a Detector Based on Staring Focal Plane Array

Optical micro-scanning technology can be used to increase spatial resolution of many optical imaging systems, especially thermal imaging system. One of its key issues is relevant image processing algorithm. A fast reconstruction algorithm is proposed for two dimensional 2 × 2 micro-scanning based on the sub-pixel imaging and reconstruction principle of two-dimensional staring focal plane arrays （FPA）. Specifically, three initialization methods are presented and implemented with the simulated data, their performances are compared according to image quality index . Experiment results show that, by the first initialization approach, timely over-sampled image can be accurately recovered, although special field diaphragm is needed. In the second initialization, the extrapolation approximation in obtaining reconstruction results is better than either bilinear interpolation or over-sampling reconstruction, without requiring any special process on system. The proposed algorithm has simple structure, low computational cost and can be realized in real-time. A high-resolution image can be obtained by low-resolution detectors. So, the algorithm has potential applications in visible light and infrared imaging area.     

Method of Infrared Image Enhancement Based on Stationary Wavelet Transform

Aiming at the problem,i.e.infrared images own the characters of bad contrast ratio and fuzzy edges,a method to enhance the contrast of infrared image is given,which is based on stationary wavelet transform.After making stationary wavelet transform to an infrared image,denoising is done by the proposed method of double-threshold shrinkage in detail coefficient matrixes that have high noisy intensity.For the approximation coefficient matrix with low noisy intensity,enhancement is done by the proposed method based on histogram.The enhanced image can be got by wavelet coefficient reconstruction.Furthermore,an evaluation criterion of enhancement performance is introduced.The results show that this algorithm ensures target enhancement and restrains additive Gauss white noise effectively.At the same time,its amount of calculation is small and operation speed is fast.     

Optimal Time-Frequency Atom Search Based on Adaptive Genetic Algorithm

Adaptive signal decomposition is an important signal processing method.The chirp-based signal representation,for example,the Gaussian chirplet decomposition,has been an active research topic in the field of signal processing.A main challenge of the Gaussian chirplet decomposition is the numerical implementation of the matching pursuit,which is an adaptive signal decomposition scheme,and the challenge remains an open research topic.In this paper,a new optimal time-frequency atom search method based on the adaptive genetic algorithm is proposed,aiming to the low precision problem of the traditional methods.Firstly,a discrete formula of finite length time-frequency atom sequence is derived.Secondly,an algorithm based on the adaptive genetic algorithm is described in detail.Finally,a simulation is carried out,and the result displays its validity and stability.     

Research on Night Vision System Based on Range-Gated Imaging

A design of low-light-level night vision system is described, which can image objects selectively in the specific space. The system can selectively image some objects in specific distances, meanwhile ignore those shelters on the way of observation by combining an intensifying charge coupled device （ICCD） with a near infrared laser assisted in vision, whose operation wavelength matches with the photocathode of the image tube, and adopting the gated mode and adjustable time- delay. A semiconductor laser diode of 100 W in peak power is chosen for illumination. The laser and the image tube operate in 150 ns pulse width and 2 kHz repeat frequency. Some images of different objects at the different distances within 100 m can be obtained clearly, and even behind a grove by using a sampling circuit and a delay control device at 100 W in peak power of semiconductor laser diode, 150 ns in pulse width of laser and image tube, 2 kHz in repeat frequency.     

An Automatic Interference Recognition Method in Spread Spectrum Communication System

An algorithm to detect and recognize interferences embedded in a direct sequence spread spectrum （DSSS） communication system is proposed. Based on Welch＇ s averaging modified periodogram method and fractional Fourier transfor- mation （FRFT）, the paper proposes a decision tree-based algorithm in which a set of decision criteria for identifying different types of interferences is developed. Simulation results demonstrate that the proposed algorithm provides a high recognition rate and is robust for various ISR and SNR.     

