This PDF file contains the front matter associated with SPIE Proceedings Volume 10008, including the Title Page, Copyright information, Table of Contents, Introduction (if any), and Conference Committee listing.

New, high-resolution aerosol products are required in urban areas to improve the spatial coverage of the products, in terms of both resolution and retrieval frequency. These new products will improve our understanding of the spatial variability of aerosols in urban areas and will be useful in the detection of localized aerosol emissions. Urban aerosol retrieval is challenging for existing algorithms because of the high spatial variability of the surface reflectance, indicating the need for improved urban surface reflectance models. This problem can be stated in the language of novelty detection as the problem of selecting aerosol parameters whose effective surface reflectance spectrum is not an outlier in some space. In this paper, empirical orthogonal functions, a reconstruction-based novelty detection technique, is used to perform single-pixel aerosol retrieval using the single angular and temporal sample provided by the MODIS sensor. The empirical orthogonal basis functions are trained for different land classes using the MODIS BRDF MCD43 product. Existing land classification products are used in training and aerosol retrieval. The retrieval is compared against the existing operational MODIS 3 KM Dark Target (DT) aerosol product and co-located AERONET data. Based on the comparison, our method allows for a significant increase in retrieval frequency and a moderate decrease in the known biases of MODIS urban aerosol retrievals.

Communities surrounding local airports are becoming increasingly concerned about the aircraft pollutants emitted during the landing-takeoff (LTO) cycle, and their potential for negative health effects. Chicago, Los Angeles, Boston and London have all recently been featured in the news regarding concerns over the amount of airport pollution being emitted on a daily basis, and several studies have been published on the increased risks of cancer for those living near airports. There are currently no inexpensive, portable, and unobtrusive sensors that can monitor the spatial and temporal nature of jet engine exhaust plumes. In this work we seek to design a multispectral imaging system that is capable of tracking exhaust plumes during the engine idle phase, with a specific focus on unburned hydrocarbon (UHC) emissions. UHCs are especially potent to local air quality, and their strong absorption features allow them to act as a spatial and temporal plume tracer. Using a Gaussian plume to radiometrically model jet engine exhaust, we have begun designing an inexpensive, portable, and unobtrusive imaging system to monitor the relative amount of pollutants emitted by aircraft in the idle phase. The LWIR system will use two broadband filters to detect emitted UHCs. This paper presents the spatial and temporal radiometric models of the exhaust plume from a typical jet engine used on 737s. We also select filters for plume tracking, and propose an imaging system layout for optimal detectibility. In terms of feasibility, a multispectral imaging system will be two orders of magnitude cheaper than current unobtrusive methods (PTR-MS) used to monitor jet engine emissions. Large-scale impacts of this work will include increased capabilities to monitor local airport pollution, and the potential for better-informed decision-making regarding future developments to airports.

Two Doppler lidars were recently used to collect data from the planetary boundary layer (PBL) in Sao Paulo city (23°32’S, 46°38’W). The measurement campaign was carried out from December-2015 to February-2016, during the summer, which is the rainy season. Although Sao Paulo is the main city of a huge metropolitan region with more than 11 million of inhabitants and 7 millions of vehicles, according to the government agencies, the lack of PBL observational data is still a limitation for the atmospheric dispersion studies. Therefore, this work should contribute to the comprehension of PBL mechanisms and also for future atmospheric modeling studies. The data revealed that the nocturnal low-level jets (LLJs) frequently occurred along those 3 months, but its height is highly variable, from 100 m up to 650 m. It was also seen that the nocturnal LLJs can extend for several hours, right before the sunset until sunrise. This work aims to investigate the turbulence production by the nocturnal LLJs and its influence into the stable boundary layer (SBL).

The market for nano- and microsatellites is developing rapidly. There is a strong focus on 2D imaging of the Earth’s surface, with limited possibilities to obtain spectral information. More demanding applications, such as monitoring trace gases, aerosols or water quality still require advanced imaging instruments, which are large, heavy and expensive. In recent years TNO has investigated and developed different innovative designs to realize advanced spectrometers for space applications in a more compact and cost-effective manner. This offers multiple advantages: a compact instrument can be flown on a much smaller platform (nano- or microsatellite); a low-cost instrument opens up the possibility to fly multiple instruments in a satellite constellation, improving both global coverage and temporal sampling (e.g. to study diurnal processes); a constellation of low-cost instruments may provide added value to the larger scientific and operational satellite missions (e.g. the Copernicus Sentinel missions); and a small, lightweight spectrometer can also be mounted easily on a high-altitude UAV (offering high spatial resolution). Last but not least, a low-cost instrument may allow to break through the ‘cost spiral’: lower cost will allow to take more risk and thus progress more quickly. This may lead to a much faster development cycle than customary for current Earth Observation instruments. To explore the potential of a constellation of low-cost instruments a consortium of Dutch partners was formed, which currently consists of Airbus Defence and Space Netherlands, ISISpace, S and T and TNO. In this paper we will illustrate this new design approach by using the most advanced design of a hyperspectral imaging spectrometer (named ‘Spectrolite’) as an example. We will discuss the different design and manufacturing techniques that were used to realize this compact and low-cost design. Laboratory tests as well as the first preliminary results of airborne measurements with the Spectrolite breadboard will be presented and discussed. The design of Spectrolite offers the flexibility to tune its performance (spectral range, spectral resolution) to a specific application. Thus, based on the same basic system design, Spectrolite offers a range of applications to different clients. To illustrate this, we will present a mission concept to monitor NO2 concentrations over urban areas at high spatial resolution, based on a constellation of small satellites.

