Difference Image Research Articles

Winter wheat planting area extraction using SAR change detection

ABSTRACT With its all-day and all-weather imaging capability, Synthetic Aperture Radar (SAR) has shown great potential in crop identification. This letter puts forward a novel method of extracting winter wheat planting area by means of SAR change detection. Based on the phenological differences between winter wheat and other crops, the Difference Image (DI) is generated firstly by the different temporal SAR images. Then, Object Markov Random Field (OMRF) model is used to DI pre-classification. Finally, Convolutional Neural Network (CNN) is utilized to obtain the better area extraction result of winter wheat. The area extraction accuracy is proved by the field survey data, with an accuracy of 90.53%. This research can provide a new idea and method for agricultural remote sensing monitoring.

SAR image change detection based on sparse representation and a capsule network

ABSTRACT Synthetic aperture radar (SAR) image change detection has good application prospects in the domain of remote sensing image processing. However, the accuracy of change detection is negatively affected by the inherent speckle noise in SAR images. To solve this problem, a method based on sparse representation (SR) and a capsule network is proposed. Firstly, sparse features of the difference image (DI) are extracted by the SR method. Secondly, a lightweight capsule network (L-CapsNet) is constructed, which is used to mine the spatial relationship between features. The network classifies the changed and unchanged pixels. Finally, the change map (CM) is generated. The proposed method can obtain more robust features while reducing the influence of speckle noise. Experiments are performed on four SAR data sets to compare the proposed method with related methods. The results confirm the superior performance of the proposed method.

Seeking a Widely Adoptable Practical Standard to Estimate Signal-to-Noise Ratio in Magnetic Resonance Imaging for Multiple-Coil Reconstructions.

Signal-to-noise ratio (SNR) is used to evaluate the performance of magnetic resonance (MR) imaging systems. Accurate and consistent estimations are needed to routinely use SNR to assess coils and image reconstruction techniques. To identify a reliable and practical method for SNR estimation in multiple-coil reconstructions. Technical evaluation and comparison. A uniform phantom and four healthy volunteers: 35, 38, 39 y/o males, 25 y/o female. Two-dimensional multislice gradient-echo pulse sequence at 3 T and 7 T. Reference-standard SNR was calculated from 100 multiple replicas. Six SNR methods were compared against it: difference image (DI), analytic array combination (AC), pseudo-multiple-replica (PMR), generalized pseudo-replica (GPR), smoothed image subtraction (SIS), and DI with temporal instability correction (TIC). The assessment was repeated for different multiple-coil reconstructions. SNR methods were evaluated in terms of relative deviation (RD) and normalized mutual information (NMI) with respect to the reference-standard, using a linear regression (0.05 significance level) to assess how different factors affect accuracy. Average RD (phantom) for DI, AC, PMR, GPR, SIS, and TIC was 7.9%, 6%, 6.7%, 10.1%, 40%, and 14.6%, respectively. RD increased with acceleration. SNR maps with AC were the most similar to the reference standard (NMI=0.358). Considering all brain regions of interest, average RD for all SNR methods varied 96% among volunteers but remained approximately 10% for AC, PMR, and GPR, whereas it was more than 30% for DI, SIS, and TIC. RD was mainly affected by image reconstruction (beta=12) for AC and SNR entropy for SIS (beta=19). AC provided accurate and robust SNR estimation. PMR and GPR are more generally applicable than AC. DI and TIC should be used only at low acceleration factors, when an additional noise-only scan cannot be acquired. SIS is a single-acquisition alternative to DI for generalized autocalibrating partial parallel acquisition (GRAPPA) reconstructions. 1 TECHNICAL EFFICACY: Stage 1.

