山区可见光—近红外遥感影像浓密植被暗像元自动识别方法研究

暗目标法是目前气溶胶光学厚度遥感反演中应用最为广泛的方法,浓密植被暗像元的识别是暗目标法的基础。针对可见光—近红外影像缺少中红外波段难以有效识别浓密植被暗像元的问题,引入红波段直方图阈值法识别山区可见光—近红外影像的浓密植被暗像元。该方法利用浓密森林像元在可见光波段反射率低的特点,通过搜索红波段直方图的最小峰值自动识别浓密植被暗像元。试验中选取Landsat TM影像前4个波段利用红波段直方图阈值法识别可见光—近红外影像的浓密植被暗像元,并与在中红外波段影像和可见光—近红外影像中广泛应用的两种暗像元识别方法进行对比分析,探讨红波段直方图阈值法的有效性,最后将该方法应用于环境减灾卫星(HJ-1)CCD影像的暗像元识别和气溶胶反演。实验结果表明:红波段直方图阈值法明显优于常用的可见光—近红外影像暗像元识别方法,识别精度接近传统的中红外波段影像识别方法,相似度指数小于2和小于3的暗像元分别为83.12%和93.48%。该方法为山区可见光—近红外影像浓密植被暗像元自动识别提供了一种新的适用方法,识别结果能够满足暗目标法反演气溶胶光学厚度的要求。     

秸秆焚烧火点遥感监测算法实现

为了获取华中区域的秸秆焚烧火点空间分布信息,实现对该区域秸秆焚烧的有效管控,以2014年MODIS L1B遥感数据为主要数据源,结合土地利用类型数据,以华中的农田为研究区域,基于增强型上下文火点遥感影像识别方法,充分利用定量遥感的理论知识及地理空间数据抽象库(GDAL)等技术手段,实现了华中区域秸秆焚烧火点的识别。利用中华人民共和国环境保护部发布的全国秸秆焚烧火点日报和MODIS标准火点产品(MYD14)进行空间和定量上的对比分析。研究结果表明,该算法能够有效地进行研究区域的秸秆焚烧火点遥感监测,并且可以依据研究区域的特点进行参数的实时调整,提高了秸秆焚烧火点提取的自动化和工作效率。     

基于混合像元分解的高光谱影像柑橘识别方法

为及时准确地监测柑橘种植信息,以江西省会昌县作为研究区,采用EO-1 Hyperion高光谱影像作为数据源,构建了基于混合像元分解的高光谱影像柑橘识别方法。首先,针对EO-1 Hyperion高光谱影像提供了242个波段,光谱范围广的特点,在波段选择、大气校正等预处理的基础上,提取研究区典型地物端元光谱曲线;然后,利用全约束线性光谱混合模型进行混合像元分解,提取出柑橘端元的丰度值,并通过对照高分遥感影像,构建柑橘端元丰度与柑橘实际种植的对应的关系。结果表明:由于典型地物端元提取中不可避免的误差及柑橘冠层覆盖度的差异,柑橘种植的准确识别与其柑橘端元丰度阈值存在对应关系。在经过反复试验的条件下,研究区柑橘端元丰度阈值设定在0.30~0.45范围之内,总精度达到90%以上,能够满足柑橘种植识别要求。     

基于PCA与EM算法的多光谱遥感影像变化检测研究

变化检测技术能够从观测数据中有效地提取各种变化信息。以多时相的多光谱遥感影像作为主要研究对象,就基于直接比较像元特征的变化检测中两方面的关键问题——差异影像的构造和变化阈值的提取来开展研究,具体采用PCA结合差值比较的方法构造差异影像,采用EM自动确定变化阈值以提取变化区域。通过与传统变化检测方法的实验比较,证明了该方法的有效性。     

面向城市区域的遥感影像云检测方法

针对高分辨率遥感影像云检测过程中,城市区域中大型建筑物对云检测精度有较大影响的问题,提出了一种面向城市区域的高分辨率遥感影像云检测算法。算法利用高分辨率卫星各季节的晴空遥感影像数据,构建晴空背景场,采用遥感数据多光谱各波段的动态阈值定位出云像元,然后基于云与建筑物在纹理和形态特征上的不同,去除矩形、多边形等高亮建筑物的错误检测区域。结果表明,该方法可有效去除城市区域的高亮地物,云检测正确率由83%提高到90%以上,证明了该算法的有效性。     

