Estimating forage quantity and quality using aerial hyperspectral imagery for northern mixed-grass prairie

Sustainable rangeland stewardship calls for synoptic estimates of rangeland biomass quantity (kg dry matter ha^− 1) and quality [carbon:nitrogen (C:N) ratio]. These data are needed to support estimates of rangeland crude protein in forage, either by percent (CP_c) or by mass (CP_m). Biomass derived from remote sensing data is often compromised by the presence of both photosynthetically active (PV) and non-photosynthetically active (NPV) vegetation. Here, we explicitly quantify PV and NPV biomass using HyMap hyperspectral imagery. Biomass quality, defined as plant C:N ratio, was also estimated using a previously published algorithm. These independent algorithms for forage quantity and quality (both PV and NPV) were evaluated in two northern mixed-grass prairie ecoregions, one in the Northwestern Glaciated Plains (NGGP) and one in the Northwestern Great Plains (NGP). Total biomass (kg ha^− 1) and C:N ratios were mapped with 18% and 8% relative error, respectively. Outputs from both models were combined to quantify crude protein (kg ha^− 1) on a pasture scale. Results suggest synoptic maps of rangeland vegetation mass (both PV and NPV) and quality may be derived from hyperspectral aerial imagery with greater than 80% accuracy.     

Rule-based interpretation of aerial imagery

In this paper, we describe the organization of a rule-based system, SPAM, that uses map and domain-specific knowledge to interpret airport scenes. This research investigates the use of a rule-based system for the control of image processing and interpretation of results with respect to a world model, as well as the representation of the world model within an image/map database. We present results on the interpretation of a high-resolution airport scene wvhere the image segmentation has been performed by a human, and by a region-based image segmentation program. The results of the system's analysis is characterized by the labeling of individual regions in the image and the collection of these regions into consistent interpretations of the major components of an airport model. These interpretations are ranked on the basis of their overall spatial and structural consistency. Some evaluations based on the results from three evolutionary versions of SPAM are presented.     

Performance evaluation and analysis of monocular buildingextraction from aerial imagery

Research in monocular building extraction from aerial imagery has neglected performance evaluation in three areas: unbiased metrics for quantifying detection and delineation performance, an evaluation methodology for applying these metrics to a representative body of test imagery, and an approach for understanding the impact of image and scene content on building extraction algorithms. This paper addresses these areas with an end-to-end performance evaluation of four existing monocular building extraction systems, using image space and object space-based metrics on 83 test images of 18 sites. This analysis is supplemented by an examination of the effects of image obliquity and object complexity on system performance, as well as a case study on the effects of edge fragmentation. This widely applicable performance evaluation approach highlights the consequences of various traditional assumptions about camera geometry, image content and scene structure, and demonstrates the utility of rigorous photogrammetric object space modeling and primitive-based representations for building extraction     

由航空影像自动重建大范围3维地形

目的 针对由航空影像自动生成大范围3维地形的立体模型配准问题,提出一种自动配准全部立体模型的方法,从而生成大范围3维地形。方法 首先由相邻影像构建独立的立体模型;然后根据特征匹配同名点在公共影像上的坐标对应关系,自动提取相邻模型的连接点;通过循环遍历搜索,自动配准全部立体模型,进而构建全航摄区的大范围3维地形。结果 采用两组数据进行实验,结果显示,两组数据全部3维模型的均方配准误差分别为5.20像素和2.63像素。本文方法生成的大范围地形的相对精度较高;对第2组数据的结果采用控制点进行绝对定向,并用检查点进行精度评估,结果显示全部检查点的均方平面和高程误差分别为0.326 m和0.502 m,生成的大范围地形达到了较高的绝对精度。结论 本文方法可自动化执行,仅需输入一系列有一定重叠的航空影像,即可自动生成按一定方式组织的大范围3维地形产品。该方法生成的大范围地形既可用于3维场景浏览,也可用于地形量测,但不适用于由激光扫描获取的点云数据的配准。     

