Search NASA⌕ Search

SEARCH · Search NASA

Results for “change detection”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 records

Imaging, object detection, and change detection with a polarized multistatic GPR array

A polarized detection system performs imaging, object detection, and change detection factoring in the orientation of an object relative to the orientation of transceivers. The polarized detection system may operate on one of several modes of operation based on whether the imaging, object detection, or change detection is performed separately for each transceiver orientation. In combined change mode, the polarized detection system performs imaging, object detection, and change detection separately for each transceiver orientation, and then combines changes across polarizations. In combined object mode, the polarized detection system performs imaging and object detection separately for each transceiver orientation, and then combines objects across polarizations and performs change detection on the result. In combined image mode, the polarized detection system performs imaging separately for each transceiver orientation, and then combines images across polarizations and performs object detection followed by change detection on the result.

47 OTHER INSTRUMENTATION↗

Change detection and change monitoring of natural and man-made features in multispectral and hyperspectral satellite imagery

An approach for land cover classification, seasonal and yearly change detection and monitoring, and identification of changes in man-made features may use a clustering of sparse approximations (CoSA) on sparse representations in learned dictionaries. A Hebbian learning rule may be used to build multispectral or hyperspectral, multiresolution dictionaries that are adapted to regional satellite image data. Sparse image representations of pixel patches over the learned dictionaries may be used to perform unsupervised k-means clustering into land cover categories. The clustering process behaves as a classifier in detecting real variability. This approach may combine spectral and spatial textural characteristics to detect geologic, vegetative, hydrologic, and man-made features, as well as changes in these features over time.

47 OTHER INSTRUMENTATION↗

Evaluation of change detection techniques for monitoring coastal zone environments

Development of satisfactory techniques for detecting change in coastal zone environments is required before operational monitoring procedures can be established. In an effort to meet this need a study was directed toward developing and evaluating different types of change detection techniques, based upon computer aided analysis of LANDSAT multispectral scanner (MSS) data, to monitor these environments. The Matagorda Bay estuarine system along the Texas coast was selected as the study area. Four change detection techniques were designed and implemented for evaluation: (1) post classification comparison change detection, (2) delta data change detection, (3) spectral/temporal change classification, and (4) layered spectral/temporal change classification. Each of the four techniques was used to analyze a LANDSAT MSS temporal data set to detect areas of change of the Matagorda Bay region.

Weismiller, R. A.↗

Scalable time series change detection for biomass monitoring using gaussian process

Biomass monitoring, specifically detecting changes in the biomass or vegetation of a geographical region, is vital for studying the carbon cycle of the system and has significant implications in the context of understanding climate change and its impacts. Recently, several time series change detection methods have been proposed to identify land cover changes in temporal profiles (time series) of vegetation collected using remote sensing instruments. In this paper, we adapt Gaussian process regression to detect changes in such time series in an online fashion. While Gaussian process (GP) has been widely used as a kernel based learning method for regression and classification, their applicability to massive spatio-temporal data sets, such as remote sensing data, has been limited owing to the high computational costs involved. In this paper we address the scalability aspect of GP based time series change detection. Specifically, we exploit the special structure of the covariance matrix generated for GP analysis to come up with methods that can efficiently estimate the hyper-parameters associated with GP as well as identify changes in the time series while requiring a memory footprint which is linear in the size of input data, as compared to traditional method which involves solving a linear system of equations for the Choleksy decomposition of the quadratic sized covariance matrix. Experimental results show that our proposed method achieves significant speedups, as high as 1000, when processing long time series, while maintaining a small memory footprint. To further improve the computational complexity of the proposed method, we provide a parallel version which can concurrently process multiple input time series using the same set of hyper-parameters. The parallel version exploits the natural parallelization potential of the serial algorithm and is shown to perform significantly better than the serial version, with speedups as high as 10. Finally, we demonstrate the effectiveness of the proposed change detection method in identifying changes in Normalized Difference Vegetation Index (NDVI) data. Moreover, we show that the scalable solution is able to process NDVI data for the entire Iowa region significantly faster than the standard method.