面向对象变化向量分析的遥感影像变化检测

为了解决基于像元的变化向量分析法在高分辨率遥感影像变化检测中精度低的问题,提出了一种面向对象变化向量分析的遥感影像变化检测方法。综合2个时期的遥感影像,首先通过影像分割获取像斑,其次提取直方图作为像斑的特征向量,再次采用直方图相交法度量2个时期像斑直方图之间的距离,构建像斑的变化向量,然后利用加权组合的方法计算像斑变化向量的模,最后依据最大熵原理获取变化检测阈值,对像斑进行变化/未变化判别。在QuickBird及Ikonos遥感影像上的实验表明:在高分辨率遥感影像变化检测中,与基于像元的变化向量分析法相比,该方法变化检测的精度较优,变化检测的正确率分别达到了0.92与0.90。     

基于MODIS数据的华北地区秸秆焚烧监测

秸秆焚烧给我国城乡生态环境造成巨大损害,利用遥感手段监测秸秆焚烧能够为禁烧治理工作提供有效的数据支持。“背景对比火点探测算法”(the Contextual Fire Detection Algorithm)是目前精度较高的自动探测算法,但固定的阈值参数难以适用于不同地区和不同的监测对象,因此依据实际观测情况对其中的关键参数和阈值进行了适当调整,以更好地监测中国地区的火点。基于EOS/Terra卫星的MODIS数据,利用调整阈值后的算法对我国华北地区2007年5月至8月的秸秆焚烧状况进行了遥感监测,监测精度能够满足实际业务化监测的需要。进一步结合IGBP地表分类数据,将火点像元分成秸秆焚烧、林火和草原火等3种生物焚烧类型,并分别对其亮度温度等多个参数进行了统计分析,在此基础上讨论了根据火点辐射特性判断火点类别的可行性,提出在目前,地表分类数据对于判断火点类别仍是必要的。     

利用指数熵的多时相遥感影像变化检测方法

针对像素级变化检测法中变化阈值的提取不够自动化和准确化,导致变化检测结果精度不高的问题,提出一种利用双阈值指数熵的多时相遥感影像变化检测方法。该方法首先采用差值法构造2个时相遥感影像的差异影像;其次采用双阈值指数熵的方法确定差异影像的最佳变化阈值,并将其用于分割差异影像,得到变化区域。采用客观评价法对变化检测结果进行精度评定。选择我国鄱阳湖局部区域2个时相的遥感影像进行试验,并与基于模糊C均值的变化检测方法进行对比。通过试验,所提出方法变化检测精度达94.22%,是一种有效、可行的变化检测方法。     

针对ETM基于BP网络像元分解模型分类研究

限制遥感图像分类精度的一个主要原因是混合像元问题。因此像元分解也一直是遥感研究的一个热点问题。本针对传统像元分解方法的缺点，首先对影像进行纯净像元提取，再对混合像元进行分解。在提取纯净像元时，利用ETM影像的全色波段和单波段不同的分辨率选取训练样本，从而克服了传统像元分解中需要两种影像或地面实测资料的缺点。然后用BP神经网络对混合像元进行分解。同时用民乐县的ETM影像作了试验。又利用对应的土地利用图作了验证。取得了比较好的效果。     

基于灰色理论的遥感图像最佳镶嵌线检测

提出了一种基于灰关联分析的多景遥感影像最佳镶嵌线的检测算法。该算法根据中心像元与周围像元之间的临近像元效应,在两景影像重叠区域选择中心像元邻域作为参考序列与比较序列,利用斜率关联度计算两者之间的灰关联度,据此找到一条影像色调和纹理差异较小的镶嵌线。以四组两景相邻影像自动寻找镶嵌线为例进行了实验研究,取得了良好效果,验证了该方法的可行性与有效性。     

基于亮温—植被指数—气溶胶光学厚度的MODIS火点监测算法研究

MODIS火灾产品的火点检测算法主要以4和11μm通道亮温数据来识别火点,在应用于不同地区和不同季节时有一定局限性。在分析MODIS现有火点检测算法的基础上,对算法相关阈值及参数进行适当调整,同时考虑火灾前后NDVI的变化,以及林火燃烧过程中伴生烟羽使火点下风方气溶胶光学厚度明显增加的特点,构建了基于亮温—植被指数—气溶胶光学厚度的火点识别算法,并应用多次火灾个例对本算法进行验证。结果表明:算法提高了对高温热点和低温焖烧火点的识别能力,有效降低了高温热点的误报率和低温火点的漏报率,使火点检测算法在不同环境的适应性有所增强。     