Partially supervised hierarchical classification for urban features from lidar data with aerial imagery

Although spatial and spectral resolutions of remotely sensed data have been improved, the usage of multispectral imagery is not sufficient for urban feature classification. This article addresses the problem of automated classification by integrating airborne lidar range data and aerial imagery. In this study, the classification procedure is divided into three phases. We first use the lidar range data to obtain the coarse lidar-based classification results, by which a lidar-driven labelled image and a lidar-driven high-rise object mask are acquired in this phase. Then, at the image-based classification level, we train samples based on the lidar-driven labelled image and conduct maximum likelihood classification experience with the lidar-driven normalized digital surface model as a high-rise object mask. Finally, we propose a knowledge-based cross-validation (KBCV) for misclassification between the lidar-based classification results and the image-based classification results. Experimental results are presented to demonstrate the benefits of the training sample selection of the lidar-driven labelled image, using the lidar-driven high-rise object mask, and the greater classification accuracy of the KBCV.     

Comparison of hyperspectral imagery with aerial photography and multispectral imagery for mapping broom snakeweed

Broom snakeweed (Gutierrezia sarothrae (Pursh) Britt. & Rusby) is one of the most widespread and abundant rangeland weeds in western North America. The objectives of this study were to evaluate airborne hyperspectral imagery and compare it with aerial colour-infrared (CIR) photography and multispectral digital imagery for mapping broom snakeweed infestations. Airborne hyperspectral imagery along with aerial CIR photographs and digital CIR images was acquired from a rangeland area in south Texas. The hyperspectral imagery was transformed using minimum noise fraction (MNF) and then classified using minimum distance, Mahalanobis distance, maximum likelihood, and spectral angle mapper (SAM) classifiers. The digitized aerial photographs and the digital images were respectively mosaicked as one photographic image and one digital image; these were then classified using the same classifiers. Accuracy assessment showed that the maximum likelihood classifier performed the best for the three types of images. The best overall accuracies for three-class classification maps (snakeweed, mixed woody and mixed herbaceous) were 91.0%, 92.5%, and 95.0%, respectively, for the CIR photographic image, the digital CIR image and the MNF-transformed hyperspectral image. Kappa analysis showed that there were no significant differences in maximum likelihood-based classifications among the three types of images. These results indicate that airborne hyperspectral imagery along with aerial photography and multispectral imagery can be used for monitoring and mapping broom snakeweed infestations on rangelands.     

Estimating census district populations from satellite imagery: Some approaches and limitations

Small-area population densities and counts were estimated for Australian census collection districts (CDs), using Landsat TM imagery. A number of mathematical and statistical refinements to previously reported methods were explored. The robustness of these techniques as a practical methodology for population estimation was investigated and evaluated using a primary image for model development and training, and a second image for validation. Correlations of up to 0.92 in the training set and up to 0.86 in the validation set were obtained between census and remote sensing estimates of CD population density, with median proportional errors of 17.4% and 18.4%, respectively. Total urban populations were estimated with errors of +1% and-3%, respectively. These results indicate a moderate level of accuracy and a substantial degree of robustness. Accuracy was greatest in suburban areas of intermediate population density. There was a general tendency towards attenuation in all models tested, with high densities being under-estimated and low densities being over-estimated. It is concluded that the level of accuracy obtainable with this methodology is limited by heterogeneity within the individual CDs, particularly large rural CDs, and that further improvements are in principle unlikely using the aggregated approach. An alternative statistical approach is foreshadowed.     

Estimating forest stand structure from high resolution imagery using the directional variogram

The height and stocking of forest stands can be estimated with relatively high precision using an empirical model relating parameters extracted from the directional variogram of high resolution images and forest structure parameters. A geometrical-optical model of the forest was first used to generate images of artificial forest stands in order to establish the relation between tree size. tree density and image texture. The resulting equations were then applied on the computer generated images as well as on high resolution MEIS II images to predict the forest structure values. The results show a good concordance between actual and predicted values, even when spatial resolution was degraded from 0·36m to 2·16m.     

