Sentinel-2卫星的多光谱轻量级船舶目标检测算法

近年来深度卷积神经网络在可见光船舶检测方面取得了显著的进展，然而，大多数相关研究是通过改进大型的网络结构来提高检测性能，因此加大了对更高计算机性能的需求。此外，可见光图像难以在云、雾、海杂波、黑夜等复杂场景检测到船舶。针对以上问题，提出了一种融合红（red, R）、绿（green, G）、蓝（blue, B）和近红外（NIR）4个波段光谱信息的由粗到精细的轻量型船舶检测算法。与现有的方法中根据光谱特性利用水体检测算法提取水体区域不同之处是该算法是利用改进的水体检测算法来提取船舶候选区域。为获取更准确的候选区域，对船舶、厚云、薄云、平静海面、杂波海面5种场景中4个波段的像素值进行了统计分析，选取近红外大于阈值作为辅助判断，并以其中心点获取候选区域32×32大小的切片，并对切片进行非极大值抑制，由此获得了船舶粗检测结果。随后构建了轻量级LSGFNet网络对船舶候选区域切片进行精细识别。构建的网络融合了1×1卷积提取的波谱特征与3×3的提取几何特征，为防止光谱特征与几何特征的信息在融合时“信息不流通”，在LSGFNet网络中引入了ShuffleNet中的通道打乱机制，并减小了模型结构，与典型的轻量级网络相比具有更好的效果且模型较小。最后，利用Sentinel-2卫星多光谱10 m分辨率数据构建了512×512大小的1 120组数据进行粗检测，以及32×32大小的6 014组数据进行精细网络训练，其中候选区域粗提取的查全率为98.99%，精细识别网络精确度为96.04%，不同场景下的平均精确度为92.98%。实验表明该算法在抑制云层、海浪杂波等干扰的复杂背景下具有较高的检测效率，且训练时间短、计算机性能需求低。     

New insights into the anti-hepatoma mechanism of Alisol G-metal ions complexes based on c-myc DNA

This study investigated the anti-hepatoma molecular mechanism of Alisol G, which is an effective component of the Chinese medicine Alisma orientalis, in the presence of metal ions Cu²⁺ and Fe³⁺ based on c-myc DNA. Here, a combination of Alisol G and metal ions (Cu²⁺, Fe³⁺) to augment anti-hepatoma efficiencies of Alisol G has been identified by methyl thiazolyl tetrazolium (MTT) assay. Network pharmacology revealed that c-myc DNA was the potential target of Alisol G with respect to its anti-hepatoma effects. By performing multi-spectroscopic analyses, we showed that the interaction of Alisol G with c-myc DNA was a process of static quenching. The binding constants and thermodynamic constants indicated that a 1:1 complex was formed between Alisol G and c-myc DNA. Moreover, metal ions strengthened the interaction between Alisol G and c-myc DNA. Molecular docking and molecular dynamics simulation further unveiled that the higher binding affinity between Alisol G-Fe³⁺ complex and c-myc DNA as compared to Alisol G-Cu²⁺ complex. This probably resulted from the polarization of metal ions and the structural flexion of Alisol G. The C22-O31-H76 and C18-O32-H77 of Alisol G were key groups in the interaction with c-myc DNA. Addition of metal ion, had greatly changed the c-myc DNA-binding domain of Alisol G while didn’t affect the kinetic stability of the interaction, thus facilitating the insertion of Alisol G into c-myc DNA A-T base pair. Importantly, the DG113 of c-myc DNA was important for its binding to metal ions. Together, our findings suggested that Alisol G in combination with metal ions may be an efficient and promising option for the treatment of liver cancer.     