森林过火面积的遥感测算方法

根据对近年来多次特大森林火灾和相应的气象卫星资料的分析,提出利用NOAA/AVHRR数据测算森林大火的过火面积的四种方法,即灰度修正像元法、植被修正像元法、坐标法和蔓延法。在GIS地面信息数据库支持下,利用这4种方法能准确、快速地计算出过火面积。经今春应急评估试运行验证,森林大火过火面积测算精度达90%。     

A contextual algorithm for AVHRR fire detection

A contextual algorithm for fire detection with NOAA-AVHRR-LAC data was developed. Unlike ‘traditional’ fire detection algorithms (e.g., multichannel thresholds), the decision to record a fire is made by comparing a fire pixel with the pixels in its immediate neighbourhood. The algorithm is self-adaptive and therefore very consistent over large areas as well as through seasons. The algorithm appears to operate successfully in most areas of the world. This Letter presents the contextual approach and describes the algorithm.     

基于显著性目标检测的弹着点定位算法

针对目前炮弹定位方法安全隐患大、人工测量效率低、精度差的问题,本文提出一种基于显著性目标检测网络BASNet(Boundary-Aware Salient Object Detection)的弹着点定位方法。采用改进的BASNet网络,结合注意力机制模块CBAM(Convolutional Block Attention Module)、金字塔池化模块PPM(Pyramid Pooling Module)与深度可分离卷积,对炮弹火焰进行显著性检测,提取弹着点图像坐标。实验结果表明,该方法在自制的炮弹火焰数据集上的检测精度F值达到0.914,MAE为0.006,推理速度为3.86 fps,优于BASNet、U2Net等显著性目标检测网络。该方法提取的弹着点图像坐标与真实坐标误差为5.92个像素值,相比于BASNet网络减少近4.85个像素值。综合可知,该算法增强了网络对显著性物体内部的检测精度,提高了模型推理效率,减少了图像弹着点坐标误差,适用于靶场小范围炮弹火焰烟雾的检测,能够满足靶场应用的实测需求。     

An effective graph embedded YOLOv5 model for forest fire detection

The existing YOLOv5-based framework has achieved great success in the field of target detection. However, in forest fire detection tasks, there are few high-quality forest fire images available, and the performance of the YOLO model has suffered a serious decline in detecting small-scale forest fires. Making full use of context information can effectively improve the performance of small target detection. To this end, this paper proposes a new graph-embedded YOLOv5 forest fire detection framework, which can improve the performance of small-scale forest fire detection using different scales of context information. To mine local context information, we design a spatial graph convolution operation based on the message passing neural network (MPNN) mechanism. To utilize global context information, we introduce a multi-head self-attention (MSA) module before each YOLO head. The experimental results on FLAME and our self-built fire dataset show that our proposed model improves the accuracy of small-scale forest fire detection. The proposed model achieves high performance in real-time performance by fully utilizing the advantages of the YOLOv5 framework.     

Quantifying burn severity in a heterogeneous landscape with a relative version of the delta Normalized Burn Ratio (dNBR)

Multi-temporal change detection is commonly used in the detection of changes to ecosystems. Differencing single band indices derived from multispectral pre- and post-fire images is one of the most frequently used change detection algorithms. In this paper we examine a commonly used index used in mapping fire effects due to wildland fire. Subtracting a post-fire from a pre-fire image derived index produces a measure of absolute change which then can be used to estimate total carbon release, biomass loss, smoke production, etc. Measuring absolute change however, may be inappropriate when assessing ecological impacts. In a pixel with a sparse tree canopy for example, differencing a vegetation index will measure a small change due stand-replacing fire. Similarly, differencing will produce a large change value in a pixel experiencing stand-replacing fire that had a dense pre-fire tree canopy. If all stand-replacing fire is defined as severe fire, then thresholding an absolute change image derived through image differencing to produce a categorical classification of burn severity can result in misclassification of low vegetated pixels. Misclassification of low vegetated pixels also happens when classifying severity in different vegetation types within the same fire perimeter with one set of thresholds. Comparisons of classifications derived from thresholds of dNBR and relative dNBR data for individual fires may result in similar classification accuracies. However, classifications of relative dNBR data can produce higher accuracies on average for the high burn severity category than dNBR classifications derived from a universal set of thresholds applied across multiple fires. This is important when mapping historic fires where precise field based severity data may not be available to aid in classification. Implementation of a relative index will also allow a more direct comparison of severity between fires across space and time which is important for landscape level analysis. In this paper we present a relative version of dNBR based upon field data from 14 fires in the Sierra Nevada mountain range of California, USA. The methods presented may have application to other types of disturbance events.     