Hyperspectral imaging for gas detection applications is an under-researched topic. The same gas model is used in most of the gas detection studies in the literature. This model aims to formulate the scene covering the gas emission as well as the background and the atmosphere. Therefore, the model requires prior knowledge on transmittance, emissivity, and temperature values of the components in the scene. The commonly used approaches to estimate these parameters include atmospheric modeling and statistical inference. However, accessing such information is costly in remote detection applications. Some studies avoid background characterization by decomposing the scene using spectral-spatial information.

There are several studies in the literature using this model. They aim to detect various types of gases on different parts of electromagnetic spectrum. Most of these studies use hyperspectral radiance information regarding the scene. However, using brightness temperature map of the data instead of radiance data is more suitable for direct analysis. For this reason, we used brightness temperature spectrum in this study.

On the other hand, the detection algorithms are generally based on pixel based investigation. Since the emission of the gas is sourced by a pipe or a chimney, investigating the emission region at the segment level increases detection accuracy. In this study, we used an iterative spectral feature based pixel clustering algorithm followed by spatial segmentation.

The ChoT project aims at analysing the potential of aerial thermal imagery to produce large scale datasets for energetic efficiency analyses and policies in urban environments. It is funded by the Italian Ministry of Education, University and Research (MIUR) in the framework of the SIR 2014 (Scientific Independence of young Researchers) programme. The city of Bologna (Italy) was chosen as the case study. The acquisition of thermal infrared images at different times by multiple aerial flights is one of the main tasks of the project. The present paper provides an overview of the ChoT project, but it delves into some specific aspects of the data processing chain: the computing of the radiometric quantities of the atmosphere, the estimation of surface emissivity (through an object-oriented classification applied on a very high resolution multispectral image, to distinguish among the major roofing materials) and sky-view factor (by means of a digital surface model). To collect ground truth data, the surface temperature of roofs and road pavings was measured at several locations at the same time as the aircraft acquired the thermal images. Furthermore, the emissivity of some roofing materials was estimated by means of a thermal camera and a contact probe. All the surveys were georeferenced by GPS. The results of the first surveying campaign demonstrate the high sensitivity of the model to the variability of the surface emissivity and the atmospheric parameters.

Solar energy is emerging as one of the top options for renewable energy sources in the Philippines, with largescale solar photovoltaic (PV) farms being built all over the country. Solar energy resource in the urban environment has great potential in making a city self-sustaining, but has not been fully explored for the country. In order to represent its potential, reliable resource assessment should be done. This study aims to assess the available solar energy resource in Davao City, a trade and commerce hub in southern Philippines. The functions of GRASS GIS, specifically the r.sun module, in modelling incoming solar radiation is discussed, along with the use of a one-meter LiDAR Digital Surface Model (DSM) and Linke Turbidity coefficients as inputs. The average Julian day of each month was used to compute the Global Horizontal Irradiation (GHI) values under clear-sky or cloudless conditions. To account for the effects of the clouds in the study area, the clear-sky indices (Kc) were computed using data from solar recording stations of the Bureau of Soils and Water Management (BSWM) found within and around the region. These were multiplied to the modelled clear-sky GHI rasters to get the real-sky GHI. The results show that the city’s average GHI potential ranges from 2693.79 Wh/m² and 4453.13 Wh/m². Average values are particularly higher around the months of March and April, while lower values are seen in the months of November and January. Areas with higher potential are seen in the southern portion of the city, consistent in built-up areas.

Cities today have two major environmental concerns – carbon emissions and air quality. Global carbon levels are increasing and cities require to show plans to tackle and reduce the amount of carbon which they are emitting. At present carbon emissions in urban areas are calculated rather than measured. In some cities where industrial activity is not carbon intensive, the major contributors are the burning of fuel for heating and the emissions from vehicles. Air quality levels have a direct impact on human health and cities are under increased pressure to demonstrate plans to control and reduce levels of air pollution. Of great importance is the way in which emissions, both carbon rich emissions and pollutants, disperse in a city environment. Little work has been reported on the movement of CO2 in the urban environment and the effect the structure of the environment exerts on the movement and dispersion. This paper describes an investigation into the dispersion of CO2 within an urban environment in the Old Town of the City of Edinburgh, using a hand carried low cost portable CO2 sensor.