ОЦЕНКА ПОГРЕШНОСТИ ПРИ МНОЖЕСТВЕННОМ СРАВНЕНИИ ЗАШУМЛЕННЫХ ИЗОБРАЖЕНИЙ

The aim of this work is to study the effect of noise on the image on the quality of comparison of afinite set of images of the same shape and size. This task inevitably arises when analyzing scenes, detectingindividual objects, detecting symmetry, etc. The noise factor must be taken into account, since thedifference between digital objects can be caused not only by the mismatch of the compared images ofreal objects, but also by distortions due to noise, which is almost always takes place. This differenceturns out to be proportional to the level of the noise component. The main result of this article is ananalytical estimate for the probability of a given level of error, which may arise in the multiple comparisonof a finite set of commensurate digital images. This estimate is based on a low-level comparison,which is a pixel-by-pixel calculation of image differences using the Euclidean metric. In this case, astandard assumption is made about the independent normal noise of image intensities with zero mathematicalexpectation and a priori established standard deviation in each pixel. The evidence presented inthe article allows us to assert that the obtained estimate should be regarded as sufficiently "cautious"and it can be expected that in reality the scatter of the measure caused by noise in the image will besignificantly less than the theoretically found boundary. The estimates obtained in this work are alsouseful for detecting various types of symmetry in images, which, as a rule, lead to the need to calculatethe difference of an arbitrary number of commensurate digital areas. In addition, they can be used astheoretically grounded threshold values in tasks requiring a decision on the coincidence or difference ofimages. Such threshold values inevitably appear at various stages of processing noisy images, and thequestion of their specific values, as a rule, remains open; at best, heuristic considerations are proposedfor their selection.

Spectral Analysis of Sandy Desertification in the Semi-Arid Zone of North Eastern Nigeria

The method of integrating remote sensing, Geographic Information System (GIS) and field survey was employed. Assessment of the rate and intensity of sand dune encroachment using multi-temporal Landsat images (Landsat.TM, 1986, Landsat.ETM, 2000 and Landsat.OLI, TIR, 2016) and GIS. The satellite images were processed by converting raw Digital Number (DNs) values to radiance images which were converted into reflectance images used for spectral analysis.The satellite images were processed accordingly for evaluating six (6) spectral indices; Crust Index (CI, Grain Size Index, Bare soil Index (BSI), Normalized Difference Sand Dune Index (NDSDI), Normalized Difference Sand Index (NDSI), Normalised Difference Soil Index (NDSLI). An aggregate index of each of the six (6) selected indices was evaluated and the long term geometric mean was determined and used for image differencing with the baseline date image. A combination of MEDALUS.ESA and Image Differencing was adopted for change detection technique. Sandy landscapes were mapped into four (4) natural classes using natural jenks classifier of the ArcGIS analytical tool based on pre-field field determination and post verification. The description of the four (4) sandy landscape classes is as follows: Active, Semi-active, Semi-fixed and Fixed sand dune/sheets. Results of overall sandy desertification based on Aggregate Sandification Index indicates that active and semi-active sandy landscapes have progressed steadily at annual rate of expansion of 1.20 and 1.28 km 2 and intensity for the of 0.13 and 0.23% respectively. This has caused a corresponding decrease in the semi-fixed and fixed sandy landscapes of 1.24 and 1.39 km 2 and intensity for the period of 0.17 and 0.47 % respectively. The highest risk of sandy desertification is in the fixed sandy landscape which is be lost an an annual rate of 1.39 km 2 and 0.47% intensity being the highest among other classes. The result of this study indicates that the natural ecology or vegetation, graze lands, irrigated lands, rainfed farmlands, settlements, infrastructure are at high risk of sandy desertification in the semi-arid zone of Nigeria. This study is also a pointer that the shelter belts have not been very effective in controlling wind erosion and thus sandy desertification. Keywords: : desertification. sandification, eco-geomorphic landscape, semi-arid, encroachment DOI: 10.7176/JNSR/12-14-06 Publication date: July 31 st 2021

A Comparison of LUCC Detection Algorithms in a Mesoamerican Lowland Tropical Forest

Land use and land cover changes occur throughout the world, but none is more concerning than tropical deforestation, much of it for agricultural purposes. Rural-rural frontier migrant farmers such as those colonizing the Sierra del Lacandón National Park in Petén, Guatemala act as a primary direct agent in this land cover conversion. This paper seeks to compare three different algorithms for monitoring changes in forested land cover, making use of freely available remotely sensed Landsat images from two years, 1991 and 2000. In the intervening 9 years, some forested land was converted to cropped, pasture, or fallow land, while other areas experienced no change. This paper contains a detailed description of the methods employed for three different change detection techniques, producing a total of five land change maps: multidate principal components analysis (PCA), normalized difference vegetation index (NDVI ) image differencing, and brightness greenness wetness (BGW) image differencing. Of the five land change maps produced, the Greenness component of the BGW transformation had the highest overall accuracy, at 86%, and is conservative in detecting change. The amount of change detected by this algorithm represents approximately 300 km2 of forest loss, or 11.9% of the area examined.