A Void Reference Sensor-Multiple Signal Classification Algorithm for More Accurate Direction of Arrival Estimation of Low Altitude Target

There exists MUSIC （multiple signal classification） algorithm for direction of arrival （DOA） estimation. This paper is to present a different MUSIC algorithm for more accurate estimation of low altitude target. The possibility of better performance is analyzed using a void reference sensor （VRS） in MUSIC algorithm. The following two topics are discussed： 1） the time delay formula and VRS-MUSIC algorithm with VRS located on the minus of z-axes; 2） the DOA estimation results of VRS-MUSIC and MUSIC algorithms. The simulation results show VRS-MUSIC algorithm has three advantages compared with MUSIC： 1） When the signal to noise ratio （SNR） is more than - 5 dB, the direction estimation error is 1/2 as much as that obtained by MUSIC; 2） The side lobe is more lower and the stability is better; 3） The size of array that the algorithm requires is smaller.     

Detection Algorithm for LFM Echo of Underwater Moving Targets Based on Discrete Fractional Fourier Transform

The mismatch between echo and replica caused by underwater moving target（UMT）＇s radial velocity degrades the detection performance of the matched filter（MF） for the linear frequency modulation（LFM） signal. By using the focusing property of fractional Fourier transform（FRFT） to that signal, a detection algorithm for UMT＇s LFM echo based on the discrete fractional Fourier transform（DFRFT） is proposed. This algorithm is less affected by the target＇s radial velocity compared with the other MF detection algorithm utilizing zero radial velocity replica（ZRVR）, and the mathematical relation between the output peak positions of these two algorithms exists in the case of existence of target echo. The algorithm can also estimate the target distance by using this relation. The simulation and experiment show that this algorithm＇s detection performance is better than or equivalent to that of the other MF algorithm utilizing ZRVR for the LFM echo of UMT with unknown radial velocity under reverberation noise background.     

一种新型红外焦平面阵列非均匀校正算法

针对现有红外热像产品多采用一点+两点的校正方式,存在增加结构复杂度和整机功耗的问题,开发研究一种新型的红外焦平面阵列非均匀性的校正方法.介绍非均匀性的3种定义,给出非均匀校正的常用方法,在以ULIS0304为试验件的基础上详细论述红外焦平面阵列非均匀性的校正方法,并比较各温度校正前后直方图.结果表明:校正后图像矩阵的非均匀性基本为0,算法有效.     

制冷焦平面高动态范围热成像非均匀校正系数与积分时间的关系

变积分时间是实现高动态范围（HDR）热成像的最有效手段,但积分时间的变化会使得系统的残留非均匀性噪声也发生变化。从红外系统响应模型出发,推导了两点非均匀校正（NUC）法的增益和偏置校正参数与积分时间的关系,结果表明：增益校正参数与积分时间无关,只有偏置校正参数与积分时间有关;并通过实际热成像系统的测试对比实验,证明了上述结论的正确性。研究结果为变积分时间的HDR热成像系统的NUC提供了合理、可行的校正依据,可极大的简化变积分时间的HDR热成像系统NUC处理过程。     

基于时域统计的非均匀校正算法

提出从输入辐射和输出响应的概率密度函数角度分析红外焦平面阵列非均匀性问题,并以此建立基于时域统计特性的自适应非均匀校正算法。增益和偏移系数的漂移被看作是离散的缓慢变化的非平稳随机过程。反映输出响应概率密度函数特性的时域直方图被用来计算探测元的增益和偏移系数;自回归模型被用来对未来的增益和偏移系数进行预测。与直接存储图像数据相比,采用时域直方图存储图像的统计信息可以节约大量的存储空间,而且获得时域直方图的方法也非常简单。把算法设计成流水线结构,便可以实现实时的自动漂移补偿的红外热成像系统。真实的红外实验数据说明该算法有显著的非均匀校正效果。     

基于神经网络的微光图像增强与复原

微光图像的退化主要取决于探測器的灵敏度、非线性和噪声。神经网络是ー个高度复杂的非线性动力学系统,对于微光图像,提出ー种基于多层前向传播（MLP)神经网络的图像增强与复原方法。为了加速训练学习过程,釆用加动量项的自适应BP学习方法对网络进行循环迭代训练。实验结果表明,MLP神经网络增强与复原方法是可行的,并且对微光图像的增强与复原具有很好的效果。     