Preterm birth (PTB), one of the major concerns in obstetrics, is conventionally defined as the delivery of a live infant before 37 completed weeks of gestation, and one of its causes may be environmental factors. Remote sensing is a valuable approach for monitoring environmental variables, including in health sciences. In this work, remote sensing data were used to explore the relation of the environment with PTB. Time-series with monthly rates of male/female ratio and PTB were obtained from Portugal in 2000-2014. The environmental variables included in this study were monthly mean temperatures (T), relative humidity (RH), NDVI, concentrations of NO₂ and PM₁₀ in 2003-2008. A temporal and spatial analysis of each health-related and environmental variable was performed, as well as their correlation. PTB has been increasing over time, from below 5% in 2000 to around 7% in 2014, with predominance of higher rates in districts with larger population. From 2003 to 2008, T and PM₁₀ decreased significantly. A positive and significant correlation was found between male/female ratio and NO₂ and RH, and to a lesser extent with PM₁₀ and NDVI. PTB was also positively and significantly correlated with NO₂ and T, and to a lesser extent with RH and PM₁₀. These preliminary results suggest an association of PTB with most of the environmental variables studied, showing that more polluted and populated districts have higher rates of PTB. Further studies are warranted to explore interaction between the considered environmental factors and other variables related with risk for PTB.

H2020-Space project URBANFLUXES (URBan ANthrpogenic heat FLUX from Earth observation Satellites) investigates the potential of Copernicus Sentinels to retrieve anthropogenic heat flux, as a key component of the Urban Energy Budget (UEB). URBANFLUXES advances the current knowledge of the impacts of UEB fluxes on urban heat island and consequently on energy consumption in cities. This will lead to the development of tools and strategies to mitigate these effects, improving thermal comfort and energy efficiency. In URBANFLUXES, the anthropogenic heat flux is estimated as a residual of UEB. Therefore, the rest UEB components, namely, the net all-wave radiation, the net change in heat storage and the turbulent sensible and latent heat fluxes are independently estimated from Earth Observation (EO), whereas the advection term is included in the error of the anthropogenic heat flux estimation from the UEB closure. The project exploits Sentinels observations, which provide improved data quality, coverage and revisit times and increase the value of EO data for scientific work and future emerging applications. These observations can reveal novel scientific insights for the detection and monitoring of the spatial distribution of the urban energy budget fluxes in cities, thereby generating new EO opportunities. URBANFLUXES thus exploits the European capacity for space-borne observations to enable the development of operational services in the field of urban environmental monitoring and energy efficiency in cities.

With the development of urbanization, urban heat island effect issue is becoming more and more severe. What's more, the vegetation eco-environmental quality (VEEQ) is severely damaged, resulting in the decline of urban ecosystem function. Therefore, it is of great significance to use remote sensing technique to analyze the response of urban heat island to VEEQ quantitatively. As is known to all, vegetation is the main body in the vegetation ecological environment system. Water and heat conditions are the important driving forces for its formation and evolution. Good soil condition is the basis for vegetation survival. Besides, the terrain is conducive to the judgment of the vegetation distribution. Accordingly, several indexes involving vegetation index, heat index, soil moisture index, soil brightness index, elevation factor and slope factor were selected and extracted from Landsat8 OLI images to establish the evaluation index system of VEEQ. Based on Landsat8 TIRS images, this paper applied the radiative transfer equation method to retrieve land surface temperature (LST) and the urban island grade was divided based on the mean and standard deviation values of LST. The principal component analysis method was utilized to determine the weigh value of each index and then a comprehensive evaluation model of VEEQ was established. Furthermore, the quantitative relationship between LST and VEEQ was analyzed. The results showed that, there existed obvious heat island effects in Haidian District of Beijing city and its surrounding areas. The poor quality areas and the high quality areas of vegetation ecological environment had strengthening and weakening thermal environment effects respectively. There was a strong negative relationship between LST and VEEQ.

This paper focuses on thermal inertia estimation of Delhi and its surrounding areas during summer season based on the diurnal temperature variations and albedo information of the region retrieved from satellite data. The study involves mapping of day and night time surface temperatures and the blue sky albedo (actual albedo on ground) over the study region using Moderate Resolution Imaging Spectroradiometer (MODIS) datasets. The study reveals that Delhi is cooler than its surrounding regions during the day time, showing the formation of cool island. On the contrary, temperature inside the city is much higher than its surrounding rural regions during the night time, thus confirming the formation of nocturnal heat island. The day and night time temperature maps are then used to obtain the diurnal temperature range and together with albedo maps of the study region, are used to estimate the thermal inertia over the region. The study reveals that the dense built-up urban area of Delhi has higher thermal inertia than that of the surrounding rural areas during the summer season. The spatial variation of thermal inertia over the region is found to explain the occurrence of day-time cool island reasonably well.