Unsupervised Change Detection from Remotely Sensed Images Based on Multi-Scale Visual Saliency Coarse-to-Fine Fusion

Unsupervised change detection(CD) from remotely sensed images is a fundamental challenge when the ground truth for supervised learning is not easily available. Inspired by the visual attention mechanism and multi-level sensation capacity of human vision, we proposed a novel multi-scale analysis framework based on multi-scale visual saliency coarse-to-fine fusion (MVSF) for unsupervised CD in this paper. As a preface of MVSF, we generalized the connotations of scale as four classes in the field of remote sensing (RS) covering the RS process from imaging to image processing, including intrinsic scale, observation scale, analysis scale and modeling scale. In MVSF, superpixels were considered as the primitives for analysing the difference image(DI) obtained by the change vector analysis method. Then, multi-scale saliency maps at the superpixel level were generated according to the global contrast of each superpixel. Finally, a weighted fusion strategy was designed to incorporate multi-scale saliency at a pixel level. The fusion weight for the pixel at each scale is adaptively obtained by considering the heterogeneity of the superpixel it belongs to and the spectral distance between the pixel and the superpixel. The experimental study was conducted on three bi-temporal remotely sensed image pairs, and the effectiveness of the proposed MVSF was verified qualitatively and quantitatively. The results suggest that it is not entirely true that finer scale brings better CD result, and fusing multi-scale superpixel based saliency at a pixel level obtained a higher F1 score in the three experiments. MVSF is capable of maintaining the detailed changed areas while resisting image noise in the final change map. Analysis of the scale factors in MVSF implied that the performance of MVSF is not sensitive to the manually selected scales in the MVSF framework.

Robust unsupervised small area change detection from SAR imagery using deep learning

Small area change detection using synthetic aperture radar (SAR) imagery is a highly challenging task, due to speckle noise and imbalance between classes (changed and unchanged). In this paper, a robust unsupervised approach is proposed for small area change detection using deep learning techniques. First, a multi-scale superpixel reconstruction method is developed to generate a difference image (DI), which can suppress the speckle noise effectively and enhance edges by exploiting local, spatially homogeneous information. Second, a two-stage centre-constrained fuzzy c-means clustering algorithm is proposed to divide the pixels of the DI into changed, unchanged and intermediate classes with a parallel clustering strategy. Image patches belonging to the first two classes are then constructed as pseudo-label training samples, and image patches of the intermediate class are treated as testing samples. Finally, a convolutional wavelet neural network (CWNN) is designed and trained to classify testing samples into changed or unchanged classes, coupled with a deep convolutional generative adversarial network (DCGAN) to increase the number of changed class within the pseudo-label training samples. Numerical experiments on four real SAR datasets demonstrate the validity and robustness of the proposed approach, achieving up to 99.61% accuracy for small area change detection.

LCS-EnsemNet: A Semisupervised Deep Neural Network for SAR Image Change Detection With Dual Feature Extraction and Label-Consistent Self-Ensemble

Change detection (CD) in synthetic aperture radar (SAR) images faces two challenging problems: 1) inherent speckle noise in SAR data causes the class confusion and affects the image understanding for inferring category of each image pixel; and 2) adequate labeled samples are quite hard to collect, which is the major limitation for supervised methods. In this article, we develop a novel deep learning-based semisupervised method to address these challenges. The method firstly incorporates a pixel-wise log-ratio difference image (DI) and its saliency map to produce a spatially enhanced (SE) DI using a reweighting scheme based on the fact that changed pixels exhibit higher saliency than unchanged pixels. As a result, prominent changed regions are highlighted and the interclass separability is increased. We construct pixel-wise and context-wise features based on the log-ratio DI and SE DI to jointly express the dissimilarity at each pixel. Second, a two-branch deep neural network taking the dual features as input is devised to reveal and refine underlying label information from abundant unlabeled samples with the help of a two-stage refinement strategy for improving generalization ability, called LCS-EnsemNet. The refinement strategy, namely label-consistent self-ensemble, exploits label consistency between dual features and across multiple classifiers to ensure the reliability of pseudo labels. Finally, the cross-entropy loss is calculated with the limited labeled data and selected pseudo labeled samples to optimize the LCS-EnsemNet. The proposed method is evaluated on three low/medium- resolution and one high-resolution SAR data sets, and experimental results have demonstrated its effectiveness.