Estimating fragmentation effects on simulated forest net primary productivity derived from satellite imagery

Conversion of native forests to agriculture and urban land leads to fragmentation of forested landscapes with significant consequences for habitat conservation and forest productivity. When quantifying land-cover patterns from airborne or spaceborne sensors, the interconnectedness of fragmented landscapes may vary depending on the spatial resolution of the sensor and the extent at which the landscape is being observed. This scale dependence can significantly affect calculation of remote sensing vegetation indices, such as the Normalized Difference Vegetation Index (NDVI) and its subsequent use to predict biophysical parameters such as the fraction of photosynthetically active radiation intercepted by forest canopies (fPAR). This means that simulated above-ground net primary productivity (NPP_A) using canopy radiation interception models such as 3-PG (Physiological Principles for Predicting Growth), coupled with remote sensing observations, can yield different results in fragmented landscapes depending on the spatial resolution of the remotely sensed data.We compared the amount of forest fragmentation in 1?km SPOT-4 VEGETATION pixels using a simultaneously acquired 20?m SPOT-4 multispectral (XS) image. We then predicted NPP_A for New Zealand native forest ecosystems using the 3-PG model with satellite-derived estimates of the fPAR obtained from the SPOT-4 VEGETATION sensor, using NDVI values with and without correction for fragmentation. We examined three methods to correct for sub-pixel fragmentation effects on NPP_A. These included: (1) a simple conversion between the broad 1?km scale NDVI values and the XS NDVI values; (2) utilization of contextural information from XS NDVI pixels to derive a single coefficient to adjust the 1?km NDVI values; and (3) calculation of the degree of fragmentation within each VEGETATION 1?km pixel and reduce NDVI by an empirically derived amount based on the proportional areal coverage of forest in each pixel. Our results indicate that predicted NPP_A derived from uncorrected 1?km VEGETATION pixels was significantly higher than estimates using adjusted NDVI values; all three methods reduced the predicted NPP_A. In areas of the landscape with a large degree of forest fragmentation (such as forest boundaries) predictions of NPP_A indicate that the fragmentation effect has implications for spatially extensive estimates of carbon uptake by forests.     

A scale robust convolutional neural network for automatic building extraction from aerial and satellite imagery

Identifying buildings from remote sensing imagery has been a challenge due to uncertainties from remote sensing imagery and variations in building structure and texture. In this study, we develop a scale robust CNN structure to improve the segmentation accuracy of building data from high-resolution aerial and satellite images. Based on a fully convolutional network, we introduce two Atrous convolutions on the first two lowest-scale layers, respectively, in the decoding step, aiming at enlarging the sight-of-view and integrate semantic information of large buildings. Then, a multi-scale aggregation strategy is applied. The last feature maps of each scale are used to predict the corresponding building labels, and further up-sampled to the original scale and concatenated for the final prediction. In addition, we introduce a combined data augmentation and relative radiometric calibration method for multi-source building extraction. The method enlarges sample spaces and hence the generalization ability of the deep learning models. We validate our developed methods with an aerial dataset of more than 180, 000 buildings with various architectural types, and a satellite image dataset consists of more than 29,000 buildings. The results are compared with several most recent studies. The comparison result shows our neural network outperformed other studies, especially in segmenting scenes of large buildings. The test on transfer learning from aerial dataset to satellite dataset showed our augmentation strategy significantly improved the prediction accuracy; however, further studies are needed to improve the generalization ability of the CNN model.     