Vegetation is essential in urban environments since it provides significant services in terms of health, heat, property value, ecology ... As part of the European Union Biodiversity Strategy Plan for 2020, the protection and development of green-infrastructures is strengthened in urban areas. In order to evaluate and monitor the quality of the green infra-structures, this article investigates contributions of Pléiades multi-angular images to extract and characterize low and high urban vegetation. From such images one can extract both spectral and elevation information from optical images. Our method is composed of 3 main steps : (1) the computation of a normalized Digital Surface Model from the multi-angular images ; (2) Extraction of spectral and contextual features ; (3) a classification of vegetation classes (tree and grass) performed with a random forest classifier. Results performed in the city of Rennes in France show the ability of multi-angular images to extract DEM in urban area despite building height. It also highlights its importance and its complementarity with contextual information to extract urban vegetation.

Because of the large increase of urban population in the last decades, the question of sustainable development in urban areas is crucial. In this context, vegetation plays a significant role in urban planning, environmental protecting, and sustainable development policy making, heating and cooling requirements of buildings, displacement of animals dispersion, concentration of pollutants, and well-being. In numerous cities, vegetation is limited to public areas using GPS surveys or aerial remote sensing data.

Recently, very high-resolution sensors as Light Detection and Ranging (LiDAR) data have permitted significant improvements in vegetation mapping in urban areas. This paper presents an evaluation of a new generation of airborne LIDAR bi-spectral discrete point (Optech titan) for mapping and characterizing urban vegetation. The methodology is based on a four-step approach: 1) the analysis of the quality of data in order to estimate noise between the green and near-infrared LIDAR point clouds; 2) this enables to remove the topographic effects and 3) a first classification, devoted to the elimination of the non-vegetation class, is performed based on the intensity value of the two channels; finally, in 4), the tree coverage is classified into seven categories of strata combination. To this end specific descriptors related to the organization of the point clouds are used.

These first results show that compared to monospectral LiDAR data, bi-spectral LiDAR enables to improve significantly both the extraction and the characterization of urban objects. This reveals new perspectives for mapping and characterizing urban patterns and other complex structures.

In this study a learning urban image spectral archive (LUISA) has been developed, that overcomes the issue of an incomplete spectral library and can be used to derive scene-specific pure material spectra. It consists of a well described starting spectral library (LUISA-A) and a tool to derive scene-based pure surface material spectra (LUISA-T). The concept is based on a three-stage approach: (1) Comparing hyperspectral image spectra with LUISA-A spectra to identify scene-specific pure materials, (2) extracting unknown pure spectra based on spatial and spectral metrics and (3) provides the framework to implement new surface material spectra into LUISA-A. The spectral comparison is based on several similarity measures, followed by an object- and spectral-based ruleset to optimize and categorize potentially new pure spectra.

The results show that the majority of pure surface materials could be identified using LUISA-A. Unknown spectra are composed of mixed pixels and real pure surface materials which could be distinguished by LUISA-T.

The value of characteristic reflectance features for mapping urban materials has been demonstrated in many experiments with airborne imaging spectrometry. Analysis of larger areas requires satellite-based multispectral imagery, which typically lacks the spatial and spectral detail of airborne data. Consequently the need arises to develop mapping methods that exploit the complementary strengths of both data sources. In this paper a workflow for sub-pixel quantification of Vegetation–Impervious–Soil urban land cover is presented, using medium resolution multispectral satellite imagery, hyperspectral endmember libraries and Support Vector Regression. A Landsat 8 Operational Land Imager surface reflectance image covering the greater metropolitan area of Brussels is selected for mapping. Two spectral libraries developed for the cities of Brussels and Berlin based on airborne hyperspectral APEX and HyMap data are used. First the combined endmember library is resampled to match the spectral response of the Landsat sensor. The library is then optimized to avoid spectral redundancy and confusion. Subsequently the spectra of the endmember library are synthetically mixed to produce training data for unmixing. Mapping is carried out using Support Vector Regression models trained with spectra selected through stratified sampling of the mixed library. Validation on building block level (mean size = 46.8 Landsat pixels) yields an overall good fit between reference data and estimation with Mean Absolute Errors of 0.06, 0.06 and 0.08 for vegetation, impervious and soil respectively. Findings of this work may contribute to the use of universal spectral libraries for regional scale land cover fraction mapping using regression approaches.

Remote sensing based land cover classification in urban areas generally requires the use of subpixel classification algorithms to take into account the high spatial heterogeneity. These spectral unmixing techniques often rely on spectral libraries, i.e. collections of pure material spectra (endmembers, EM), which ideally cover the large EM variability typically present in urban scenes. Despite the advent of several (semi-) automated EM detection algorithms, the collection of such image-specific libraries remains a tedious and time-consuming task. As an alternative, we suggest the use of a generic urban EM library, containing material spectra under varying conditions, acquired from different locations and sensors. This approach requires an efficient EM selection technique, capable of only selecting those spectra relevant for a specific image. In this paper, we evaluate and compare the potential of different existing library pruning algorithms (Iterative Endmember Selection and MUSIC) using simulated hyperspectral (APEX) data of the Brussels metropolitan area. In addition, we develop a new hybrid EM selection method which is shown to be highly efficient in dealing with both imagespecific and generic libraries, subsequently yielding more robust land cover classification results compared to existing methods. Future research will include further optimization of the proposed algorithm and additional tests on both simulated and real hyperspectral data.

