空-谱融合的条件随机场高光谱影像分类方法

高光谱遥感数据具有光谱信息丰富、图谱合一的特点,目前已经广泛地应用在对地观测中。传统的高光谱分类模型大多过分依赖影像光谱信息,没有充分利用空间特征信息,这使得分类精度还有很大的提升空间。条件随机场是一种概率模型,能够较好地融合空间上下文信息,在高光谱影像分类中已经得到越来越多的关注,但大部分条件随机场模型存在超平滑的现象,会导致影像细节丢失。针对该问题,本文提出了一种优化融合影像空-谱信息的高分辨率/高光谱影像分类方法,该方法将影像的纹理信息与原始光谱信息进行融合,利用SVM分类器对其进行预分类,并将各类概率定义为一元势函数,以融合空间特征信息;然后将空间平滑项和局部类别标签成本项加入二元势函数中,以考虑空间背景信息,并保留各类别中的详细信息。最后,通过两组的高分辨率/高光谱影像数据进行试验。结果表明,与SVM算法、传统的条件随机场方法和面向对象的分类方法相比,本文提出的算法在整体分类精度上分别提高了10%、9%和8%以上,同时在保持地物边缘完整性、避免“同谱异物”与“同物异谱”的现象方面有较明显的优势。     

基于深度学习理论的无人机影像分类

针对无人机影像深度学习分类方法缺乏现状,本文利用深度学习理论卷积神经网络方法对无人机影像进行了分类。该法首先抽取无人机影像作为训练集和检验集,然后建立一个2个卷积层-池化层的卷积神经网络模型进行深度学习,通过设定参数并运行模型实现无人机影像分类。实验表明,本文提出的方法可完成较复杂地区无人机影像分类,其分类精度与支持向量机方法相当,为无人机遥感影像分类提供了一个崭新的技术视点。     

Integrating spectral and non-spectral data to improve urban settlement mapping in a large Latin-American city

ABSTRACT

Information on urban settlements is crucial for sustainability planning and management. While remote sensing has been used to derive such information, its applicability can be compromised due to the complexity in the urban environment. In this study, we developed a remote sensing method to map land cover types in a large Latin-American city, which is well known for its mushrooming unplanned and informal settlements. After carefully considering the landscape complexity there, we designed a data fusion method combining multispectral imagery and non-spectral data for urban and land mapping. Specifically, we acquired a cloud-free Landsat-8 image and two non-spectral datasets, i.e., digital elevation models and road networks. Then, we implemented a set of experiments with different inputs to evaluate their merits in thematic mapping through a supervised protocol. We found that the map generated with the multispectral data alone had an overall accuracy of 73.3% but combining multispectral imagery and non-spectral data yielded a land cover map with 90.7% overall accuracy. Interestingly, the thermal infrared information helped substantially improve both the overall and categorical accuracies, particularly for the two urban classes. The two types of non-spectral data were critical in resolving several spectrally confused categories, thus considerably increasing the mapping accuracy. However, the panchromatic band with higher spatial resolution and its derived textural measurement only generated a marginal accuracy improvement. The novelties of our work are with the successful separation between the two major types of urban settlements in a complex environment using a carefully designed data fusion approach and the insight into the relative merits of the thermal infrared information and non-spectral data in helping resolve the issue of class ambiguity. These findings should be valuable in deriving accurate urban settlement information which can further advance the research on socio-ecological dynamics and urban sustainability.     

中国热带6种海参骨片的种类和形态研究

利用扫描电子显微镜对我国6种热带海参(蛇目白尼参、玉足海参、花刺参、梅花参、子安辐肛参和巨梅花参)背脊部骨片的组成和结构进行了观察。结果表明,蛇目白尼参骨片有3种类型的花纹样体和1种杆状体;玉足海参骨片有2种类型的桌形体和1种扣形体;花刺参骨片有2种类型的桌形体、1种C形体和1种花纹样体;梅花参骨片有4种类型的杆状体;子安辐肛参骨片有1种花纹样体;巨梅花参骨片有1种杆状体、1种拟桌形体和1种颗粒体。结果显示,与之前学者光镜下观察的结果存在差异,玉足海参桌形体底盘中央是单孔而非4孔;花刺参桌形体立柱之间没有横梁连接。并发现了新的骨片类型:蛇目白尼参Ⅲ型花纹样体和杆状体;玉足海参顶部周缘小孔桌形体;花刺参顶部5簇小齿桌形体;梅花参Ⅱ型、Ⅲ型、Ⅳ型杆状体。     

