Use Of Change Detection Research Articles

Innovative use of change detection in large numbers of satellite scenes, with geological applications

The large number of remote-sensing datasets available necessitates the development of efficient methods when assessing change between such data. A series of techniques, optimizing the analysis of change detection, specifically on large remote-sensing dataset collections, is demonstrated. Iterative (online) statistical measures for mean and standard deviation give the ability to gain a measure of change over potentially hundreds of datasets without excessive computing power being needed. From this, the coefficient of variation can be used to provide further insight. Using such measures, seasonal change can be detected on outcrop (as opposed to vegetation), illustrating that change detection can be used to further extend a spectral signature for rocks. Twelve Sentinel-2 scenes over a 3 year period were used in this study. Thematic collection: This article is part of the Remote sensing for site investigations on Earth and other planets collection available at: https://www.lyellcollection.org/cc/remote-sensing-for-site-investigations-on-earth-and-other-planets

AN ASSESSMENT OF SPATIAL VARIATION OF LAND SURFACE CHARACTERISTICS OF MINNA, NIGER STATE NIGERIA FOR SUSTAINABLE URBANIZATION USING GEOSPATIAL TECHNIQUES

AN ASSESSMENT OF SPATIAL VARIATION OF LAND SURFACE CHARACTERISTICS OF MINNA, NIGER STATE NIGERIA FOR SUSTAINABLE URBANIZATION USING GEOSPATIAL TECHNIQUES

Enhanced change detection index for disaster response, recovery assessment and monitoring of buildings and critical facilities—A case study for Muzzaffarabad, Pakistan

The availability of Very High Resolution (VHR) optical sensors and a growing image archive that is frequently updated, allows the use of change detection in post-disaster recovery and monitoring for robust and rapid results. The proposed semi-automated GIS object-based method uses readily available pre-disaster GIS data and adds existing knowledge into the processing to enhance change detection. It also allows targeting specific types of changes pertaining to similar man-made objects such as buildings and critical facilities. The change detection method is based on pre/post normalized index, gradient of intensity, texture and edge similarity filters within the object and a set of training data. More emphasis is put on the building edges to capture the structural damage in quantifying change after disaster. Once the change is quantified, based on training data, the method can be used automatically to detect change in order to observe recovery over time in potentially large areas. Analysis over time can also contribute to obtaining a full picture of the recovery and development after disaster, thereby giving managers a better understanding of productive management and recovery practices. The recovery and monitoring can be analyzed using the index in zones extending from to epicentre of disaster or administrative boundaries over time.

Enhanced change detection index for disaster response, recovery assessment and monitoring of accessibility and open spaces (camp sites)

The availability of Very High Resolution (VHR) optical sensors and a growing image archive that is frequently updated, allows the use of change detection in post-disaster recovery and monitoring for robust and rapid results. The proposed semi-automated GIS object-based method uses readily available pre-disaster GIS data and adds existing knowledge into the processing to enhance change detection. It also allows targeting specific types of changes pertaining to similar man-made objects. This change detection method is based on pre/post normalized index, gradient of intensity, texture and edge similarity filters within the object and a set of training data. Once the change is quantified, based on training data, the method can be used automatically to detect change in order to observe recovery over time in large areas. Analysis over time can also contribute to obtaining a full picture of the recovery and development after disaster, thereby giving managers a better understanding of productive management practices.

INTERACTIVE CHANGE DETECTION USING HIGH RESOLUTION REMOTE SENSING IMAGES BASED ON ACTIVE LEARNING WITH GAUSSIAN PROCESSES

Although there have been many studies for change detection, the effective and efficient use of high resolution remote sensing images is still a problem. Conventional supervised methods need lots of annotations to classify the land cover categories and detect their changes. Besides, the training set in supervised methods often has lots of redundant samples without any essential information. In this study, we present a method for interactive change detection using high resolution remote sensing images with active learning to overcome the shortages of existing remote sensing image change detection techniques. In our method, there is no annotation of actual land cover category at the beginning. First, we find a certain number of the most representative objects in unsupervised way. Then, we can detect the change areas from multi-temporal high resolution remote sensing images by active learning with Gaussian processes in an interactive way gradually until the detection results do not change notably. The artificial labelling can be reduced substantially, and a desirable detection result can be obtained in a few iterations. The experiments on Geo-Eye1 and WorldView2 remote sensing images demonstrate the effectiveness and efficiency of our proposed method.