The combination of rapid global urban growth and climate change has resulted in increased occurrence of major urban flood events across the globe. The distribution of flooded area is one of the key information layers for applications of emergency planning and response management. While SAR systems and technologies have been widely used for flood area delineation, radar images suffer from range ambiguities arising from corner reflection effects and shadowing in dense urban settings. A new mapping framework is proposed for the extraction and quantification of flood extent based on aerial optical multi-spectral imagery and ancillary data. This involves first mapping of flood areas directly visible to the sensor. Subsequently, the complete area of submergence is estimated from this initial mapping and inference techniques based on baseline data such as land cover and GIS information such as available digital elevation models. The methodology has been tested and proven effective using aerial photography for the case of the 2013 flood in Calgary, Canada.

Circular synthetic aperture radar (CSAR) can provide a full aspect coverage on interesting scenes in one run. Over the city of Karlsruhe a Ka-band dataset was generated in CSAR mode. The data was focused using subapertures in a step of 1.5°, each SAR image representing the scene from a slightly different aspect. The potential of non-coherent fusion of full aspect coverage to reveal small targets was demonstrated. By a manual selection of the viewing angle, parking cars next to high buildings could be revealed and a full view on selected targets with reduced shadow and overlay effects was shown. We studied the effect of varying aspects on the focused image pixels and developed a first metric to automatically select the best viewing angle to a local scene. Areas containing ground information like grass or asphalt and which are not hidden between high objects could be identified and used to deliver a good aspect view on neighboring areas which suffer from shadowing effects.

Every year thousands of people are displaced by conflicts or natural disasters and often gather in large camps. Knowing how many people have been gathered is crucial for an efficient relief operation. However, it is often difficult to collect exact information on the total number of the population.

This paper presents the improved morphological methodology for the estimation of dwellings structures located in several Internally Displaced Persons (IDPs) Camps, based on Very High Resolution (VHR) multispectral satellite imagery with pixel sizes of 1 meter or less including GeoEye-1, WorldView-2, QuickBird-2, Ikonos-2, Pléiades-A and Pléiades-B.

The main topic of this paper is the approach enhancement with selection of feature extraction algorithm, the improvement and automation of pre-processing and results verification. For the informal and temporary dwellings extraction purpose the high quality of data has to be ensured. The pre-processing has been extended by including the input data hierarchy level assignment and data fusion method selection and evaluation. The feature extraction algorithm follows the procedure presented in Jenerowicz, M., Kemper, T., 2011. Optical data are analysed in a cyclic approach comprising image segmentation, geometrical, textural and spectral class modeling aiming at camp area identification. The successive steps of morphological processing have been combined in a one stand-alone application for automatic dwellings detection and enumeration. Actively implemented, these approaches can provide a reliable and consistent results, independent of the imaging satellite type and different study sites location, providing decision support in emergency response for the humanitarian community like United Nations, European Union and Non-Governmental relief organizations.

Infrared thermography (IRT, or thermal video) uses thermographic cameras to detect and record radiation in the longwavelength infrared range of the electromagnetic spectrum. It allows sensing environments beyond the visual perception limitations, and thus has been widely used in many civilian and military applications. Even though current thermal cameras are able to provide high resolution and bit-depth images, there are significant challenges to be addressed in specific applications such as poor contrast, low target signature resolution, etc. This paper addresses quality improvement in IRT images for object recognition. A systematic approach based on image bias correction and deep learning is proposed to increase target signature resolution and optimise the baseline quality of inputs for object recognition. Our main objective is to maximise the useful information on the object to be detected even when the number of pixels on target is adversely small. The experimental results show that our approach can significantly improve target resolution and thus helps making object recognition more efficient in automatic target detection/recognition systems (ATD/R).

Building detection has been a prominent area in the area of image classification. Most of the research effort is adapted to the specific application requirements and available datasets. Our dataset includes aerial orthophotos (with spatial resolution 20cm), a DSM generated from LiDAR (with spatial resolution 1m and elevation resolution 20 cm) and DTM (spatial resolution 2m) from an area of Athens, Greece. Our aim is to classify these data by means of Markov Random Fields (MRFs) in a Bayesian framework for building block extraction and perform a comparative analysis with other supervised classification techniques namely Feed Forward Neural Net (FFNN), Cascade-Correlation Neural Network (CCNN), Learning Vector Quantization (LVQ) and Support Vector Machines (SVM). We evaluated the performance of each method using a subset of the test area. We present the classified images, and statistical measures (confusion matrix, kappa coefficient and overall accuracy). Our results demonstrate that the MRFs and FFNN perform better than the other methods.

The algorithm ICARE-HS (Inversion Code for urban Areas Reflectance Extraction using HyperSpectral imagery) is presented in this paper. ICARE-HS processes airborne hyperspectral images for atmospheric compensation taking into account the strong relief of urban areas. A digital surface model is used to provide the 3D information, which is key to simulating relief-related effects such as shadow casting, multiple reflections between objects and variable illumination depending on local solid angle of view of the sky. Some of these effects are modeled using ray tracing techniques. ICARE-HS is applied to airborne hyperspectral data of the city center of Toulouse, which are also processed by a standard atmospheric correction method for comparison.