Chandola, Varun↗

Potentials for change detection using Seasat synthetic aperture radar data

The use of synthetic aperture radar (SAR) images for detecting change on the earth's surface is highly dependent on target orientation, azimuth angle, and sensor depression angle. SAR data can be used for change detection when consistency is maintained in radar wavelength, polarization, azimuth directions, and off-nadir depression angle. The interaction of these parameters and the imaged surface for change detection are shown in examples drawn from (1) Los Angeles, CA, (2) southern Florida, (3) Imperial Valley, CA, (4) a desert region west of Tucson, AZ, and (5) western Kansas. SAR imagery is used to emphasize the geometric form, and roughness, of the earth's surface. As changes in the roughness of the surface occur over time, temporal SAR images will indicate those differences. Several guidelines for change detection studies using imaging radar are derived from the examples.

Bryan, M. L.↗

Detecting Change in Urban Areas at Continental Scales with MODIS Data

Urbanization is one of the most important components of global environmental change, yet most of what we know about urban areas is at the local scale. Remote sensing of urban expansion across large areas provides information on the spatial and temporal patterns of growth that are essential for understanding differences in socioeconomic and political factors that spur different forms of development, as well the social, environmental, and climatic impacts that result. However, mapping urban expansion globally is challenging: urban areas have a small footprint compared to other land cover types, their features are small, they are heterogeneous in both material composition and configuration, and the form and rates of new development are often highly variable across locations. Here we demonstrate a methodology for monitoring urban land expansion at continental to global scales using Moderate Resolution Imaging Spectroradiometer (MODIS) data. The new method focuses on resolving the spectral and temporal ambiguities between urban/non-urban land and stable/changed areas by: (1) spatially constraining the study extent to known locations of urban land; (2) integrating multitemporal data from multiple satellite data sources to classify c. 2010 urban extent; and (3) mapping newly built areas (2000-2010) within the 2010 urban land extent using a multi-temporal composite change detection approach based on MODIS 250 m annual maximum enhanced vegetation index (EVI). We test the method in 15 countries in East-Southeast Asia experiencing different rates and manifestations of urban expansion. A two-tiered accuracy assessment shows that the approach characterizes urban change across a variety of socioeconomic/political and ecological/climatic conditions with good accuracy (70-91% overall accuracy by country, 69-89% by biome). The 250 m EVI data not only improve the classification results, but are capable of distinguishing between change and no-change areas in urban areas. Over 80% of the error in the change detection can be related to definitional issues or error propagation, rather than algorithm error. As such, these methods hold great potential for routine monitoring of urban change, as well as for providing a consistent and up-to-date dataset on urban extent and expansion for a rapidly evolving region.

Urban areas↗

Flood Extent Mapping for Namibia Using Change Detection and Thresholding with SAR

A new method for flood detection change detection and thresholding (CDAT) was used with synthetic aperture radar (SAR) imagery to delineate the extent of flooding for the Chobe floodplain in the Caprivi region of Namibia. This region experiences annual seasonal flooding and has seen a recent renewal of severe flooding after a long dry period in the 1990s. Flooding in this area has caused loss of life and livelihoods for the surrounding communities and has caught the attention of disaster relief agencies. There is a need for flood extent mapping techniques that can be used to process images quickly, providing near real-time flooding information to relief agencies. ENVISAT/ASAR and Radarsat-2 images were acquired for several flooding seasons from February 2008 to March 2013. The CDAT method was used to determine flooding from these images and includes the use of image subtraction, decision based classification with threshold values, and segmentation of SAR images. The total extent of flooding determined for 2009, 2011 and 2012 was about 542 km2, 720 km2, and 673 km2 respectively. Pixels determined to be flooded in vegetation were typically <0.5 % of the entire scene, with the exception of 2009 where the detection of flooding in vegetation was much greater (almost one third of the total flooded area). The time to maximum flooding for the 2013 flood season was determined to be about 27 days. Landsat water classification was used to compare the results from the new CDAT with SAR method; the results show good spatial agreement with Landsat scenes.