Building change detection from multi-source remote sensing images based on multi-feature fusion and extreme learning machine

ABSTRACT In this paper, in order to improve the accuracy of multi-source remote sensing image building change detection, we propose a method based on multi-feature fusion (MFF) and extreme learning machine (ELM) training. Firstly, we fuse the spectral feature difference image (DI) and textural feature (grey level co-occurrence matrix) DI obtained by change vector analysis (CVA), morphological building index DI, and shape feature DI obtained by subtraction to construct the final DI. Secondly, the coarse change detection map obtained by selecting a threshold for the DI saliency map obtained by the use of the frequency-domain significance (FDS) method is pre-classified by the fuzzy c-means (FCM) clustering algorithm. Finally, the neighborhood features obtained from the original images and the feature images of the changed pixels (buildings) and the unchanged pixels in the coarse change map are extracted and used as reliable samples for the ELM training. By using the trained ELM classifier, undetermined pixels are further separated into changed and unchanged classes. Finally, we combine the ELM classification result and the preclassification result together to form the final building change map. Experiments on two real multi-source datasets show that the proposed method can result in a significant improvement in multi-source remote sensing image building change detection performance.

Combining SAR and Optical Earth Observation with Hydraulic Simulation for Flood Mapping and Impact Assessment

Timely mapping, measuring and impact assessment of flood events are crucial for the coordination of flood relief efforts and the elaboration of flood management and risk mitigation plans. However, this task is often challenging and time consuming with traditional land-based techniques. In this study, Sentinel-1 radar and Landsat images were utilized in collaboration with hydraulic modelling to obtain flood characteristics and land use/cover (LULC), and to assess flood impact in agricultural areas. Furthermore, indirect estimation of the recurrence interval of a flood event in a poorly gauged catchment was attempted by combining remote sensing (RS) and hydraulic modelling. To this end, a major flood event that occurred in Sperchios river catchment, in Central Greece, which is characterized by extensive farming activity was used as a case study. The synergistic usage of multitemporal RS products and hydraulic modelling has allowed the estimation of flood characteristics, such as extent, inundation depth, peak discharge, recurrence interval and inundation duration, providing valuable information for flood impact estimation and the future examination of flood hazard in poorly gauged basins. The capabilities of the ESA Sentinel-1 mission, which provides improved spatial and temporal analysis, allowing thus the mapping of the extent and temporal dynamics of flood events more accurately and independently from the weather conditions, were also highlighted. Both radar and optical data processing methods, i.e., thresholding, image differencing and water index calculation, provided similar and satisfactory results. Conclusively, multitemporal RS data and hydraulic modelling, with the selected techniques, can provide timely and useful flood observations during and right after flood disasters, applicable in a large part of the world where instrumental hydrological data are scarce and when an apace survey of the condition and information about temporal dynamics in the influenced region is crucial. However, future missions that will reduce further revisiting times will be valuable in this endeavor.

Change detection in SAR images based on superpixel segmentation and image regression

Change detection (CD) is one of the most important application in remote sensing domain. The difference image (DI) generated by traditional change detection methods are sensitive to several factors, such as atmospheric condition changes, illumination variations, sensor calibration, and speckle noise, greatly affecting the detection performance. To avoid the aforementioned problem, in this paper, a novel approach based on superpixel segmentation and image regression is proposed to detect changes between bitemporal synthetic aperture radar (SAR) images. Specifically, the bitemporal images are firstly divided into a number of superpixel pairs under the guidance of segmentation result of a pre-DI. Next, each pixel in pre-event image is reconstructed utilizing its nearest neighbor to reduce the influence of noise. Then, a set of preselected unchanged sample are selected to learn the local regression model and to estimate the post-event image. After that, the final DI can be obtained by measuring the difference between estimated post-event image and the actual one. Finally, the fuzzy c-means (FCM) clustering algorithm is adopted to generate the binary change map. Adequate experiments on four SAR datasets have been tested, and the experimental results compared with the state-of-the-art methods have proved the superiority of the proposed method.