Estimating energy consumption from night-time DMPS/OLS imagery after correcting for saturation effects

A methodology is presented to accurately estimate electric power consumption from saturated night-time Defense Meteorological Satellite Program (DMSP) Operational Linescan System (OLS) imagery using a stable light correction. An area correction for the stable light image of DMSP/OLS for the year 1999 was performed and the build-up area rate data were used to clarify the intensity distribution characteristics of the stable light. Based on the spatial distribution characteristics of the stable light, the saturation light of the electric power supply area of Japan was corrected using a cubic regression equation. The regression between the correction calculations by the cubic regression equation and the statistical electric power consumption data was applied in Japan and also in China, India and 10 other Asian countries. The correction method was then evaluated. This study confirms that electric power consumption can be estimated with high precision from the stable light.     

Semantic segmentation and colorization of grayscale aerial imagery with W-Net models

The semantic segmentation of remotely sensed aerial imagery is nowadays an extensively explored task, concerned with determining, for each pixel in an input image, the most likely class label from a finite set of possible labels. Most previous work in the area has addressed the analysis of high-resolution modern images, although the semantic segmentation of historical grayscale aerial photos can also have important applications. Examples include supporting the development of historical road maps, or the development of dasymetric disaggregation approaches leveraging historical building footprints. Following recent work in the area related to the use of fully-convolutional neural networks for semantic segmentation, and specifically envisioning the segmentation of grayscale aerial imagery, we evaluated the performance of an adapted version of the W-Net architecture, which has achieved very good results on other types of image segmentation tasks. Our W-Net model is trained to simultaneously segment images and reconstruct, or predict, the colour of the input images from intermediate representations. Through experiments with distinct data sets frequently used in previous studies, we show that the proposed W-Net architecture is quite effective in colouring and segmenting the input images. The proposed approach outperforms a baseline corresponding to the U-Net model for the segmentation of both coloured and grayscale imagery, and it also outperforms some of the other recently proposed approaches when considering coloured imagery.     

Estimating impervious surfaces from medium spatial resolution imagery: a comparison between fuzzy classification and LSMA

Impervious surfaces are important environmental indicators and are related to many environmental issues, such as water quality, stream health and the urban heat island effect. Therefore, detailed impervious surface information is crucial for urban planning and environment management. To extract impervious surfaces from remote sensing imagery, many algorithms and techniques have been developed. However, there are still debates over the strengths and limitations of linear versus nonlinear algorithms in handling mixed pixels in the urban landscapes. In the meantime, although many previous studies have compared various techniques, few comparisons were made between linear and nonlinear techniques. The objective of this study is to compare the performance between nonlinear and linear methods for impervious surface extraction from medium spatial resolution imagery. A linear spectral mixture analysis (LSMA) and a fuzzy classifier were applied to three Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) images acquired on 5 April 2004, 16 June 2001 and 3 October 2000, which covered Marion County, Indiana, United States. An aerial photo of Marion County with a spatial resolution of 0.14 m was used for validation of estimation results. Six impervious surface maps were yielded, and an accuracy assessment was performed. The root mean square error (RMSE), the mean average error (MAE), and the coefficient of determination (R ²) were calculated to indicate the accuracy of impervious surface maps. The results show that the fuzzy classification outperformed LSMA in impervious surface estimation in all seasons. For the June image, LSMA yielded a result with an RMSE of 13.2%, while the fuzzy classifier yielded an RMSE of 12.4%. For the April image, LSMA yielded an accuracy of 21.1% and the fuzzy classifier yielded 17.0%. For the October image, LSMA yielded a result with an RMSE of 19.8%, but the fuzzy classifier yielded an RMSE of 17.5%. Moreover, a subset image of the commercial, high-density and low-density residential areas was selected in order to compare the effectiveness of the developed algorithms for estimating impervious surfaces of different land use types. The result shows that the fuzzy classification was more effective than LSMA in both high-density and low-density residential areas. These areas prevailed with mixed pixels in the medium resolution imagery, such as ASTER. The results from the tested commercial area had a very high RMSE value due to the prevalence of shade in the area. It is suggested that the fuzzy classifier based on the nonlinear assumption can handle mixed pixels more effectively than LSMA.     

