Direct Change Detection Research Articles

Change Detection over mining areas using Deep Neural Networks from Planetscope image time series

The processes of change detection (CD) in land use and land cover through satellite images acquired in different temporal phases represent a key process to monitor the land, the environment and evaluate disasters. Currently the methodologies used for CD based on pixels do not allow processing a large amount of images massively, so it is not possible to analyze all the available information. Main aim of this research is to evaluate the performance of Deep learning methods for CD, grouped in two approaches, across open mining areas from PlanetScope (PS) image time series of the Cerrejon mine in Colombia. Two approaches for change generation are proposed, one based on post-classification comparison in which two convolutional neural network (CNN) architectures are evaluated: U-Net and Feature Pyramidal Network (FPN) for the classification of mining areas along the time series, for this purpose different models with different hyperparameters were generated and trained to select the most suitable for such process and subsequently perform the difference between the periods of the series; and a second approach based on direct change detection in which a modified U-Net network was evaluated from pairs of images. For this purpose, different models were also trained selecting the most appropriate for the detection of changes for each period of the time series, obtaining a map of changes for each approach, each one the results were validated; the most appropriate approach was number 2 (Direct change detection), with kappa accuracies greater than 0.9 in each period of the time series.

Operational earthquake-induced building damage assessment using CNN-based direct remote sensing change detection on superpixel level

• Superpixel-based direct change detection is capable of finding damages efficiently. • Extra feature enhancement bands accelerate the process of damage assessment. • Channel-expanded CNN model benefits simultaneous building damage analysis. • The potential of the proposed method in post-earthquake emergency is recognized. Accurate and quick building damage assessment is an indispensable step after a destructive earthquake. Acquiring building damage information of the seismic area in a remotely sensed way enables a timely emergency response. Existing remote sensing building damage detection methods based on convolutional neural network (CNN) mainly need two-step processing or only use single post-event image, leading to low efficiency and inaccurate building boundary. Considering the practical needs of emergency rescue and post-disaster reconstruction, this study proposed a hierarchical building damage assessment workflow using CNN-based direct remote sensing change detection on superpixel level. First, vulnerable building areas close to the epicenter are extracted using extra feature enhancement bands (EFEBs) to narrow the extent of image processing. Then, fine scale building damage is detected in the extracted building areas based on a direct change detection method with pre-event superpixel constraint (PreSC) strategy to improve the precision and efficiency. Finally, a rapid remote sensing earthquake damage index (rRSEDI) is used to quantitatively assess the damage. Experimental results of the case study show that damaged buildings can be effectively and accurately localized and classified using the proposed workflow. Comparative experiments with single-temporal image and post-event segmentation further embody the superiority of the direct change detection. The damage assessment result matches the official report after Ludian earthquake, proving the reliability of the proposed workflow. For future natural hazard events, the workflow can contribute to formulating appropriate disaster management, prevention and mitigation policies.

Enhancing the sweat resistance of sunscreens.

While sunbathing of performing outdoor sport activities, sunscreens are important for protection of uncovered skin against ultraviolet (UV) radiation. However, perspiration negatively affects the performance of a sunscreen film by weakening its substantivity and uniformity through the activation of two mechanisms, namely sunscreen wash-off and sunscreen redistribution. We used a perspiring skin simulator to investigate the effect of sunscreen formulation on its efficiency upon sweating. Specifically, we modified the sunscreen formulation by incorporating a hydrophobic film former and adding water-absorbing particles. Sunscreen performance before and after perspiration is assessed by in vitro sun protection factor measurements, direct detection of changes in the sunscreen distribution using UV reflectance imaging, and by coherent anti-Stokes Raman scattering (CARS) microscopy for microscopic characterization of the UV filter relocation. The results show that incorporating a hydrophobic film former can decrease sunscreen wash-off due to sweating, while an excessive amount of film former might negatively affect the sunscreen distribution. The addition of water-absorbing particles, on the other hand, had either a negative or positive impact on the sunscreen substantivity, depending on the particle properties. While the addition of large water-absorbing particles appeared to increase sunscreen redistribution, smaller particles that could form a gel-like structure upon contact with water, appeared to change sunscreen wetting and sweat droplet spreading, thereby decreasing sunscreen wash-off and sunscreen redistribution. We find that using a combination of hydrophobic film formers, which increase water resistance, and small water-absorbing particles, which change the wetting behavior, can make sunscreen formulations more sweat-resistant and less runny.