我国海岸潟湖卫星遥感分类方法研究

参照Landsat-5TM和HJ-1CCD遥感影像,依据实地踏勘资料及解译经验,分析各种海岸潟湖地貌学成因和发育演变阶段,提出基于潟湖地貌学成因的遥感分类方法。并将我国海岸潟湖划分为滨外坝型、沙坝型、沙嘴型、湾顶坝型、连岛坝型和河口型共6种地貌类型。建立基于发育演变阶段的潟湖遥感分类标准,将海岸潟湖分为青壮年期、老年期和死亡期共3个发育演变过程;结合潟湖在影像中的特征,以近红外、红、绿波段组合后的假彩色影像为例,从色彩、纹理、地物邻接关系等方面建立不同潟湖类型的遥感解译标志;根据所建立的遥感解译标志,对山东省53个海岸潟湖进行遥感分类验证,结果吻合。     

一种基于分割图斑的海岸带遥感图像分类方法

传统的基于象元的分类方法在处理海岸带这类动态多变、混杂度大的地物时常出现“椒盐现象”,难以矢量化形成专题图。提出了一种新的遥感图像分类方法——基于分割图斑的分类法,即选择一个合适的尺度,利用光谱信息按照一定的策略将图像分割为一系列的图斑,并确保图斑内大多数象元的光谱特征相近,分别对图斑内诸象元进行统计,求出各个波段的均值,用该均值替换图斑内所有象元的原始亮度值,这样分类时可以确保同一图斑内的所有象元被分到同一个类别中,从而有效地避免了“椒盐现象”,使分类结果便于矢量化成图,也便于对特定地物的提取。实验结果表明,该方法还能在一定程度上提高分类精度。     

盐田水体遥感分类方法研究

以连云港台北盐场为研究区，介绍了监督分类法和神经网络分类法及其在盐田水体遥感分类中的具体应用。研究结果表明，用神经网络分类法进行遥感影像自动分类，其分类精度高，显示了其在遥感领域较为广阔的应用前景。     

保持局域特征的多源海冰图像融合与分类

本文提出一种保持局域特征的多源海冰图像融合方法,并在此基础上进行海冰分类。本文提出的多源海冰图像融合方法包括保持空间局域融合和保持特征局域融合两方面。首先,通过学习得到投影矩阵和相似矩阵。投影矩阵将多源像素进行投影变换,得到保留像素空间局域特性的融合向量。相似矩阵度量像素特征间的相似性,通过拉普拉斯特征分解,得到保留像素特征局域相似性的融合向量。然后,将空间融合向量和特征融合向量进行像素综合,得到融合图像。在此基础上,本文设计一种滑动集成分类方法进行融合图像像素分类。提出的分类方法利用滑动集成的特点,在分类时增强刻画了海冰局域特性。由于本文的保持局域融合框架不仅刻画了海冰在物理空间中的邻接关系,而且考虑不同海冰类型的特征关系,因此其在多源图像（多光谱和合成孔径雷达（SAR）图像）的海冰分类任务中表现优异。实验结果表明本文提出的基于保持局域特征融合的多源海冰图像分类方法有效提升了海冰分类精度。     

用于海底目标识别与底质分类的多波束水体波形预处理

多波束水体数据是多波束系统获取的最原始数据,记录了波束从发射到接收整个过程全部的反向散射强度信息,可以为目标识别、水下栖息环境探测等提供重要的数据支撑.目前,针对多波束水体强度时间序列所表现的波形信息的处理及研究仍处于起步阶段,另外水体波形数据易受噪声影响,且存在明显的入射角效应问题,对此,本文提出了一种基于分区异构的...     

A New Retracking Technique for Brown Peaky Altimetric Waveforms

A new Brown-Peaky (BP) retracker has been developed for peaky waveforms that usually appear within ～10 km to the coastline. The main feature of the BP is that it fits peaky waveforms using the Brown model without introducing a peak function. The retracking strategy first detects the peak location and width of a waveform using an adaptive peak detection method, and then estimates retracking parameters using a weighted least squares (WLS) estimator. The WLS assigns a downsized weight to corrupted waveform gates, but an equal weight to other normal waveform gates. The BP retracker has been applied to 4-year Jason-1 waveform (2002–2006) in two Australian coastal zones. The results retracked by BP, MLE4 and ALES retrackers have been validated against tide-gauge observations located at Burnie, Lorne and Broome. The comparison results show that three retrackers have similar performance over open oceans with the correlation coefficient (～0.7) and RMSE (～13 cm) between altimetric and tide-gauge sea levels for distance >7 km offshore. The main improvement of BP retracker occurs for distance ≤7 km to the coastline, where validation results indicate that data retracked by BP are more accurate (15–21 cm) than those by ALES (16–24 cm) and MLE4 (19–37 cm).