Flood↗

Impact of LANDSAT MSS Sensor Differences on Change Detection Analysis

Change detection techniques were used to pinpoint differences in the multispectral band scanners on LANDSAT 2, 3, and 4 satellites. The method of analysis was to co-register 512 by 512 pixel subwindows for all data pairs followed by scattergram generation and analysis. In all cases, the LANDSAT-4 data were used as the base to which other images were registered. There appear to be no major problems preventing use of LANDSAT-4 MSS with previous MSS sensors for charge detection, provided the interference noise can be removed or minimized. This noise may result in detection of spurious changes, as well as affect other uses of the data, including image classification. Analysis of dark (water and forests), rather than light features will be most impacted because the noise will form a higher percentage of the total response at low DN values. Any data normalizations for change detection should be based upon the data, rather than solely upon calibration information. While the observed relative radiometric transfer function between LANDSAT 3 and 4 was approximately as predicted, there were still significant deviations. Normalizing based upon data content also can have the advantage of allowing simultaneous normalization of the atmosphere as well as the radiometry.

Likens, W. C.↗

The use of multispectral difference data for urban change detection

The current work describes an experiment in which ERTS-1 digital multispectral scanner data from two times was used to detect change in an urban scene. The computer analysis implementation approach explored here was to register image data from the two times and to subtract the data to form multispectral difference images. The difference imagery was then analyzed using statistical pattern recognition to separate and classify different types of change. Construction was successfully detected.

Anuta, P. E.↗

Evaluation of change detection techniques for monitoring coastal zone environments

Procedures for detecting changes in Landsat multispectral scanning imagery of coastal zone environments are discussed. Four detection procedures are examined: a comparison of independently produced spectral classifications; a classification of a multispectral difference data set; a single analysis of a multidate data set; and a maximum likelihood classification using multistage decision logic. The relatively complex maximum likelihood classification technique was found to yield results closest to those obtained with the comparison of independently produced spectral classifications, the chosen standard.

Weismiller, R. A.↗

A New Method for High Resolution Surface Change Detection: Data Collection and Validation of Measurements from UAS at the Nevada National Security Site, Nevada, USA

The use of uncrewed aerial systems (UAS) increases the opportunities for detecting surface changes in remote areas and in challenging terrain. Detecting surface topographic changes offers an important constraint for understanding earthquake damage, groundwater depletion, effects of mining, and other events. For these purposes, changes on the order of 5–10 cm are readily detected, but sometimes it is necessary to detect smaller changes. An example is the surface changes that result from underground explosions, which can be as small as 3 cm. Previous studies that described change detection methodologies were generally not aimed at detecting sub-5-cm changes. Additionally, studies focused on high-fidelity accuracy were either computationally modeled or did not fully provide the necessary examples to highlight the usability of these workflows. Detecting changes at this threshold may be critical in certain applications, such as global security research and monitoring for high-consequence natural hazards, including landslides. Here we provide a detailed description of the methodology we used to detect 2–3 cm changes in an important applied research setting—surface changes related to underground explosions. This methodology improves the accuracy of change detection data collection and analysis through the optimization of pre-field planning, surveying, flight operations, and post-processing the collected data, all of which are critical to obtaining the highest output data resolution possible. We applied this methodology to a field study location, collecting 1.4 Tb of images over the course of 30 flights, and location data for 239 ground control points (GCPs). We independently verified changes with orthoimagery, and found that structure-from-motion, software-reported root mean square errors (RMSEs) for both control and check points underestimated the actual error. We found that 3 cm changes are detectable with this methodology, thereby improving our knowledge of a rock’s response to underground explosions.

47 OTHER INSTRUMENTATION↗

Image Change Detection via Ensemble Learning

The concept of geographic change detection is relevant in many areas. Changes in geography can reveal much information about a particular location. For example, analysis of changes in geography can identify regions of population growth, change in land use, and potential environmental disturbance. A common way to perform change detection is to use a simple method such as differencing to detect regions of change. Though these techniques are simple, often the application of these techniques is very limited. Recently, use of machine learning methods such as neural networks for change detection has been explored with great success. In this work, we explore the use of ensemble learning methodologies for detecting changes in bitemporal synthetic aperture radar (SAR) images. Ensemble learning uses a collection of weak machine learning classifiers to create a stronger classifier which has higher accuracy than the individual classifiers in the ensemble. The strength of the ensemble lies in the fact that the individual classifiers in the ensemble create a mixture of experts in which the final classification made by the ensemble classifier is calculated from the outputs of the individual classifiers. Our methodology leverages this aspect of ensemble learning by training collections of weak decision tree based classifiers to identify regions of change in SAR images collected of a region in the Staten Island, New York area during Hurricane Sandy. Preliminary studies show that the ensemble method has approximately 11.5% higher change detection accuracy than an individual classifier.