Reflectance calibration of aerial video imagery with automatic gain compensation on and off

Abstract

Video imaging systems are noi radiometrically calibrated, thus it is difficult to obtain quantitative remotely-sensed imagery for natural resource applications. Video camera automatic gain controls (AGC) present potential problems in calibrating video systems for quantitative analysis because they compensate for changing solar illumination conditions. In this experiment aerial video calibrations to ground reflectance standards were compared for AGC turned on and off, The calibrated aerial video was evaluated for guinea grass (Panicum maximum L.) biomass treatments on two dates. Results showed that there was more atmospheric light scattering in the red than in the NIR video band. Light scattering affects could be detected only when the AGC was ofT because when AGC was on light scattering affects were masked. These results also showed that video imagery produced significant correlations with guinea grass biomass that were comparable to ground reflectance measurements.     

Technical Note: Pseudo natural colour aerial imagery for urban and suburban mapping

Due to their near‐infrared data channel, digital airborne four‐channel imagers provide a potentially good discrimination between vegetation and human‐made materials, which is very useful in automated mapping. Due to their red, green and blue data channels, they also provide natural colour images, which are very useful in traditional (manual) mapping. In this paper, an algorithm is described which provides an approximation to the spectral capabilities of the four‐channel imagers by using a colour‐infrared aerial photo as input. The algorithm is tailored to urban/suburban surroundings, where the quality of the generated (pseudo) natural colour images are fully acceptable for manual mapping. This brings the combined availability of near‐infrared and (pseudo) natural colours within reach for mapping projects based on traditional photogrammetry, which is valuable since traditional analytical cameras still by far outnumber the relatively new family of digital airborne four‐channel imagers.     

Estimating impervious surfaces from medium spatial resolution imagery using the self-organizing map and multi-layer perceptron neural networks

The studies of impervious surfaces are important because they are related to many environmental problems, such as water quality, stream health, and the urban heat island effect. Previous studies have discussed that the self-organizing map (SOM) can provide a promising alternative to the multi-layer perceptron (MLP) neural networks for image classification at both per-pixel and sub-pixel level. However, the performances of SOM and MLP have not been compared in the estimation and mapping of urban impervious surfaces. In mid-latitude areas, plant phenology has a significant influence on remote sensing of the environment. When the neural networks approaches are applied, how satellite images acquired in different seasons impact impervious surface estimation of various urban surfaces (such as commercial, residential, and suburban/rural areas) remains to be answered. In this paper, an SOM and an MLP neural network were applied to three ASTER images acquired on April 5, 2004, June 16, 2001, and October 3, 2000, respectively, which covered Marion County, Indiana, United States. Six impervious surface maps were yielded, and an accuracy assessment was performed. The root mean square error (RMSE), the mean average error (MAE), and the coefficient of determination (R²) were calculated to indicate the accuracy of impervious surface maps. The results indicated that the SOM can generate a slightly better estimation of impervious surfaces than the MLP. Moreover, the results from three test areas showed that, in the residential areas, more accurate results were yielded by the SOM, which indicates that the SOM was more effective in coping with the mixed pixels than the MLP, because the residential area prevailed with mixed pixels. Results obtained from the commercial area possessed very high RMSE values due to the prevalence of shade, which indicates that both algorithms cannot handle the shade problem well. The lowest RMSE value was obtained from the rural area due to containing of less mixed pixels and shade. This research supports previous observations that the SOM can provide a promising alternative to the MLP neural network. This study also found that the impact of different map sizes on the impervious surface estimation is significant.     