INTERACTIVE CHANGE DETECTION USING HIGH RESOLUTION REMOTE SENSING IMAGES BASED ON ACTIVE LEARNING WITH GAUSSIAN PROCESSES

Abstract. Although there have been many studies for change detection, the effective and efficient use of high resolution remote sensing images is still a problem. Conventional supervised methods need lots of annotations to classify the land cover categories and detect their changes. Besides, the training set in supervised methods often has lots of redundant samples without any essential information. In this study, we present a method for interactive change detection using high resolution remote sensing images with active learning to overcome the shortages of existing remote sensing image change detection techniques. In our method, there is no annotation of actual land cover category at the beginning. First, we find a certain number of the most representative objects in unsupervised way. Then, we can detect the change areas from multi-temporal high resolution remote sensing images by active learning with Gaussian processes in an interactive way gradually until the detection results do not change notably. The artificial labelling can be reduced substantially, and a desirable detection result can be obtained in a few iterations. The experiments on Geo-Eye1 and WorldView2 remote sensing images demonstrate the effectiveness and efficiency of our proposed method.

A framework for moving object segmentation using dynamic background modelling and shadow suppression in complex wavelet domain

In this paper, a framework for dynamic background modelling and shadow suppression under rapidly changing illumination conditions for moving object segmentation in complex wavelet domain is proposed which deals with the problems of ghosts, object shadows, noise, object distortion in dynamic background changes. The proposed method consists of eight steps applied on given video frames which include: wavelet de-composition of frame using complex wavelet transform; use of change detection on detail coefficients; use of dynamic background modelling on approximate co-efficient; cast shadow suppression; use of soft thresholding for noise removal; strong edge detection; inverse wavelet transformation for reconstruction and finally using closing morphology operator. A comparative analysis of the proposed method is presented both qualitatively and quantitatively with other standard methods available in the literature for four datasets in terms of various performance measures. Experimental results demonstrate the efficacy of the proposed method is better in terms of relative foreground area measure, misclassification penalty, relative position based measure, normalised cross-correlation, normalised absolute error, peak signal-to-noise ratio and pixel classification based measure as compared to other standard methods.

A Novel Asymmetrical Probability Density Function for Modeling Log-Ratio SAR Images

Multitemporal synthetic aperture radar (SAR) images are often employed for change detection use due to the stable and periodic monitoring of the ground by SAR sensors. Because of the multiplicative nature of speckle noise in SAR data, the comparison between multitemporal images is often carried out according to a standard log-ratio operator. Therefore, in this letter, the probability density function (pdf) of the log-ratio SAR image is studied. Assuming that the SAR speckle is gamma distributed with unit mean, we present the analytic distribution form of the log-ratio image. The distribution is capable of capturing the nonsymmetrical feature of the histogram of the log-ratio data. At the same time, we provide a moment-based method for estimating the parameters of the pdf using empirical data. In the experiments, we use chi-square goodness-of-fit test and one-sample Kolmogorov–Smirnov test to evaluate the performance of the approach. The Gaussian distribution and the generalized Gaussian distribution are also employed for comparison. Experiment on two real SAR data sets demonstrates the effectiveness of the proposed method.

Framework for dynamic background modeling and shadow suppression for moving object segmentation in complex wavelet domain

Moving object segmentation using change detection in wavelet domain under continuous variations of lighting condition is a challenging problem in video surveillance systems. There are several methods proposed in the literature for change detection in wavelet domain for moving object segmentation having static backgrounds, but it has not been addressed effectively for dynamic background changes. The methods proposed in the literature suffer from various problems, such as ghostlike appearance, object shadows, and noise. To deal with these issues, a framework for dynamic background modeling and shadow suppression under rapidly changing illumination conditions for moving object segmentation in complex wavelet domain is proposed. The proposed method consists of eight steps applied on given video frames, which include wavelet decomposition of frame using complex wavelet transform; use of change detection on detail coefficients (LH, HL, and HH), use of improved Gaussian mixture-based dynamic background modeling on approximate coefficient (LL subband); cast shadow suppression; use of soft thresholding for noise removal; strong edge detection; inverse wavelet transformation for reconstruction; and finally using closing morphology operator. A comparative analysis of the proposed method is presented both qualitatively and quantitatively with other standard methods available in the literature for six datasets in terms of various performance measures. Experimental results demonstrate the efficacy of the proposed method.