基于红外目标局部灰度特性分析的管道滤波方法

提出了一种新的管道滤波方法并将其应用于红外图像序列中弱小运动目标检测。首先采用局部灰度特征分析的方法对图像序列的一帧进行逐点检测,检出候选目标;然后对每个候选目标建立管道,利用局部灰度特征的分析方法判断后续帧的管道区域内是否仍然存在目标,保留仍然存在目标特征的管道,清除不具备目标特征的管道。该方法不需要对每一帧图像都进行复杂的管道建立、维持、消除等操作,具有丢失目标后快速重新捕获能力。实测红外图像序列的实验结果证明了所提方法的有效性和抗干扰能力。     

High-resolution Image Reconstruction by Neural Network and Its Application in Infrared Imaging

As digital image techniques have been widely used, the requirements for high-resolution images become increasingly stringent. Traditional single-frame interpolation techniques cannot add new high frequency information to the expanded images, and cannot improve resolution in deed. Multiframe-based techniques are effective ways for high-resolution image reconstruction, but their computation complexities and the difficulties in achieving image sequences limit their applications. An original method using an artificial neural network is proposed in this paper. Using the inherent merits in neural network, we can establish the mapping between high frequency components in low-resolution images and high-resolution images. Example applications and their results demonstrated the images reconstructed by our method are aesthetically and quantitatively (using the criteria of MSE and MAE) superior to the images acquired by common methods. Even for infrared images this method can give satisfactory results with high definition. In addition, a single-layer linear neural network is used in this paper, the computational complexity is very low, and this method can be realized in real time.     

移动机器人路径跟踪的自适应控制

提出一种结构简单的自适应控制器，控制器是由模糊神经网络和PD控制器并行控制移动机器人路径跟踪。在初始阶段，PD控制器控制路径跟踪并提供控制经验给模糊神经网络学习。在学习信号触发器的管理下，可以在线学习自适应调整模糊神经网络的参数。模糊神经网络控制器既推理产生控制规律，也辩识移动机器人动力学模型，通过BP学习算法实时在线调整自身参数达到路径跟踪自适应控制目的。     

飞机目标红外图像序列双阈值分割方法

针对飞机目标红外图像序列,提出了一种新型双阈值分割方法.它依据实测飞机目标红外图像序列的统计特性,以及目标在上一帧图像中所占比例,确定当前帧图像中两个阈值的取值范围;再利用最大熵准则和遗传算法搜索最佳双阈值.仿真结果表明,这种方法能够实时地分割出飞机目标和作为目标特征点的尾喷管,可用于工作在成像目标跟踪阶段的红外成像制导、红外型自动目标识别(ATR)系统.     

无人艇非线性航迹鲁棒自适应跟踪控制

针对欠驱动水面无人艇（USV）的非线性航迹跟踪控制问题,提出一种基于单调3次埃尔米特样条插值（CHSI）和神经网络的鲁棒自适应控制方法。采用CHSI方法对航路点进行拟合,得到光滑的非线性期望航迹,解决了传统线性航迹容易使USV出现的摇摆、曲折问题;引入Serret-Frenet坐标系,并构建了自适应视线制导律,提高了收敛速度且减少了振荡;考虑USV模型的不确定性和环境干扰力影响,设计了简捷的鲁棒自适应神经网络控制器。稳定性分析结果证明了控制系统的收敛性;仿真实例验证了所提出的控制方法能够有效地改善USV航迹跟踪控制的精度和品质,并具有学习参数少、运算负载小的特点。     

基于人眼视觉响应的低光照图像增强

为解决光照图像可视信息不足的问题,提出一种基于人眼视觉响应的图像增强算法.利用导向滤波对图像亮度通道的反相图进行局部平滑计算掩摸,根据亮度通道的全局均值和标准差计算关键参数,对图像光亮度通道进行自适应曲线调整,全面提升动态范围.通过增强前后的亮度通道计算颜色饱和度增益和偏差进行颜色校正,结合灰度直方图分布进一步拉伸全局对比度.采用SSIM、VLD和MOS评价指标对不同图像增强方法结果进行对比.实验结果表明:该算法对低光照图像具有良好的增强效果,能够有效增强图像亮度和细节,提高视觉可视性.