融合注意力机制的无锚点森林火灾检测算法

森林火灾、野火是一个重大的自然灾害问题,每年全球各地植被都会受到严重的破坏。为了提高森林火灾的防控精度,针对传统方法具有火灾背景复杂、准确率低、效率低等问题,本文提出一种基于CenterNet的森林火灾检测算法。CenterNet作为一种无锚的方法,将目标定义为一个点,通过关键点估计定位目标的中心点,可以有效避免小目标的漏检。同时基于高效深层特征提取网络ResNet50,融合ECA模块以抑制无用信息,增加模型的特征提取能力。在公开森林火灾数据集上进行实验表明,与其他算法相比,本文提出的森林火灾检测算法误检率低,识别精度达到92.39%,F1值为0.86,Recall值为79.75%,FPS为43.31。本文提出的方法检测精度高,可满足实时检测森林火灾和实施精准施救的要求。     

An improved algorithm for the detection of plumes caused by natural or technological hazards using AVHRR imagery

Fires caused by natural or technological disasters emit large amounts of smoke which, once formed into plumes, may affect the human health and the environment. Satellite remote sensing data provide an effective tool to achieve detection and monitoring of these plumes over large areas on a routine basis. Discrimination of plumes on satellite images is a prerequisite to study and retrieve physical, chemical and optical properties of emitted smoke. An improved algorithm for the detection of plumes caused by natural or technological hazards using AVHRR imagery is presented in this study. The method is based on a multi-temporal and multi-spectral change detection algorithm. It is performed in two main steps: a) appropriate spectral and spatial filters are applied on the images acquired before and after a fire event in visible and near-infrared ranges in order to extract the core of the plume; b) a criterion on spectral information is defined as an homogeneity measure that enables, through a modified version of the region-growing method, the spatial expansion of the detected core to include the complete area covered by the plume. Through this approach, a pixel is identified as a plume pixel if it is “close” to the core plume pixels in both spatial and spectral spaces. The algorithm was developed and calibrated using AVHRR images acquired over Spain before and during a major forest fire event on July 16, 2005. It was applied using past events of natural and technological hazards in several locations to ensure its global applicability and robustness. The algorithm produced accurate results in all cases of plumes, either in natural or in technological fire events. Three application cases are presented in this study: A major fire in an industrial installation in London (December 11, 2005), a major fire in Baghdad during the recent war in Iraq (April 1, 2003) and a forest fire in California (September 29, 2005).     

Fire detection based on hidden Markov models

In this paper, a novel method of real-time fire detection based on HMMs is presented. First, we present an analysis of fire characteristics that provides evidence supporting the use of HMMs to detect fire; second, we propose an algorithm for detecting candidate fire pixels that entails the detection of moving pixels, fire-color inspection, and pixels clustering. The main contribution of this paper is the establishment and application of a hidden Markov fire model by combining the state transition between fire and non-fire with fire motion information to reduce data redundancy. The final decision is based on this model which includes training and application; the training provides parameters for the HMM application. The experimental results show that the method provides both a high detection rate and a low false alarm rate. Furthermore, real-time detection has been effectively realized via the learned parameters of the HMM, since the most time-consuming components such as HMM training are performed off-line.     

基于视频的嵌入式森林火灾预警系统

针对目前我国森林火情远程监控中长时间不间断监测、建设成本高的不足,提出了一种基于视频的嵌入式森林火灾监测和报警的系统。嵌入式处理器对安插在森林监测点摄像头采集到的视频图像,通过昼间或夜间火情识别算法进行智能处理,当潜在异常确认为异常发生时,通过无线网络将火情信息传送到监控中心,交由工作人员做进一步处理,避免了长时间不间断监控。与现有其他同类系统的比较表明：该系统对于我国森林火情监测有较大的实用价值。