Remote sensing imagery needs to be converted into tangible information which can be utilised in conjunction with other data sets, often within widely used Geographic Information Systems (GIS). As long as pixel sizes remained typically coarser than, or at the best, similar in size to the objects of interest, emphasis was placed on per-pixel analysis, or even sub-pixel analysis for this conversion, but with increasing spatial resolutions alternative paths have been followed, aimed at deriving objects that are made up of several pixels. This paper gives an overview of the development of object based methods, which aim to delineate readily usable objects from imagery while at the same time combining image processing and GIS functionalities in order to utilize spectral and contextual information in an integrative way. The most common approach used for building objects is image segmentation, which dates back to the 1970s. Around the year 2000 GIS and image processing started to grow together rapidly through object based image analysis (OBIA - or GEOBIA for geospatial object based image analysis). In contrast to typical Landsat resolutions, high resolution images support several scales within their images. Through a comprehensive literature review several thousand abstracts have been screened, and more than 820 OBIA-related articles comprising 145 journal papers, 84 book chapters and nearly 600 conference papers, are analysed in detail. It becomes evident that the first years of the OBIA/GEOBIA developments were characterised by the dominance of ‘grey’ literature, but that the number of peer-reviewed journal articles has increased sharply over the last four to five years. The pixel paradigm is beginning to show cracks and the OBIA methods are making considerable progress towards a spatially explicit information extraction workflow, such as is required for spatial planning as well as for many monitoring programmes.

Object based image analysis for remote sensing

This paper is a survey of the theory and methods of photogrammetric bundle adjustment, aimed at potential implementors in the computer vision community. Bundle adjustment is the problem of refining a visual reconstruction to produce jointly optimal structure and viewing parameter estimates. Topics covered include: the choice of cost function and robustness; numerical optimization including sparse Newton methods, linearly convergent approximations, updating and recursive methods; gauge (datum) invariance; and quality control. The theory is developed for general robust cost functions rather than restricting attention to traditional nonlinear least squares.

/pdf/bundle-adjustment-a-modern-synthesis-449tfs1h4g.pdf

Bundle Adjustment - A Modern Synthesis

https://pure.aber.ac.uk/portal/files/4916016/Westoby_et_al._GEOMORPH_2012_Accepted.pdf

‘Structure-from-Motion’ photogrammetry: A low-cost, effective tool for geoscience applications

https://researchweb.iiit.ac.in/~sai.deepak/lectures/14.%20Thermal%20remote%20sensing%20of%20urban%20climates%202003%20Voogt%20and%20Oke.pdf

Thermal remote sensing of urban climates

Articles R emote sensing has facilitated extraordinary advances in the modeling, mapping, and understanding of ecosystems. Typical applications of remote sensing involve either images from passive optical systems, such as aerial photography and Landsat Thematic Mapper (Goward and Williams 1997), or to a lesser degree, active radar sensors such as RADARSAT (Waring et al. 1995). These types of sensors have proven to be satisfactory for many ecological applications , such as mapping land cover into broad classes and, in some biomes, estimating aboveground biomass and leaf area index (LAI). Moreover, they enable researchers to analyze the spatial pattern of these images. However, conventional sensors have significant limitations for ecological applications. The sensitivity and accuracy of these devices have repeatedly been shown to fall with increasing aboveground biomass and leaf area index (Waring et al. 1995, Carlson and Ripley 1997, Turner et al. 1999). They are also limited in their ability to represent spatial patterns: They produce only two-dimensional (x and y) images, which cannot fully represent the three-dimensional structure of, for instance, an old-growth forest canopy.Yet ecologists have long understood that the presence of specific organisms, and the overall richness of wildlife communities, can be highly dependent on the three-dimensional spatial pattern of vegetation (MacArthur and MacArthur 1961), especially in systems where biomass accumulation is significant (Hansen and Rotella 2000). Individual bird species, in particular, are often associated with specific three-dimensional features in forests (Carey et al. 1991). In addition, other functional aspects of forests, such as productivity, may be related to forest canopy structure. Laser altimetry, or lidar (light detection and ranging), is an alternative remote sensing technology that promises to both increase the accuracy of biophysical measurements and extend spatial analysis into the third (z) dimension. Lidar sensors directly measure the three-dimensional distribution of plant canopies as well as subcanopy topography, thus providing high-resolution topographic maps and highly accurate estimates of vegetation height, cover, and canopy structure. In addition , lidar has been shown to accurately estimate LAI and aboveground biomass even in those high-biomass ecosystems where passive optical and active radar sensors typically fail to do so. The basic measurement made by a lidar device is the distance between the sensor and a target surface, obtained by determining the elapsed time between the emission of a short-duration laser pulse and the arrival of the reflection of that pulse (the return signal) at the sensor's receiver. Multiplying this …

https://www.jstor.org/stable/10.1641/0006-3568(2002)052%5B0019:LRSFES%5D2.0.CO;2

Lidar Remote Sensing for Ecosystem Studies

Geometric potential of MOMS-02/D2 data for point positioning, DTM and orthoimage generation