Robust detection of road network in aerial imagery is a challenging task since roads have different pavement texture, road-side surroundings, as well as grades. Roads of different grade have different curvilinear saliency in the aerial imagery. This paper is motivated to incrementally extract roads and construct the topology of the road network of aerial imagery from the higher-grade-first perspective. Inspired by the spanning tree technique, the proposed method starts from the robust extraction of the most salient road segment(s) of the road network, and incrementally connects segments of less saliency of curvilinear structure until all road segments in the network are extracted. The proposed algorithm includes: curvilinear path-based road morphological enhancement, extraction of road segments, and spanning tree search for the incremental road discovery. It is tested on a diverse set of aerial imagery acquired in the city and inter-city areas. Experimental results show that the proposed curvilinear spanning tree (CST) can detect roads efficiently and construct the topology of the road network effectively. It is promising for the change detection of the road network.

Pole-like objects, including trees, lampposts and traffic signs, are indispensable part of urban infrastructure. With the advance of vehicle-based laser scanning (VLS), massive point cloud of roadside urban areas becomes applied in 3D digital city modeling. Based on the property that different pole-like objects have various canopy parts and similar trunk parts, this paper proposed the 3D part-based shape analysis to robustly extract, identify and model the pole-like objects. The proposed method includes: 3D clustering and recognition of trunks, voxel growing and part-based 3D modeling. After preprocessing, the trunk center is identified as the point that has local density peak and the largest minimum inter-cluster distance. Starting from the trunk centers, the remaining points are iteratively clustered to the same centers of their nearest point with higher density. To eliminate the noisy points, cluster border is refined by trimming boundary outliers. Then, candidate trunks are extracted based on the clustering results in three orthogonal planes by shape analysis. Voxel growing obtains the completed pole-like objects regardless of overlaying. Finally, entire trunk, branch and crown part are analyzed to obtain seven feature parameters. These parameters are utilized to model three parts respectively and get signal part-assembled 3D model. The proposed method is tested using the VLS-based point cloud of Wuhan University, China. The point cloud includes many kinds of trees, lampposts and other pole-like posters under different occlusions and overlaying. Experimental results show that the proposed method can extract the exact attributes and model the roadside pole-like objects efficiently.

Recent advancements in lidar technologies have led to significant improvements in Teledyne Optech’s airborne lidar systems. This paper will present the performance enhancements that have led to the creation of the Galaxy, a compact scanning lidar system. Unlike the previous generation of conventional airborne lidar, the Galaxy offers fundamentally improved specifications for long-range airborne lidar systems. The Galaxy system is capable of acquiring high-density, multiple-return data with unique pulse separation characteristics and exceptional precision. Utilizing discrete time-of-flight measurement electronics, this new system is capable of seamlessly operating at very high laser repetition rates through blind zones and with multiple pulses in the air. By utilizing even higher scan products , the system outperforms previous generations of systems and optimizes point density during collection.

Both in military and civil applications, there is an urgent need for a highly up-to-date road data, which should be ideally semantically structured (into main roads, walking paths, escape ways, etc.) with application-driven attributes, such as road width, road type, surface condition and many others. A vectorization algorithm processing aerial images recently acquired yields an up-to-date road vector data, which are, however, often represented by wriggly, noisy polylines without semantics. The reasons for zigzagged street courses are insufficiencies in the intermediate results of sensor data processing (orthophotos, elevation maps) and occlusions caused by trees, buildings, and others. In the current contribution, an improved computation of geometric attributes will be explained which makes a difference between straight and circular (or elliptic) polylines. Using improved attributes, the candidates for polylines having identical course and sharing a junction are determined. From such candidates, we form chains of polylines. These chains correspond better to the intuitive perception of the term street than the previously used road polylines, because, even after being interrupted by narrower side roads, a chain maintains its label. The generalization of chains with simultaneously adjusting positions of junctions is evidently performed. We apply a generalization with the purpose-based modification of a well-known polyline simplification algorithm once chain-wise and once polyline-wise in order to show - by means of qualitative results - the advantages of the chain-wise generalization.

Deterioration of asphalt road pavements is inevitable throughout its life cycle. There are several types of deterioration that take place on these surfaces, like surface defects and deformations. One of the most common asphalt defects is cracking. Fatigue, transverse, longitudinal, reflective, edge, block and slippage are types of cracking that can be observed anywhere in the world. Monitoring and preventative/periodic maintenance of these types of wears are two very important actions that have to take place to avoid “costly” solutions. This paper aims to introduce the spectral characteristics of uncracked (healthy) and cracked asphalt surfaces which can give a new asphalt crack index. This is performed through remote sensing applications in the area of asphalt pavements. Multispectral images can be elaborated using the index to enhance crack marks on asphalt surfaces. Ground spectral signatures were acquired from both uncracked and cracked asphalted areas of Cyprus (Limassol). Evaluation separability indices can be used to identify the optimum wavelength regions that can distinguish better the uncracked and cracked asphalt surfaces. The results revealed that the spectral sensitivity for the enhancement of cracked asphalt was detected using the Euclidean, Mahalanobis and Cosine Distance Indices in the Vis range (approximately at 450 nm) and in the SWIR 1 range (approximately at 1750 nm).