Fine-Grained Building Change Detection From Very High-Spatial-Resolution Remote Sensing Images Based on Deep Multitask Learning

Building change detection from very high-spatial-resolution (VHR) remote sensing images has gained increasing popularity in a variety of applications, such as urban planning and damage assessment. Detecting fine-grained “from–to” changes (change transition from one land cover type to another) of buildings from the VHR images is still challenging as multitemporal representation is complicated. Recently, fully convolutional neural networks (FCNs) have been proven to be capable of feature extraction and semantic segmentation of VHR images, but its ability in change detection is untested and unknown. In this letter, we leverage the semantic segmentation of buildings as an auxiliary source of information for the fine-grained “from–to” change detection. A deep multitask learning framework for change detection (MTL-CD) is proposed for detecting building changes from the VHR images. MTL-CD adopts the encoder–decoder architecture and solves the main task of change detection and the auxiliary tasks of semantic segmentation simultaneously. Accordingly, the change detection loss function is constrained by the auxiliary semantic segmentation tasks and enables the back-propagation of the building footprints’ detection errors for the improvement of change detection. A building change detection data set named the Guangzhou data set is also developed for model evaluation, in which the bitemporal R–G–B images were collected by airplane (2009) and unmanned aerial vehicle (UAV, 2019) with different flight heights. Experiments on the Guangzhou data set demonstrate that the MTL-CD method effectively detects fine-grained “from–to” changes and outperforms the postclassification methods and the direct change detection methods.

Integrating cross-sensor high spatial resolution satellite images to detect subtle forest vegetation change in the Purple Mountains, a national scenic spot in Nanjing, China

Accurate information on the location and magnitude of vegetation change in scenic areas can guide the configuration of tourism facilities and the formulation of vegetation protection measures. High spatial resolution remote sensing images can be used to detect subtle vegetation changes. The major objective of this study was to map and quantify forest vegetation changes in a national scenic location, the Purple Mountains of Nanjing, China, using multi-temporal cross-sensor high spatial resolution satellite images to identify the main drivers of the vegetation changes and provide a reference for sustainable management. We used Quickbird images acquired in 2004, IKONOS images acquired in 2009, and WorldView2 images acquired in 2015. Four pixel-based direct change detection methods including the normalized difference vegetation index difference method, multi-index integrated change analysis (MIICA), principal component analysis, and spectral gradient difference analysis were compared in terms of their change detection performances. Subsequently, the best pixel-based detection method in conjunction with object-oriented image analysis was used to extract subtle forest vegetation changes. An accuracy assessment using the stratified random sampling points was conducted to evaluate the performance of the change detection results. The results showed that the MIICA method was the best pixel-based change detection method. And the object-oriented MIICA with an overall accuracy of 0.907 and a kappa coefficient of 0.846 was superior to the pixel-based MIICA. From 2004 to 2009, areas of vegetation gain mainly occurred around the periphery of the study area, while areas of vegetation loss were observed in the interior and along the boundary of the study area due to construction activities, which contributed to 79% of the total area of vegetation loss. During 2009–2015, the greening initiatives around the construction areas increased the forest vegetation coverage, accounting for 84% of the total area of vegetation gain. In spite of this, vegetation loss occurred in the interior of the Purple Mountains due to infrastructure development that caused conversion from vegetation to impervious areas. We recommend that: (1) a local multi-agency team inspect and assess law enforcement regarding natural resource utilization; and (2) strengthen environmental awareness education.

MONITORING CORAL GROWTH – THE DICHOTOMY BETWEEN UNDERWATER PHOTOGRAMMETRY AND GEODETIC CONTROL NETWORK

Abstract. Creating 3-dimensional (3D) models of underwater scenes has become a common approach for monitoring coral reef changes and its structural complexity. Also in underwater archeology, 3D models are often created using underwater optical imagery. In this paper, we focus on the aspect of detecting small changes in the coral reef using a multi-temporal photogrammetric modelling approach, which requires a high quality control network. We show that the quality of a good geodetic network limits the direct change detection, i.e., without any further registration process. As the photogrammetric accuracy is expected to exceed the geodetic network accuracy by at least one order of magnitude, we suggest to do a fine registration based on a number of signalized points. This work is part of the Moorea Island Digital Ecosystem Avatar (IDEA) project that has been initiated in 2013 by a group of international researchers (https://mooreaidea.ethz.ch/).