The use of change detection as a method of objectively evaluating labels

AbstractThis study introduces the use of change detection, a technique used in cognitive psychology to measure attentional scan paths, as a way to objectively evaluate the prominence of varied label elements. There are two major objectives related to this work: (a) to develop change detection software and methodology for label use; and (b) to compare the relative prominence of different label elements on a beverage container. Six label elements (i.e. treatments) were analysed, namely: the manufacturer name, the product name and a warning dot with text in three colours. Study results suggest that experimental set‐up can significantly impact results, specifically the position of the change (p = 0.0078) and the order of appearance (p = 0.069). This is not only important from an experimental design perspective, but also could lead to insights regarding the attentive behaviours of people as they purchase, select and use products. With regard to the elements of the labels tested, we identified a significant difference on time to detect a change (p < 0.0001). Time required for the manufacturer's name, Asahi Breweries, was significantly longer than for any of the other label elements (p < 0.0001). Pairwise comparisons indicated that for the warning dot, red text was located marginally faster than the warning printed in black (p = 0.0566). Change detection offers the promise to objectively evaluate the relative prominence of a label (or a scene) but is quicker and cheaper than other methods, such as eye tracking, that are currently utilized for this purpose. Copyright © 2010 John Wiley & Sons, Ltd.

Something Overlooked? How experts in change detection use visual saliency

AbstractHow does expertise in the analysis of particular images influence the effects of visual saliency upon attention? Expert analysts of aerial photographs and untrained viewers undertook change‐detection and location memory tasks using aerial photographs with eye movements recorded throughout. Experts were more accurate in both tasks. Significant differences were also seen in the scanpaths: Untrained viewers fixated preferentially upon salient features throughout stimulus presentation whereas experts did not. However, both groups showed a strong influence of saliency in change detection and memory tasks. We interpret this apparent contradiction by: (i) assuming that the use of saliency in visual search is discretionary, and experts can use semantic information to prioritise where to fixate next; whereas, (ii) in tasks requiring spatial memory, analysis of visual saliency delivers easily acquired landmarks to reference the location of items in an image; a previously overlooked function used by expert and untrained viewers alike. Copyright © 2009 John Wiley & Sons, Ltd.

Interference from filled delays on visual change detection

Failure to detect changes to salient visual input across a brief interval has popularized the use of change detection, a paradigm that plays important roles in recent studies of visual perception, short-term memory, and consciousness. Much research has focused on the nature of visual representation for the pre- and postchange displays, yet little is known about how visual change detection is interfered with by events inserted between the pre- and postchange displays. To address this question, we tested change detection of colors, spatial locations, and natural scenes, when the interval between changes was (1) blank, (2) filled with a visual scene, or (3) filled with an auditory word. Participants were asked to either ignore the filled visual or auditory event or attend to it by categorizing it as animate or inanimate. Results showed that the ability to detect visual changes was dramatically impaired by attending to a secondary task during the delay. This interference was significant for auditory as well as for visual interfering events and was invariant to the complexity of the prechange displays. Passive listening produced no interference, whereas passive viewing produced small but significant interference. We conclude that visual change detection relies significantly on central, amodal attention.

A LANDSCAPE APPROACH FOR DETECTING AND EVALUATING CHANGE IN A SEMI-ARID ENVIRONMENT

Vegetation change in the American West has been a subject of concern throughout the twentieth century. Although many of the changes have been recorded qualitatively through the use of comparative photography and historical reports, little quantitative information has been available on the regional or watershed scale. It is currently possible to measure change over large areas and determine trends in ecological and hydrological condition using advanced space-based technologies. Specifically, this process is being tested in a community-based watershed in southeast Arizona and northeast Sonora, Mexico using a system of landscape pattern measurements derived from satellite remote sensing, spatial statistics, process modeling, and geographic information systems technology. These technologies provide the basis for developing landscape composition and pattern indicators as sensitive measures of large-scale environmental change and thus may provide an effective and economical method for evaluating watershed condition related to disturbance from human and natural stresses. The project utilizes the database from the North American Landscape Characterization (NALC) project which incorporates triplicate Landsat Multi-Spectral Scanner (MSS) imagery from the early 1970s, mid 1980s, and the 1990s. Landscape composition and pattern metrics have been generated from digital land cover maps derived from the NALC images and compared across a nearly 20-year period. Results about changes in land cover for the study period indicate that extensive, highly connected grassland and desertscrub areas are the most vulnerable ecosystems to fragmentation and actual loss due to encroachment of xerophytic mesquite woodland. In the study period, grasslands and desertscrub not only decreased in extent but also became more fragmented. That is, the number of grassland and desertscrub patches increased and their average patch sizes decreased. In stark contrast, the mesquite woodland patches increased in size, number, and connectivity. These changes have important impact for the hydrology of the region, since the energy and water balance characteristics for these cover types are significantly different. The process demonstrates a simple procedure to document changes and determine ecosystem vulnerabilities through the use of change detection and indicator development, especially in regard to traditional degradation processes that have occurred throughout the western rangelands involving changes of vegetative cover and acceleration of water and wind erosion.