This paper describes the application of the monolithic UNISA Folded Pendulum, optimized as inertial sensor (seismometer) for low frequency characterization of sites (including underground sites) and structures (e.g. buildings, bridges, historical monuments), but, in general, for applications requiring large band low-frequency performances coupled with high sensitivities. The main characteristics of this class of sensors are high sensitivity, large measurement band, compactness, lightness, scalability, tunability of the resonance frequency, low thermal noise and very good immunity to environmental noises. The horizontal and vertical versions of folded pendulum allow an effective state-of-the-art mechanical implementation of triaxial sensors, configurable both as seismometer and/or as accelerometer.

With the popularity of very-high-resolution (VHR) aerial imagery, the shape, color, and context attribute of vehicles are better characterized. Due to the various road surroundings and imaging conditions, vehicle attributes could be adversely affected so that vehicle is mistakenly detected or missed. This paper is motivated to robustly extract the rich attribute feature for detecting the vehicles of VHR imagery under different scenarios. Based on the hierarchical component tree of vehicle context, attribute belief propagation (ABP) is proposed to detect salient vehicles from the statistical perspective. With the Max-tree data structure, the multi-level component tree around the road network is efficiently created. The spatial relationship between vehicle and its belonging context is established with the belief definition of vehicle attribute. To effectively correct single-level belief error, the inter-level belief linkages enforce consistency of belief assignment between corresponding components at different levels. ABP starts from an initial set of vehicle belief calculated by vehicle attribute, and then iterates through each component by applying inter-level belief passing until convergence. The optimal value of vehicle belief of each component is obtained via minimizing its belief function iteratively. The proposed algorithm is tested on a diverse set of VHR imagery acquired in the city and inter-city areas of the West and South China. Experimental results show that the proposed algorithm can detect vehicle efficiently and suppress the erroneous effectively. The proposed ABP framework is promising to robustly classify the vehicles from VHR Aerial imagery.

Increased emissions of anthropogenic aerosols associated with economic growth can lead to increased concentrations of hazardous air pollutants. In particular, large cities in East Asia have experienced numerous heavy haze episodes. Atmospheric aerosol distributions in East Asia are complex, being influenced by both natural phenomena and human activity, with urban areas in particular being dominated by fine anthropogenic aerosols released from diesel-powered vehicles and industrial activity. In Japan, air pollution levels have been reduced; nevertheless, in recent years, there is increasing concern regarding air pollution caused by fine particulate matter. The origins of air pollution were examined, focusing on the comparison between aerosol properties observed from satellites and that on the ground. Because of their short life spans, concentrations of anthropogenic aerosols are highest over the source regions, and as a result, the climatic impacts of anthropogenic aerosols are also found to be most pronounced in these regions. In this study, aerosol impacts on climate are assessed by numerical model simulations. The direct effects of aerosols include reduced solar radiation, and hence a decrease in surface temperatures. In addition to these changes in the radiation budget, aerosols have a significant potential to change cloud and precipitation fields. These climatic responses to aerosols can manifest far from their source regions with high industrial activities.

Excessive emission of light or light pollution at night is one of the elements of environmental pollution. Indirectly light pollution causes increase of fossil fuel use, greenhouse gasses and pollution in the atmosphere. Direct effects of light pollution including: disturbance of animals life, human’s psychology and environmental degradation. Light pollution in an area is related with the existence of built-up area and the lack of vegetation as a manifestation of economic and population growth. This research aims to know the relation of land cover changes with light pollution in Bandar Lampung and surrounding with 40 km radius over the last ten years. This research used satellite imagery to obtained data and later does the verification and accuracy tests on the field. The variables used in this research include light pollution radiance value, percentages in the built-up area and vegetation density. Light pollution radiance value is obtained from DMSP-OLS Version 4 satellite images, while the changes of built up and vegetation density data obtained from NDBI dan NDVI from Landsat 8 satellite images. The research area is divided into a grid with a size of 30”×30” which is the same as spatial resolution of DMSP. From sample grids, regression analysis between the percentage of light pollution radiance value with the percentage of NDVI and NDBI index on each grids. The percentages of built up areas and vegetation has 58 % of fair correlation with light emission.

Increased emissions of anthropogenic aerosols associated with economic growth can lead to increased concentrations of hazardous air pollutants. Furthermore, dust storms or biomass burning plumes can cause serious environmental hazards, yet their aerosol properties are poorly understood. Our research group has worked on the development of an efficient algorithm for aerosol retrieval during hazy episodes (dense concentrations of atmospheric aerosols). It is noted that near UV measurements are available for detection of carbonaceous aerosols. The biomass burning aerosols (BBA) due to large-scale forest fires and/or burn agriculture exacerbated the severe air pollution. It is known that global warming and climate change have caused increasing instances of forest fires, which have in turn accelerated climate change. It is well known that this negative cycle decreases the quality of the global environment and human health.