In VitroDiagnostic Device with a Bio-Coupled Gate Field Effect Transistor

In this study, we report a simple and rapid biosensing method for the analysis andin vitromonitoring of biological processes, including DNAbinding events, antigen-antibody interactions, and cellular functions, using a semiconductor device. Most biological phenomena involve cell-cell communication processes that are mediated by the transport of sodium or potassium ions and other charged biomolecules, such as DNA, across ion channels in the cell membrane. Therefore, our approach focused on the direct detection of changes in ion concentrations by utilizing a semiconductor-based biosensor device. Our results demonstrated that our semiconductor-based biosensor platform achieves label-free and noninvasive biosensing that is suitable forin vitrodiagnosis.

Label-Free Ratiometric Imaging of Serotonin in Live Cells.

Ratiometric imaging can quantitatively measure changes in cellular analyte concentrations using specially designed fluorescent labels. We describe a label-free ratiometric imaging technique for direct detection of changes in intravesicular serotonin concentration in live cells. At higher concentrations, serotonin forms transient oligomers whose ultraviolet emission is shifted to longer wavelengths. We access the ultraviolet/blue emission using relatively benign three-photon excitation and split it into two imaging channels, whose ratio reports the concentration. The technique is sensitive at a physiologically relevant concentration range (10-150 mM serotonin). As a proof of principle, we measure the increase of intravesicular serotonin concentration with the addition of external serotonin. In general, since emission spectra of molecules are often sensitive to concentration, our method may be applicable to other natively fluorescent intracellular molecules which are present at high concentrations.

Land cover change analysis in Mexico using 30m Landsat and 250m MODIS data

Abstract. Change detection is one of the most important and widely requested applications of terrestrial remote sensing. Despite a wealth of techniques and successful studies, there is still a need for research in remote sensing science. This paper addresses two important issues: the temporal and spatial scales of change maps. Temporal scales relate to the time interval between observations for successful change detection. We compare annual change detection maps accumulated over five years against direct change detection over that period. Spatial scales relate to the spatial resolution of remote sensing products. We compare fractions from 30m Landsat change maps to 250m grid cells that match MODIS change products. Results suggest that change detection at annual scales better detect abrupt changes, in particular those that do not persist over a longer period. The analysis across spatial scales strongly recommends the use of an appropriate analysis technique, such as change fractions from fine spatial resolution data for comparison with coarse spatial resolution maps. Plotting those results in bi-dimensional error space and analyzing various criteria, the “lowest cost”, according to a user defined (here hyperbolic) cost function, was found most useful. In general, we found a poor match between Landsat and MODIS-based change maps which, besides obvious differences in the capabilities to detect change, is likely related to change detection errors in both data sets.

Determining changes in the distribution and abundance of a Rhyzopertha dominica phosphine resistance allele in farm grain storages using a DNA marker

The lesser grain borer, Rhyzopertha dominica (F.), is a highly destructive pest of stored grain that is strongly resistant to the fumigant phosphine (PH3 ). Phosphine resistance is due to genetic variants at the rph2 locus that alter the function of the dihydrolipoamide dehydrogenase (DLD) gene. This discovery now enables direct detection of resistance variants at the rph2 locus in field populations. A genotype assay was developed for direct detection of changes in distribution and frequency of a phosphine resistance allele in field populations of R. dominica. Beetles were collected from ten farms in south-east Queensland in 2006 and resampled in 2011. Resistance allele frequency increased in the period from 2006 to 2011 on organic farms with no history of phosphine use, implying that migration of phosphine-resistant R. dominica had occurred from nearby storages. Increasing resistance allele frequencies on organic farms suggest local movement of beetles and dispersal of insects from areas where phosphine has been used. This research also highlighted for the first time the utility of a genetic DNA marker in accurate and rapid determination of the distribution of phosphine-resistant insects in the grain value chain. Extending this research over larger landscapes would help in identifying resistance problems and enable timely pest management decisions.

Comparison of Remote Sensing Image Processing Techniques to Identify Tornado Damage Areas from Landsat TM Data.