Photogrammetry and remote sensing have proven their efficiency for spatial data collection in many ways. Interactive mapping at digital workstations is performed by skilled operators, which guarantees excellent quality in particular of the geometric data. In this way, worldwide acquisition of a large number of national GIS databases has been supported and still a lot of production effort is devoted to this task. In the field of image analysis, it has become evident that algorithms for scene interpretation and 3D reconstruction of topographic objects, which rely on a single data source, cannot function efficiently. Research in two directions promises to be more successful. Multiple largely complementary sensor data like range data from laser scanners or SAR and panchromatic or multispectral aerial images have been used to achieve robustness and better performance in image analysis. On the other hand, given GIS databases, e.g. layers from topographic maps, can be considered as virtual sensor data which contain geometric information together with its explicitly given semantics. In this case, image analysis aims at supplementing missing information, e.g. the extraction of the third dimension for 2D databases. A second goal, which is expected to become more important in future, is the revision and update of existing GIS databases. In this paper, we review recent developments in the overlapping area of image analysis and GIS. On the data side, we focus on different sensor data in conjunction with GIS databases. Analysis of these data addresses almost all aspects of knowledge based image analysis like detection, localisation, reconstruction and identification. The paper will focus on use of GIS databases to support image analysis and the opposite, as well as fusion of multiple cues from cooperative algorithms for object extraction, reconstruction and classification. Conceptual aspects behind new developments will also be described. In general, processes exploiting different information sources often have a lower algorithmic complexity compared with single sensor data processing. This, for example, was shown with building reconstruction based on range data and a given ground plan of the buildings. Another example is map update using spectral and spatial resolution satellite images. Given a topographic map supervised classification can be executed with the result that inconsistencies between map information and image information regarding geometry and semantics can be detected and localised. With this review we aim at summarising work of our InterCommission Working Group IV/III.2 having in mind to promote further activities in this exciting field.

/pdf/integration-of-image-analysis-and-gis-mdgepu10al.pdf

Integration of image analysis and GIS

Within the EU project Cloudmap, cloud-top heights should be estimated using satellite stereo images at very high temporal resolution. Since such images are not provided by operational sensors, MOMS-2P and ATSR2 images of lower temporal resolution were used instead. The MOMS stereopair was oriented with subpixel accuracy using GCPs from 1:50,000 topographic maps and Kratky’s sensor model. Preprocessing for noise reduction, cloud stripes removal and contrast enhancement was applied. For the estimation of cloud-top heights, the images were resampled to 288 m and geometrically constrained least squares matching, using image pyramids and an interest operator, as well as varying parameters was used. The results were checked by visual inspection and comparison to semi-automatically measured points in the original resolution images. Automatic blunder detection using two tests were also applied. Matching led to large blunders in land areas between clouds or close to cloud boundaries. Excluding these blunders (error > 1100 m), matching showed an RMS of ca. 0.2 pixel, exhibiting a very high accuracy potential. A matching geometric transformation using rotations and radiometric equalization during the iterations showed slightly better results compared to the other matching versions. ATSR2 images were matched with a similar approach, however without geometric constraints, as the input images were rectified. Due to differences between the images which vary spatially, varying matching parameters are optimal for each image region. First steps in combining matching results from such varying matching versions have been performed. Both datasets showed similar matching problems due to surface discontinuities, mixing of surfaces than are neighbouring in image space but differ in height, and often large illumination differences (even with along-track stereo and small time acquisition differences).

/pdf/cloud-top-height-estimation-from-satellite-stereopairs-for-asq7vq5lcr.pdf

Cloud-top height estimation from satellite stereopairs for weather forecasting and climate change analysis

The extraction of road networks from aerial images is one of the current challenges in digital photogrammetry and computer vision. In this paper, we present our system for 3D road network reconstruction from aerial images using knowledge-based image analysis. In contrast to other approaches, the developed system integrates processing of color image data and information from digital spatial databases, extracts and fuses multiple object cues, takes into account context information, employs existing knowledge, rules and models, and treats each road subclass accordingly. The key of the system is the use of knowledge as much as possible to increase success rate and reliability of the results, working in 2D images and 3D object space, and use of 2D and 3D interaction when needed. Another advantage of the developed system is that it can correctly and reliably handle problematic areas caused by shadows and occlusions. This work is part of a project to improve and update the 1:25,000 vector maps of Switzerland. The system has been implemented as a standalone software package. We tested the system on a number of images in different landscapes. In this paper we present the results of our system in recent benchmark tests conducted independently by our project partner in Switzerland, and test results with black and white images in a test site in Belgium.

/pdf/improving-cartographic-road-databases-by-image-analysis-4dvaw1sruy.pdf

Improving cartographic road databases by image analysis

Investigations on the use of stereo GEO and multitemporal GEO Ikonos images for DSM generation are reported. Both an own developed matching algorithm and the digital photogrammetric systems LH Systems DPW770 and VirtuoZo were used. In our own matching method, we start with quasi-epipolar image generation, and derivation of approximations using grid point matching based on relaxation. The final matching is a modified multi-image geometrically constrained method. Test results on both methods are presented. Firstly, DSMs were generated from multitemporal GEO images, which theoretically should be more difficult to match, by using our own matching procedure. Different matching versions, with and without exclusion of the different cloud areas, were tried and compared to a reference DTM with an accuracy of ca. 3 m. The results showed a very promising RMS of slightly over 3 m without any manual editing. The second example refers to a DSMs generation from a GEO stereo-pair by matching using LHS DPW770 and VirtuoZo and compared to results from aerial imagery. Due to large multitemporal differences between the two datasets, stable points were manually selected and measured. These points and other points in stable open areas that were automatically matched in the aerial imagery using VirtuoZo served as reference data. The results showed a high relative accuracy of ca. 1 m but the presence of a larger systematic error of 2-3 m.

Emmanuel P. Baltsavias

Papers

Geometric potential of MOMS-02/D2 data for point positioning, DTM and orthoimage generation

Integration of image analysis and GIS

Cloud-top height estimation from satellite stereopairs for weather forecasting and climate change analysis

Improving cartographic road databases by image analysis

Matching of Ikonos stereo and multitemporal GEO Images for DSM Generation