The Japan Aerospace Exploration Agency (JAXA) has been developing a new Earth observing system, the GCOM (Global Change Observation Mission) project, which consists of two satellite series: GCOM-W1 and GCOM-C1. The first GCOM-C satellite will board the SGLI (second generation GLI [global imager]) to be launched in early 2017. The SGLI is capable of multi-channel (19) observation, including a near UV channel (0.380 μm) and two polarization channels at red and near-infrared wavelengths of 0.67 and 0.87 μm. Thus, global aerosol retrieval will be achieved with simultaneous polarization and total radiance.

In this study, algorithm improvement for aerosol remote sensing, especially of BBA episodes, is examined using Terra/MODIS measurements from 2003, when the GLI and POLDER-2 sensors were working onboard the Japanese satellite ADEOS-2.

This study investigated the influences of urban growth and heat waves events on Urban Heat Island in relationship with several biophysical variables in Bucharest metropolitan area of Romania through satellite and in-situ monitoring data. Remote sensing data from Landsat TM/ETM+ and time series MODIS Terra/Aqua sensors have been used to assess urban land cover– temperature interactions over period between 2000 and 2016 years. Vegetation abundances and percent impervious surfaces were derived by means of linear spectral mixture model, and a method for effectively enhancing impervious surface has been developed to accurately examine the urban growth. The land surface temperature (Ts), a key parameter for urban thermal characteristics analysis, was also analyzed in relation with the Normalized Difference Vegetation Index (NDVI) at city level. Based on these parameters, the urban growth, urban heat island effect (UHI) and the relationships of Ts to other biophysical parameters (surface albedo, precipitations, wind intensity and direction) have been analyzed. Results show that in the metropolitan area ratio of impervious surface in Bucharest increased significantly during investigated period, the intensity of urban heat island and heat wave events being most significant. The correlation analyses revealed that, at the pixel-scale, Ts possessed a strong positive correlation with percent impervious surfaces and negative correlation with vegetation abundances at the regional scale, respectively. This analysis provided an integrated research scheme and the findings can be very useful for urban ecosystem modeling.

The increase of urban atmospheric pollution due to particulate matters (PM) in different fraction sizes affects seriously not only human health and environment, but also city climate directly and indirectly. In the last decades, with the economic development and the increased emissions from industrial, traffic and domestic pollutants, the urban atmospheric pollution with remarkable high PM2.5 (particulate matters with aerodynamic diameter less than 2.5 μm) and PM10 (particulate matters with aerodynamic diameter less than 10 μm) concentration levels became serious in the metropolitan area of Bucharest in Romania. Both active as well as satellite remote sensing are key applications in global change science and urban climatology. The aerosol parameters can be measured directly in situ or derived from satellite remote sensing observations. All these methods are important and complementary. The current study presents a spatiotemporal analysis of the aerosol concentrations in relation with climate parameters in two size fractions (PM10 and PM2.5) in Bucharest metropolitan area. Daily average particle matters concentrations PM10 and PM2.5 for Bucharest metropolitan area have been provided by 8 monitoring stations belonging to air pollution network of Environmental Protection Agency. The C005 (version 5.1) Level 2 and Level 3 Terra and Aqua MODIS AOD550 time-series satellite data for period 01/01/2011- 31/12/2012 have been also used. Meteorological variables (air temperature, relative humidity, sea level atmospheric pressure) have been provided by in-situ measurements. Both in-situ monitoring data as well as MODIS Terra/Aqua time-series satellite data for 2011-2012 period provided useful tools for particle matter PM2.5 and PM10 monitoring.

This paper presents a system developed by an application of a neural network Multilayer Perceptron for drone acquired agricultural image segmentation. This application allows a supervised user training the classes that will posteriorly be interpreted by neural network. These classes will be generated manually with pre-selected attributes in the application. After the attribute selection a segmentation process is made to allow the relevant information extraction for different types of images, RGB or Hyperspectral. The application allows extracting the geographical coordinates from the image metadata, geo referencing all pixels on the image. In spite of excessive memory consume on hyperspectral images regions of interest, is possible to perform segmentation, using bands chosen by user that can be combined in different ways to obtain different results.

Falls are the most critical health problem for elderly people, which are often, cause significant injuries. To tackle a serious risk that made by the fall, we develop an automatic wearable fall detection system utilizing two devices (mobile phone and wireless sensor) based on three axes accelerometer signals. The goal of this study is to find an effective machine learning method that distinguish falls from activities of daily living (ADL) using only a single triaxial accelerometer. In addition, comparing the performance results for wearable sensor and mobile device data .The proposed model detects the fall by using seven different classifiers and the significant performance is demonstrated using accuracy, recall, precision and F-measure. Our model obtained accuracy over 99% on wearable device data and over 97% on mobile phone data.