基于遗传小波神经网络的海底声学底质识别分类

分割海底声纳探测图像,提取单元特征向量进行主成份分析,选取均值、标准差、对比度、相关系数、能量及同质性作为训练特征向量,构建小波神经网络。利用遗传算法优化小波神经网络的初始权值及小波参数,对砂、礁石、泥3种底质类型分别进行训练,并得到3种底质的测试精度都在90%以上,优于单独利用小波神经网络进行训练时的测试精度,克服了小波神经网络训练时易陷入局部极小的固有缺陷,表明基于遗传算法的小波神经网络可有效用于海底底质声纳图像的识别和分类。     

中国北方沿海典型澙湖—山东月湖水深遥感反演

澙湖是中国北方重要的滨海生态系统之一,水深较浅、透明度较低、底质类型复杂、人类干扰频繁等特点导致澙湖水深资料长期缺乏或存在较大误差,极大地制约了澙湖生态系统的研究和保护。本研究在现场实测水深的基础上,使用IKONOS高分辨率卫星影像,将山东荣成月湖水域划分为浅水区、植被区和深水区三类分别进行水深反演。研究表明红光波段(Band3)对水深的敏感性最高,分区建立线性回归模型反演水深的精度高于不分区的精度,多波段组合模型的反演精度最高,植被对水深反演的精度影响较大。反演结果表明,月湖水深最深处达271.23cm,水深反演的平均相对误差为13.16%。     

Remote Sensing Observations of Ocean Surface Chlorophyll and Temperature with the Impact of Cyclones and Depressions over the Bay of Bengal Water

MODIS-Aqua derived eight-day composite chlorophyll concentration data analyzed to study the impact of cyclones and depressions on the Bay of Bengal productivity. A total of 15 cyclonic storms and depressions picked up from the India Meteorological Department datasets. MODIS-Aqua data analyzed during October 2002 to December 2009. There was observation of depressions (D), deep depressions (DD), cyclonic storms (CS), severe cyclonic storms (SCS), and very severe cyclonic storms (VSCS) with categories of intensities (“T” numbers 1.5–6) and wind speeds (25–108 knots). The chlorophyll concentration observed to be high (1.0–5.0 mg/m³) with the impact in the coastal and offshore waters. Quickscat scatterometer data showed high wind speed (about 10 meters/second). Sea surface temperature (SST) observed to be decreased (roughly 2°C) with effect of cyclones. The cyclone numbers, intensity, and chlorophyll concentration has been observed to be increasing from 2002 to 2009, with observation of VSCS “Sidr” during 2007. The study would be interesting to link carbon flux/sequestration, marine food chain, and harnessing fishery resources in a postcyclone period.     

Distribution and invasion of Spartina alterniflora within the Jiaozhou Bay monitored by remote sensing image

Spartina alterniflora as an alien invasive plant, poses a serious threat to the ecological functions of the coastal wetland of the Jiaozhou Bay. As of 2019, the distribution area of S. alterniflora in the Jiaozhou Bay has reached more than 500 hm². For this reason, combined with field surveys, remote sensing monitoring of the invasion S.alterniflora in the Jiaozhou Bay has been carried out. To accurately identify S. alterniflora within the Jiaozhou Bay coastal wetland, we used a new m...     

Airborne imagery of a disintegrating Sargassum drift line

Airborne hyperspectral and thermal infrared imagery collected over the Florida Current provide a view of the disintegration of a Sargassum drift line in 5 m s⁻¹ winds. The drift line consists mostly of rafts 20-80 m² in size, though aggregations larger than 1000 m² also occur. Rafts tend to be elongated, curved in the upwind direction, and 0.1-0.5 °C warmer than the surrounding ocean surface. Long weed ‘trails’ extending upwind from the rafts are evidence of plants dropping out and being left behind more rapidly drifting rafts. The raft line may be a remnant of an earlier Sargassum frontal band, which is detectible as an upwind thermal front and areas of submerged weed. Issues are identified that require future field measurements.     

基于灰度匹配的序列影像的目标跟踪研究与实现

利用匹配算法对视频帧进行处理可以实现对特定目标的跟踪.通过在Matlab平台上利用基于灰度的匹配算法实现简单背景下的单目标跟踪,来说明目标跟踪的基本过程,并且对在Matlab平台上实现目标跟踪所遇到的坐标系转换、运算速度以及适用范围等问题进行了分析讨论.     

基于遥感增强嵌套模型红树林带水沙环境对码头建设的响应研究

南海北部湾铁山港码头建设期间,邻近区域红树林带植株出现受损死亡现象。本研究构建基于遥感增强包含红树林潮间带的海湾水沙嵌套模型,评估码头建设对红树林带潮流、泥沙输移及冲淤的影响。结果表明：海湾水沙嵌套模型可充分利用南海海流模拟信息,且精细化模拟了工程建设附近海域泥沙输移规律。采用全球地表水覆盖几率遥感反演数据与当地潮位的信息融合技术,为红树林带水沙模拟提供了可靠地形信息。码头建设主要影响红树林带南部潮流,涨急时流速减小,落急时部分潮沟通道流速增大。当进港方案更改为过水钢栈桥,红树林带东南部流速略有增加。码头施工期,红树林带69%～72%范围的悬浮泥沙浓度增量介于20～50 mg/L之间。码头建设前,红树林带泥沙平均冲淤量为0.27 cm/a;进港方案分别为不过水通道和过水钢栈桥,码头建设后红树林带淤积量增加的面积占比分别为96.6%和89.3%,平均冲淤量分别为0.45 cm/a和0.36 cm/a。过水钢栈桥替换不过水通道,红树林带冲淤增量下降了50.0%。研究结论可为潮间带水沙环境模拟和红树林带保护决策提供科学依据。     

基于静态小波变换改进邻域系数法的遥感图像滤波

针对NeighShrink滤波算法存在的不足,提出了改进的阈值估计策略和系数收缩方案,并将改进算法与静态小波变换相结合进行图像滤波,取到了较好的滤波效果.实验验证了改进算法的有效性,以峰值信噪比PSNR作为客观评价指标,通过与经典方法的比较,对该算法的滤波性能做出了客观评价.     

一种基于自适应窗口滤波的DEM获取算法

在分析了一种基于固定窗口滤波的DEM获取算法的基础上,提出了一种自适应窗口滤波的DEM获取算法,改变移动窗口的大小进行分块拟合来反演地面的高程值,获取DEM数据.利用海岸带区域IKONOS卫星影像匹配生成的三维坐标数据集进行了实验,实验证明了该算法的有效性.     

基于高斯混合模型的海冰图像非监督聚类分割研究

为了利用海冰图像识别技术获取海冰冰况信息, 探索了利用高斯混合模型进行海冰图像分割的技术途径, 描述了具体算法, 并利用高斯混合模型的最大期望值（EM）算法以及最小描述长度（MDL）准则对渤海海冰图像进行目标提取。研究结果表明, 该方法可以很好地实现海冰信息的有效提取和海冰图像的有效分割, 从而证明了建立在图像分割技术...