高光谱数据对损伤长枣的检测判别

灵武长枣作为宁夏优势特色枣果,具有重要的经济社会价值和科学研究意义.利用可见近红外(Vis/NIR)高光谱成像系统采集60颗完整长枣光谱图像,然后利用损伤装置对60颗完整长枣进行损伤实验,最终得到60颗损伤(内部瘀伤)长枣,高光谱成像系统采集损伤后五个时间段(损伤后2,4,8,12和24 h)长枣的光谱图像.对采集的长...     

支持向量机算法对鼻咽癌与正常鼻咽细胞株拉曼光谱分析

拉曼光谱技术在肿瘤与正常细胞株的鉴别方面有着广泛的应用。对一个已有的诊断模型进行可靠性验证是非常重要和有意义的工作。采用两种不同的支持向量机分类算法对鼻咽癌和正常鼻咽细胞株的拉曼光谱进行分析和识别,结果显示灵敏度和特异性均在90%以上,并且与已知的相关线性判别分析结果一致。结论表明,两种支持向量机算法都能较好地对细胞株进行鉴别,同时也表明拉曼光谱技术结合相关统计分类算法的方法可以实现对肿瘤细胞的准确鉴别,这一结果将进一步证实拉曼光谱可以作为鼻咽癌诊断的一种方式。     

基于稀疏表示和学习图正则的高光谱图像特征提取

针对传统局部特征提取算法难以确定邻域参数,以及仅考虑数据间的单一结构而漏掉重要信息的问题,提出一种基于稀疏表示和学习图正则的局部判别与全局稀疏保持投影算法。该算法首先对稀疏表示模型施加基于学习的图正则器,用该改进的稀疏表示模型自适应揭示样本数据间的局部线性结构,通过局部判别模型全局集成算法来提取局部线性结构中的判别信息;利用基于学习图正则稀疏表示模型构建的新型稀疏图来揭示数据间的全局稀疏结构;使得数据的局部判别结构和全局稀疏结构在低维特征空间得以保持。通过1-近邻和支持向量机分类器对实验结果进行评估,在PaviaU和Indian Pines两个高光谱公共数据集上的实验显示,提出的局部判别与全局稀疏保持投影算法较对比算法取得了最好的性能,由于提取了全局和局部的判别信息,有效提升了高光谱图像的地物分类精度。     

不同波长提取方法的高光谱成像技术检测番茄叶片早疫病的研究

提出了基于连续投影算法(successive projections algorithm,SPA)、载荷系数法(x-loading weights,x-LW)和格拉姆-施密特正交(gram-schmidt orthogonalization,GSO)提取特征波长的高光谱成像技术检测番茄叶片早疫病的方法。首先获取380～1 023 nm波段范围内70个健康和70个染病番茄叶片的高光谱图像信息,然后提取健康和染病叶片感兴趣区域(region of interest, ROI)的光谱反射率值,建立番茄叶片早疫病的最小二乘-支持向量机(least squares-support vector machine,LS-SVM)鉴别模型,建模集和预测集的鉴别率都是100%。再通过SPA 、x-LW和GSO提取特征波长(effective wavelengths,EW),并建立EW-LS-SVM和特征波长-线性判别分析(ew-linear discriminant analysis,EW-LDA)鉴别模型。结果显示,每个模型的鉴别效果都很好,EW-LS-SVM模型中预测集的鉴别率都达到了100%,EW-LDA模型中预测集的鉴别率分别是100%, 100%和97.83%。基于SPA, x-LW和GSO所建模型的输入变量分别是4个(492,550,633和680nm),3个(631,719和747 nm)和2个(533和657 nm),较少的特征波长便于实时检测仪器的开发。结果表明,高光谱成像技术检测番茄叶片早疫病是可行的,SPA,x-LW和GSO都是非常有效的特征波长提取方法。     

基于高光谱图像的玉米种子特征提取与识别

玉米种子的形态特征是玉米品种识别的重要因素之一.采用高光谱成像系统获取9个品种共432粒玉米种子的高光谱反射图像,对图像进行校正和预处理,提取每个样本在563.6～911.4nm共55个波段范围内的形状特征.分别利用单波段、多波段和全波段下的玉米种子形状特征结合偏最小二乘判别法进行模型分类.结果显示,全波段范围内训练集和测试集的平均正确识别率达到98.31%和93.98%,均优于多波段和单波段的正确识别率.研究表明,该方法能充分利用高光谱图像中可见光和近红外区域的有效特征信息,较准确地鉴别玉米品种,为玉米品种的自动识别领域提供了一种新方法.     

基于高光谱图像的玉米种子特征提取与识别

玉米种子的形态特征是玉米品种识别的重要因素之一.采用高光谱成像系统获取9个品种共432粒玉米种子的高光谱反射图像,对图像进行校正和预处理,提取每个样本在563.6~911.4 nm共55个波段范围内的形状特征.分别利用单波段、多波段和全波段下的玉米种子形状特征结合偏最小二乘判别法进行模型分类.结果显示,全波段范围内训练集和测试集的平均正确识别率达到98.31%和93.98%,均优于多波段和单波段的正确识别率.研究表明,该方法能充分利用高光谱图像中可见光和近红外区域的有效特征信息,较准确地鉴别玉米品种,为玉米品种的自动识别领域提供了一种新方法.     