基于无人机多光谱图像的土壤水分检测方法研究

以表层土壤为对象,探究土壤的多光谱反射率与土壤水分含量相关性,进行基于无人机多光谱图像的土壤水分含量预测模型方法的探究。选取中国农业大学通州实验站为研究区域,实地采集试验田的土壤样本100组,按照一定梯度配制土壤含水量,配成的土壤含水率为10%~50%之间,土壤含量的真实值采用土壤烘干法进行测定。多光谱相机灵巧便捷,可搭载在无人机上对土壤进行监测。将RedEdged-M型多光谱相机搭载在Phantom 3型无人机上,选择阳光充足的采集环境,实时采集土壤样本的多光谱图像,建立土壤多光谱信息与水分含量之间的模型。利用处理光谱数据的ENVI5.3软件提取土壤样本多光谱信息,以多光谱相机自带的标准白板反射率为100%,计算出土壤样本在蓝、绿、红、红边、近红外五个波段的光谱反射率。采用BP神经网络算法、支持向量机算法、偏最小二乘算法分别建立基于无人机多光谱图像的土壤水分含量的预测模型。以80组土壤样本数据作为训练集,建立基于多光谱图像的土壤水分含量预测模型。采用莱文贝格-马夸特算法对BPNN进行改进,提高了其训练速度,当网络结构为5-10-1时,训练效果最好,本文选择该网络结构;SVM采取高斯核函数,当参数为0.56时,模型效果最好。本研究采用归一化均方根误差(NRMSE)和决策系数(R 2)对三种土壤水分含量的预测模型进行定量对比。以20组土壤样本数据作为测试集,结果可知,基于BP神经网络土壤水分含量预测模型的NRMSE为0.268,R 2为0.872;基于支持向量机的土壤水分含量预测模型的NRMSE为0.298,R 2为0.821;基于偏最小二乘土壤水分含量预测模型的NRMSE为0.316,R 2为0.789。对三种模型分析可知,基于BPNN的土壤水分含量预测模型效果均较好。结果可知,土壤的光谱反射率与含水率间存在较密切的相关性,将多光谱相机搭载在无人机上可以对土壤水分含量进行有效的实时监测,对监测土壤墒情提供技术支持和理论支撑。     

Euclidian distance-weighted smoothing for quantitative MRI: application to intervoxel anisotropy index mapping with DTI

During the computation of intervoxel anisotropy features, the inclusion of both eigenvalues and eigenvectors reduces the effect of noise, but spatial averaging blurs the resulting maps. We propose a new adaptive technique that uses data-dependent weights in the averaging process so that the influence of each neighbor in the local window is proportional to the similarity of characteristics of the neighbor considered to those of the reference central voxel. This likeness criterion is based on the multidimensional Euclidian distance using the entire available multispectral information contained in the diffusion-weighted images. This solution is controlled by a single parameter beta that results from a compromise between edge-preserving and noise-smoothing abilities. This Euclidian distance-weighted technique is a generic solution for filtering noise during parametric reconstruction. It was applied to map anisotropy using an intervoxel lattice index (LI) from experimental images of mouse brain in vivo and achieves noise reduction without distorting small anatomical structures. We also show how to employ in the discrimination scheme the images not used in the estimation of the considered feature.     

Luminance adaptation model for increasing the dynamic range of an imaging system based on a CCD camera

We propose a luminance adaptation model (LAM) to increase the dynamic range of an imaging system when scenes containing areas of low and high illumination are imaged. The LAM that we developed is based on capturing images at different exposure times to obtain digital levels within the linear response zone for all the pixels in the image. The levels are subsequently transformed to a reference exposure time that is common to all pixels. We use a linear transformation whose coefficients are determined by the digital levels obtained for a set of flat-spectrum samples. In this study, the LAM is applied to a multispectral imaging system that is based on a CCD camera used for color measurements and spectral reconstructions. It is shown to be a very useful method for increasing the dynamic range of the system, whilst maintaining its accuracy.     

多光谱成像无损识别冻融猪肉中危害级碎骨

冻融猪肉作为肉制品加工原料,被广泛应用于无骨肉制品加工。该原料中的危害级碎骨（1～2.5 cm）对后期加工及食用安全均有较大风险。因此,开展多光谱成像技术（405～970 nm）快速无损识别冻融猪肉中碎骨的可行性研究十分必要。将195块肉片制备成65个无骨肉样、65个碎骨表面嵌入式肉样和65个碎骨内部嵌入式肉样,经冻融处理后采集其多光谱图像;再利用经典判别分析（CDA）进行图像分割,获得两类感兴趣区域（ROIs-1和ROIs-2）,并提取相应光谱和图像信息;最后运用支持向量机（SVM）和神经网络（NN）建立冻融猪肉危害级碎骨识别模型。结果显示：ROIs-2全光谱比ROIs-1全光谱有更好的识别能力,SVM和NN模型的精度均为100%,表明区域分割与模型精度密切相关。基于连续投影算法（SPA）筛选出六个关键波长（505,590,700,850,890和970 nm）,所提取的ROIs-2特征光谱可实现样品碎骨高精度识别,准确率为100%,进一步提升了识别效率。利用图像信息既能建立优越的SVM和NN碎骨识别模型,准确率分别为93.8%和93.33%,又能实现结果可视化,体现出优良的技术优势,但精度低于光谱识别模型。综上所述,多光谱成像技术可实现冻融猪肉危害级碎骨的高精度识别,为工业在线检测提供理论基础。     

Multispectral and hyperspectral image analysis of elemental and micro-Raman maps of cross-sections from a 16th century painting