Similarity Matrix Entropy for Multitemporal Polarimetric SAR Change Detection

For cumulative change detection using time-series polarimetric synthetic aperture radar (SAR) (PolSAR) data, pairwise comparisons using bi-temporal data change detection are often performed over each successive pair of images in the entire time series. In this way, due to the impact of the quality of the difference image (DI) of bitemporal data and the classification accuracy of the DI into the change and nonchange classes, there is a great risk of many false alarms or omissions caused by gradually occurred changes. Therefore, this letter proposes a symmetric similarity matrix based on a likelihood ratio test (LRT), which groups the similarity between any two-time point data in time series. Then, Shannon entropy is applied to the similarity matrix to calculate the DI of the entire time-series data. Finally, an image segmentation method based on Markov random field (MRF) is used to classify the calculated DI into the change and nonchange classes. The experiments using the multitemporal fully polarimetric Radarsat-2 data acquired at three different time points prove the effectiveness of the proposed method compared with some existing methods.

Modelling and Delineation of Hydrocarbon Micro - Seepage Prone Zones on Soil and Sediment in Ugwueme, South Eastern Nigeria with Soil Adjustment Vegetation Index (SAVI)

Remote Sensing is an excellent tool in monitoring, mapping and interpreting areas, associated with hydrocarbon micro-seepage. An important technique in remote sensing known as the Soil Adjusted Vegetation Index (SAVI), adopted in many studies is often used to minimize the effect of brightness reflectance in the Normalized Difference Vegetation Index (NDVI), related with soil in areas of spare vegetation cover, and mostly in areas of arid and semi–arid regions. The study aim at analyzing the effect of hydrocarbon micro – seepage on soil and sediments in Ugwueme, Southern Eastern Nigeria, with SAVI image classification method. To achieve this aim, three cloud free Landsat images, of Landsat 7 TM 1996 and ETM+ 2006 and Landsat 8 OLI 2016 were utilized to produce different SAVI image classification maps for the study. The SAVI image classification analysis for the study showed three classes viz Low class cover, Moderate class cover and high class cover. The category of high SAVI density classification was observed to increase progressive from 31.95% in 1996 to 34.92% in 2006 and then to 36.77% in 2016. Moderately SAVI density classification reduced from 40.53% in 1996 to 38.77% in 2006 and then to 36.96% in 2016 while Low SAVI density classification decrease progressive from 27.51% in 1996 to 26.31% in 2006 and then increased to 28.26% in 2016. The SAVI model is categorized into three classes viz increase, decrease and unchanged. The un – changed category increased from 12.32km2 (15.06%) in 1996 to 17.17 km2 (20.96%) in 2006 and then decelerate to 13.50 km2 (16.51%) in 2016. The decrease category changed from 39.89km2 (48.78%) in 1996 to 40.45 km2 (49.45%) in 2006 and to 51.52 km2 (63.0%) in 2016 while the increase category changed from 29.57km2 (36.16%) in 1996 to 24.18 km2 (29.58%) in 2006 and to 16.75 km2 (20.49%) in 2016. Image differencing, cross tabulation and overlay operations were some of the techniques performed in the study, to ascertain the effect of hydrocarbon micro - seepage. The Markov chain analysis was adopted to model and predict the effect of the hydrocarbon micro - seepage for the study for 2030. The study expound that the SAVI is an effective technique in remote sensing to identify, map and model the effect of hydrocarbon micro - seepage on soil and sediment particularly in areas characterized with low vegetation cover and bare soil cover.

Hybrid modelling approach for water body change detection at Chalan Beel area in northern Bangladesh

Water is a strategic resource for both socio-economic development and human life. The current study has been carried out for spatio-temporal change detection of surface water bodies during winter period using hybrid modelling approach. The study area has fallen in the northern part of Bangladesh and is locally called Chalan Beel with 5 million of in habitants, a prominent intensive crop production, surface and groundwater irrigation, high evapotranspiration, and water scarcity. For the detection of water body changes, satellite images of 1999 and 2011 were used, and the following image fusion techniques were applied: (a) Gram-Schmidt (GS), (b) modified intensity hue saturation (IHS), (c) high-pass filter (HPF), and (d) wavelet. Landsat 7/ETM + panchromatic (PAN) band of 15 m × 15 m resolution in 1999 and Landsat 5/TM multispectral (MS) bands of 30 m × 30 m resolution in 2011 were allied each other to generate high-resolution image that contains information of two different years. The fused images were classified to extract the water bodies using four classification methods: (a) artificial neural network (ANN), (b) support vector machine (SVM) and (c) maximum likelihood (ML). To analyze the quality of the fused images, statistical calculation (quantitatively) and Laplacian edge detection (qualitatively) were used. To validate the fused image classification results, the multispectral images from 1999 and 2011 were again individually classified using principal component analysis (PCA), normalized difference water index (NDWI), and image differencing (ID) processes and compared with the previous classification. Surprisingly, the results showed that two-third of the areas dried up in 10 years.