Remote sensing techniques have been shown effective for large-scale damagesurveys after a hazardous event in both near real-time or post-event analyses. The paperaims to compare accuracy of common imaging processing techniques to detect tornadodamage tracks from Landsat TM data. We employed the direct change detection approachusing two sets of images acquired before and after the tornado event to produce a principalcomponent composite images and a set of image difference bands. Techniques in thecomparison include supervised classification, unsupervised classification, and object-oriented classification approach with a nearest neighbor classifier. Accuracy assessment isbased on Kappa coefficient calculated from error matrices which cross tabulate correctlyidentified cells on the TM image and commission and omission errors in the result. Overall,the Object-oriented Approach exhibits the highest degree of accuracy in tornado damagedetection. PCA and Image Differencing methods show comparable outcomes. Whileselected PCs can improve detection accuracy 5 to 10%, the Object-oriented Approachperforms significantly better with 15-20% higher accuracy than the other two techniques.

Monitoring of conformational change in maltose binding protein using split green fluorescent protein

In this study, we describe a novel method for the detection of conformational changes in proteins, which is predicated on the reconstitution of split green fluorescent protein (GFP). We employed fluorescence complementation assays for the monitoring of the conformationally altered proteins. In particular, we used maltose binding protein (MBP) as a model protein, as MBP undergoes a characteristic hinge-twist movement upon substrate binding. The common feature of this approach is that GFP, as a reporter protein, splits into two non-fluorescent fragments, which are genetically fused to the N- and C-termini of MBP. Upon binding to maltose, the chromophores move closer together, resulting in the generation of fluorescence. This split GFP method also involves the reconstitution of GFP, which is determined via observations of the degree to which fluorescence intensity is restored. As a result, reconstituted GFP has been observed to generate fluorescence upon maltose binding in vitro, thereby allowing for the direct detection of changes in fluorescence intensity in response to maltose, in a concentration- and time-dependent fashion. Our findings showed that the fluorescence complementation assay can be used to monitor the conformational alterations of a target protein, and this ability may prove useful in a number of scientific and medical applications.

Peroxidation of linoleate at physiological pH: hemichrome formation by substrate binding protects against metmyoglobin activation by hydrogen peroxide

Peroxidation by metmyoglobin, MbFe(III), by metmyoglobin/hydrogen peroxide, MbFe(III)/H 2O 2 , to yield the myoglobin ferryl radical ( MbFe(IV)=O), or by ferrylmyoglobin, MbFe(IV)=O, was investigated at physiological pH (7.4) in oil-in-water linoleate emulsions. Linoleate peroxidation was followed using second derivative ultraviolet (UV)-spectroscopy for monitoring formation of conjugated dienes and quantitative determination of specific linoleate hydroperoxides by liquid chromatography with photodiode absorption detection. Modifications of myoglobins during lipid peroxidation were followed simultaneously by changes in the Soret absorption band (410 or 424 nm), and in the visible absorption region (from 450 to 700 nm), combined with electron spin resonance (ESR) spectroscopy for direct detection of changes in the spin state of the iron center. In contrast to MbFe(IV)=O, MbFe(III) and MbFe(III)/H 2O 2 were not able to initiate linoleate peroxidation in oil-in-water emulsions, and MbFe(III) was converted, by binding of linoleate (but not methyl linoleate), to a low-spin hemichrome derivate, HMbFe(III), with the distal histidine reversibly bound to the iron center. HMbFe(III) is ineffective in initiating lipid peroxidation and cannot be activated to MbFe(IV)=O or MbFe(IV)=O by addition of moderate amounts of H 2O 2. Addition of MbFe(III) to linoleate emulsions containing H 2O 2 results in the competitive formation of MbFe(IV)=O and HMbFe(III) in favor of HMbFe(III), and little linoleate peroxidation is detected, demonstrating the inherent protection, at physiologic pH, against peroxidation by reversible binding of the substrate to the potential myoglobin catalyst.

Monophasic action potentials, electrocardiograms and mechanical performance in normal and ischaemic epicardial segments of the pig ventricle in situ.

Few studies report simultaneous electrical and mechanical recordings from the epicardium of intact beating hearts in situ during ischaemia. We use suction to apply transducers and electrodes to areas of the epicardium. This interferes little with its behaviour and allows: i) free mobility over the surface; ii) simultaneous tridirectional length changes to be recorded and summed for an overall impression of mechanical behaviour, iii) detection of changes in direction of movement; iv) simultaneous recordings of monophasic action potential and epicardial ECG. During ischaemia we can detect impaired contraction with dyskinesis and a change in direction of epicardial forces. The action potential duration shortens and we have noted impaired conduction, inexitability and recordings consistent with re-entry leading to ventricular fibrillation.