Spectroscopic imaging is well suited to the study of micro-samples from artworks, where the sample material is limited and the maximum amount of information needs to be obtained. In this study, a new approach to imaging elemental data from energy dispersive X-ray analysis maps was used in conjunction with micro-Raman spectroscopic imaging to characterise the paint layers within micro-samples. Cross-sections from the 16th century painting Portrait of a Youth were found to contain vermilion, lead-tin yellow type 1 and a blue-green pigment consistent with terre-verte. The mid-preparatory layer (imprimatura) contains a high proportion of elements and mineral inclusions that indicates a clay-type composition. The ground layer was identified as anhydrite with large gypsum inclusions. The pigments and composition of the preparatory layers are consistent with those used by Italian Renaissance artist Dosso Dossi.     

双卤化物共掺新型硫系多光谱玻璃的制备及性能

利用传统熔融-淬冷工艺制备了65GeS_2-15Ga_2S_3-(20-x)CsCl-xCsI(x=0,5,10,15,20)系列硫卤玻璃;通过测试该系列玻璃样品的密度、显微硬度、可见/近红外吸收光谱、红外透射光谱、喇曼谱、XRD衍射谱、玻璃转变温度等,对其进行了系统研究.结果表明:该系列玻璃具有较宽的成玻范围,在0.42~12μm范围内具有良好的透过率;随着CsI含量(mol%)的增加,玻璃的密度逐渐增大;随着CsCl含量(mol%)的增加,光学带隙以及硬度逐渐增大;玻璃样品的玻璃转变温度随Cl~-、I~-的共掺比例发生明显变化,当Cl~-(mol%):I-(mol%)=1时,玻璃转变温度最低.     

多光谱法结合分子对接研究柠檬黄与牛血清白蛋白的相互作用

采用荧光光谱法、同步荧光光谱法、三维荧光光谱法、紫外光谱法以及分子对接法研究了柠檬黄与牛血清白蛋白(BSA)的相互作用.柠檬黄与BSA相互作用的荧光光谱分析表明柠檬黄能有效猝灭BSA的内源荧光,根据Stern-Volmer方程计算得到柠檬黄对BSA的荧光猝灭常数KSV随着温度的升高而逐渐降低,表明柠檬黄对BSA的荧光猝...     

多目标普朗克极小值优化法的多光谱真温反演研究

光谱发射率是辐射体辐射能力的重要参数,通过光谱发射率可以建立辐射体与黑体的之间的桥梁,从而黑体辐射的相关理论就可以应用于辐射体。采用普朗克公式,光谱高温计的每一个光谱通道可以构成一个方程,这个方程中包含有真温、亮度温度和光谱发射率。对于N个光谱通道可以构成N个方程,这N个方程中也包含一个真温、N个亮度温度和N个光谱发射率,其中亮度温度是已知量,真温和光谱发射率是未知量。由于方程组是欠定的,理论上存在着大量的解。为了求解这个方程组常需要假设光谱发射率与波长和温度之间的数学模型,使方程组未知数的个数降为N个,实现真温的求解。当光谱发射率与波长或温度之间的规律被正确获得后,多光谱辐射测温法才能反演出正确的真温。通过对上述较为常用两种光谱发射率模型的分析可知,这两种方法的基本思想都是试图找到光谱发射率与波长或温度之间的函数关系,确立光谱发射率与波长或温度之间数学模型。用含有波长或温度的表达式代替光谱发射率,实现方程的求解。由于光谱发射率具有一定的不确定性,假设的光谱发射率模型与实际光谱发射率的变化之间存在一定的差异,有可能导致真温反演产生较大的误差。光谱发射率与波长或温度之间的数学模型是需要通过大量的实验和经验才能获得的,而且这种数学模型通用性较差,尤其是当待测辐射体发生改变时,这种数学模型也就失去了意义。为了解决多光谱高温计在实际测量中存在的问题,找到一种无需假定光谱发射率与波长或温度之间数学模型而且又具有一定通用性的多光谱真温反演方法成为一种迫切的需要。为此,将优化的思想引入到了多光谱求解过程中,将多光谱真温的求解问题转化为多目标普朗克极小值优化(MMP)问题,从而不再需要建立光谱发射率与波长或温度之间的数学模型,降低了系统的复杂性与难度。该方法以普朗克公式和光谱发射率之间的等式约束条件为基础,构造了六个目标函数,实现了真温的求解。新方法在反演精度上得到了较大幅度的提高,仿真数据的误差都小于1%。借助于以往的真实测量数据,利用多目标普朗克极小值优化法实现了真温的反演。