Martin, Benjamin W↗

Open-source Tip-and-Cue for Change-Detection Tasking

Currently, no tool exists to detect daily change across the globe, let alone to do so cost- and resource-efficiently. The proposed tool will use open-source data to detect change in near-real-time to identify targets of interest for tasking, providing a unique capability for constant global situational awareness.

97 MATHEMATICS AND COMPUTING↗

Parametric probability distributions for anomalous change detection

The problem of anomalous change detection arises when two (or possibly more) images are taken of the same scene, but at different times. The aim is to discount the 'pervasive differences' that occur thoughout the imagery, due to the inevitably different conditions under which the images were taken (caused, for instance, by differences in illumination, atmospheric conditions, sensor calibration, or misregistration), and to focus instead on the 'anomalous changes' that actually take place in the scene. In general, anomalous change detection algorithms attempt to model these normal or pervasive differences, based on data taken directly from the imagery, and then identify as anomalous those pixels for which the model does not hold. For many algorithms, these models are expressed in terms of probability distributions, and there is a class of such algorithms that assume the distributions are Gaussian. By considering a broader class of distributions, however, a new class of anomalous change detection algorithms can be developed. We consider several parametric families of such distributions, derive the associated change detection algorithms, and compare the performance with standard algorithms that are based on Gaussian distributions. We find that it is often possible to significantly outperform these standard algorithms, even using relatively simple non-Gaussian models.

97 MATHEMATICS AND COMPUTING↗

The impact of misregistration on change detection

The impact of images misregistration on the detection of changes in land cover was studied using spatially degraded Landsat MSS images. Emphasis is placed on simulated images of the Normalized Difference Vegetation Index (NDVI) at spatial resolutions of 250 and 500 m. It is pointed out that there is the need to achieve high values of registration accuracy. The evidence from simulations suggests that misregistrations can have a marked effect on the ability of remotely sensed data to detect changes in land cover. Even subpixel misregistrations can have a major impact, and the most marked proportional changes will tend to occur at the finest misregistrations.

Townshend, John R. G.↗

Unsupervised Change Detection for Space Habitats Using 3D Point Clouds

This work presents an algorithm for scene change detection from point clouds to enable autonomous robotic caretaking in future space habitats. Autonomous robotic systems will help maintain future deep-space habitats, such as the Gateway space station, which will be uncrewed for extended periods. Existing scene analysis software used on the International Space Station (ISS) relies on manually-labeled images for detecting changes. In contrast, the algorithm presented in this work uses raw, unlabeled point clouds as inputs. The algorithm first applies modified Expectation-Maximization Gaussian Mixture Model (GMM) clustering to two input point clouds. It then performs change detection by comparing the GMMs using the Earth Mover’s Distance. The algorithm is validated quantitatively and qualitatively using a test dataset collected by an Astrobee robot in the NASA Ames Granite Lab comprising single frame depth images taken directly by Astrobee and full-scene reconstructed maps built with RGB-D and pose data from Astrobee. The runtimes of the approach are also analyzed in depth. The source code is publicly released to promote further development.

robotics↗

Point pattern match-based change detection in a constellation of previously detected objects

A method and system is provided that applies attribute- and topology-based change detection to objects that were detected on previous scans of a medium. The attributes capture properties or characteristics of the previously detected objects, such as location, time of detection, detection strength, size, elongation, orientation, etc. The locations define a three-dimensional network topology forming a constellation of previously detected objects. The change detection system stores attributes of the previously detected objects in a constellation database. The change detection system detects changes by comparing the attributes and topological consistency of newly detected objects encountered during a new scan of the medium to previously detected objects in the constellation database. The change detection system may receive the attributes of the newly detected objects as the objects are detected by an object detection system in real time.

47 OTHER INSTRUMENTATION↗