甘肃马衔山和田玉化学成分分析和产地判别研究

甘肃马衔山和田玉具有重要的宝玉石价值和考古文化价值,其中产地区分为甘肃马衔山和田玉的研究重点.通过测试分析马衔山和田玉的化学成分,对比其与国内主要产地和田玉的化学成分差异,建立微量元素产地判别模型,对甘肃马衔山和田玉的产地鉴别进行探究,为甘肃齐家文化中古玉器溯源提供数据支撑.以电子探针分析甘肃马衔山和田玉的主量元素,结...     

基于脑部磁共振图像三维局部模式变换特征提取进行阿尔茨海默病病程预测分类

本文提出一种三维局部模式变换提取进行纹理特征并与常规特征相融合的方法,基于脑部磁共振图像,对认知功能正常的健康人体（CN）、轻度认知障碍（MCI）患者和阿尔茨海默病（AD）患者进行预测分类.首先对46例CN对照组、61例MCI患者和25例AD患者的脑部磁共振图像提取感兴趣区域,然后提取双侧海马体组织、灰质和白质的三维局部模式变换纹理特征和常规特征,并将两类特征融合,使用支持向量机分类算法进行分类.结果显示利用本方法,基于双侧海马体组织对AD组和CN组进行分类的准确率为88.73%、敏感度为78.00%、特异度为95.7%、受试者工作特征（ROC）曲线下面积（AUC）为0.886 5;基于灰质的准确率为85.92%、敏感度为80.00%、特异度为86.6%、AUC为0.854 3.这证明基于海马体磁共振图像,利用本文提出的改进三维局部模式变换提取的纹理特征进行阿尔茨海默病病程分类效果较好,融合常规特征后更可提高分类预测的精度.     

空间望远镜主镜的热光学特性分析

主镜是空间望远镜光学系统的重要组成元件,其热光学特性将为光学系统的热控设计提供依据。利用建立在集成分析基础上的热光学分析方法,分析了某空间望远镜主镜在三种不同的温度分布形式,即在径向温度、轴向温度、周向温度梯度的作用下对光学系统成像质量的影响,结合设计的要求,根据计算结果给出了许用温度梯度的范围。     

Study on the effect of sound duration on the annoyance of helicopter noise by applying a technique of time compression and expansion of sound signals

The number of helicopter operations has rapidly increased during the last 20 years in Japan. Helicopter noise sounds different from other aircraft; the waveform of the sound pressure is impulsive and the signal duration is relatively long. The Environmental Agency of Japan implemented new guidelines for evaluating noise exposure around small airports, including heliports, in 1990. This study was executed in connection with the development of provisional guidelines. Psychoacoustic experiments were carried out to identify an evaluation index for helicopter noise. In order to examine the effect of duration independently, we not only used original sound recordings, but also synthesized sound signals. The durations of these sound signals were time compressed or expanded without degrading the quality of the original sound recording. The test results show that the effect of duration is significant, and that the A-weighted sound exposure level is a better index than the maximum A-weighted sound pressure level for the evaluation of helicopter noise.     

FPGA implementation and image encryption application of a new PRNG based on a memristive Hopfield neural network with a special activation gradient

A memristive Hopfield neural network(MHNN)with a special activation gradient is proposed by adding a suitable memristor to the Hopfield neural network(HNN)with a special activation gradient.The MHNN is simulated and dynamically analyzed,and implemented on FPGA.Then,a new pseudo-random number generator(PRNG)based on MHNN is proposed.The post-processing unit of the PRNG is composed of nonlinear post-processor and XOR calculator,which effectively ensures the randomness of PRNG.The experiments in this paper comply with the IEEE 754-1985 high precision32-bit floating point standard and are done on the Vivado design tool using a Xilinx XC7 Z020 CLG400-2 FPGA chip and the Verilog-HDL hardware programming language.The random sequence generated by the PRNG proposed in this paper has passed the NIST SP800-22 test suite and security analysis,proving its randomness and high performance.Finally,an image encryption system based on PRNG is proposed and implemented on FPGA,which proves the value of the image encryption system in the field of data encryption connected to the Internet of Things(Io T).     

Study of Sm-doped ZnO samples sintered in a nitrogen atmosphere and deposited on n-Si(1 0 0) by evaporation technique

The samarium doping zinc oxide (Zn_1-xSm_xO) with (x=0.0, 0.04, 0.05 and 0.17) polycrystalline thin films have been deposited on n-Si(1 0 0) substrate using thermal evaporation technique. Ceramic targets for deposition were prepared by the standard solid-state reaction method and sintered in nitrogen atmospheres. X-ray diffraction and scanning electron microscopy analyses show that the bulk and films features reveal wurtzite crystal structure with a preferential (1 0 1) crystallographic orientation and grows as hexagonal shape grains. According to the results of the Hall effect measurements, all the films show p-type conductivity, possibly a result of nitrogen incorporation into the Sm-doped ZnO samples. Magnetic measurements show that ferromagnetic behavior depends on the Sm³⁺ concentration. For a film with lower Sm₂O₃ contents (x=0.04), a phenomenon of paramagnetism has been observed. While, with further increase of Sm³⁺ contents (x=0.05) the ferromagnetic behavior has been observed at room temperature. However, at higher doping content of Sm³⁺, the ferromagnetic behavior was suppressed. The decrease of ferromagnetism with increasing doping concentration demonstrates that ferromagnetism observed at room temperature is an intrinsic property of Zn_1-xSm_xO films.     

Design and construction of a TPC prototype based on GEM detector readout

A time projection chamber (TPC) readout by gas electron multipliers (GEM) detector is a very promising candidate for the central tracking system of ILC (International Linear Collider). A prototype is designed and set up in our lab and introduced here. The tests during and after the assembly prove that the prototype TPC has been constructed successfully. It is ready for further study.