Convolutional neural networks for object detection in aerial imagery for disaster response and recovery

Accurate and timely access to data describing disaster impact and extent of damage is key to successful disaster management (a process that includes prevention, mitigation, preparedness, response, and recovery). Airborne data acquisition using helicopter and unmanned aerial vehicle (UAV) helps obtain a bird’s-eye view of disaster-affected areas. However, a major challenge to this approach is robustly processing a large amount of data to identify and map objects of interest on the ground in real-time. The current process is resource-intensive (must be carried out manually) and requires offline computing (through post-processing of aerial videos). This research introduces and evaluates a series of convolutional neural network (CNN) models for ground object detection from aerial views of disaster’s aftermath. These models are capable of recognizing critical ground assets including building roofs (both damaged and undamaged), vehicles, vegetation, debris, and flooded areas. The CNN models are trained on an in-house aerial video dataset (named Volan2018) that is created using web mining techniques. Volan2018 contains eight annotated aerial videos (65,580 frames) collected by drone or helicopter from eight different locations in various hurricanes that struck the United States in 2017–2018. Eight CNN models based on You-Only-Look-Once (YOLO) algorithm are trained by transfer learning, i.e., pre-trained on the COCO/VOC dataset and re-trained on Volan2018 dataset, and achieve 80.69% mAP for high altitude (helicopter footage) and 74.48% for low altitude (drone footage), respectively. This paper also presents a thorough investigation of the effect of camera altitude, data balance, and pre-trained weights on model performance, and finds that models trained and tested on videos taken from similar altitude outperform those trained and tested on videos taken from different altitudes. Moreover, the CNN model pre-trained on the VOC dataset and re-trained on balanced drone video yields the best result in significantly shorter training time.     

Detection of forest reserve incursion in south-western Nigeria from a combination of multi-date aerial photographs and high resolution satellite imagery

Forest reserves were established in Nigeria to serve as repositories of the primary habitats of the forest ecosystems.However, with increasing population pressure and the need to feed the population, the peasant farmers are already making incursion into these reserves. This phenomenon has not attracted the serious attention of relevant government establishments, perhaps because they do not know the extent of the incursion in order to appreciate the magnitude of the danger it poses to the environment at large. This paper is therefore an attempt to quantify the extent of biotic degradation consequent upon the incursion. The study makes use of aerial photos and SPOT XS imagery and demonstrates the importance of combination of the two data sources to environmental monitoring in a country like Nigeria, with a poor geo-information/data bank.     

Using aerial imagery and digital photography to monitor growth and yield in winter wheat

Monitoring wheat (Triticum aestivum L.) performance throughout the growing season provides information on productivity and yield potential. Remote sensing tools have provided easy and quick measurements without destructive sampling. The objective of this study was to evaluate genetic variability in growth and performance of 20 wheat genotypes under two water regimes (rainfed and irrigated), using spectral vegetation indices (SVI) estimated from aerial imagery and percentage ground cover (%GC) estimated from digital photos. Field experiments were conducted at Bushland, Texas in two growing seasons (2014–2015 and 2015–2016). Digital photographs were taken using a digital camera in each plot, while a manned aircraft collected images of the entire field using a 12-band multiple camera array Tetracam system at three growth stages (tillering, jointing and heading). Results showed that a significant variation exists in SVI, %GC, aboveground biomass and yield among the wheat genotypes mostly at tillering and jointing. Significant relationships for %GC from digital photo at jointing was recorded with Normalized Difference Vegetation Index (NDVI) at tillering (coefficient of determination, R² = 0.84, p< 0.0001) and with %GC estimated from Perpendicular Vegetation Index (PVI) at tillering (R² = 0.83, p< 0.0001). Among the indices, Ratio Vegetation Index (RVI), Green-Red VI, Green Leaf Index (GLI), Generalized DVI (squared), DVI, Enhanced VI, Enhanced NDVI, and NDVI explained 37–99% of the variability in aboveground biomass and yield. Results indicate that these indices could be used as an indirect selection tool for screening a large number of early-generation and advanced wheat lines.