Change Detection in Multispectral Remote Sensing Images with Leader Intelligence PSO and NSCT Feature Fusion

Change detection (CD) using Remote sensing images have been a challenging problem over the years. Particularly in the unsupervised domain it is even more difficult. A novel automatic change detection technique in the unsupervised framework is proposed to address the real challenges involved in remote sensing change detection. As the accuracy of change map is highly dependent on quality of difference image (DI), a set of Normalized difference images and a complementary set of Normalized Ratio images are fused in the Nonsubsampled Contourlet Transform (NSCT) domain to generate high quality difference images. The NSCT is chosen as it is efficient in suppressing noise by utilizing its unique characteristics such as multidirectionality and shift-invariance that are suitable for change detection. The low frequency sub bands are fused by averaging to combine the complementary information in the two DIs, and, the higher frequency sub bands are merged by minimum energy rule, for preserving the edges and salient features in the image. By employing a novel Particle Swarm Optimization algorithm with Leader Intelligence (LIPSO), change maps are generated from fused sub bands in two different ways: (i) single spectral band, and (ii) combination of spectral bands. In LIPSO, the concept of leader and followers has been modified with intelligent particles performing Lévy flight randomly for better exploration, to achieve global optima. The proposed method achieved an overall accuracy of 99.64%, 98.49% and 97.66% on the three datasets considered, which is very high. The results have been compared with relevant algorithms. The quantitative metrics demonstrate the superiority of the proposed techniques over the other methods and are found to be statistically significant with McNemar’s test. Visual quality of the results also corroborate the superiority of the proposed method.

RETRACTED ARTICLE: A feature based change detection approach using multi-scale orientation for multi-temporal SAR images

ABSTRACT Excellent operation regardless of weather conditions and superior resolution independent of sensor light are the most attractive and desired features of synthetic aperture radar (SAR) imagery. This paper proposes an exclusive multi-scale with multiple orientation approach for multi-temporal SAR images. This approach integrates pre-processing and change detection. Pre-processing is performed on the SAR imagery through speckle reducing anisotropic diffusion and discrete wavelet transform. The processed speckle-free images are designed by Log-Gabor filter bank in terms of multi-scale with multiple orientations. The maximum magnitude of multiple orientations is concatenated to obtain feature-based scale representation. Each scale is dealt with multiple orientations and is compared by band-wise subtraction to retrieve difference image (DI) coefficient. The series of the difference coefficients from each scale are add-on together to estimate a DI. Thus, the resultant image of multi-scale orientation gives perception of detailed information with specific contour. Constrained k-means clustering algorithm is preferred to achieve change and un-change map. Performance of the proposed approach is validated on three real SAR image datasets. The effective change detection is examined by using confusion matrix parameters. Experimental results are described to show the efficacy of the proposed approach.

Self-Supervised Representation Learning for Remote Sensing Image Change Detection Based on Temporal Prediction

Traditional change detection (CD) methods operate in the simple image domain or hand-crafted features, which has less robustness to the inconsistencies (e.g., brightness and noise distribution, etc.) between bitemporal satellite images. Recently, deep learning techniques have reported compelling performance on robust feature learning. However, generating accurate semantic supervision that reveals real change information in satellite images still remains challenging, especially for manual annotation. To solve this problem, we propose a novel self-supervised representation learning method based on temporal prediction for remote sensing image CD. The main idea of our algorithm is to transform two satellite images into more consistent feature representations through a self-supervised mechanism without semantic supervision and any additional computations. Based on the transformed feature representations, a better difference image (DI) can be obtained, which reduces the propagated error of DI on the final detection result. In the self-supervised mechanism, the network is asked to identify different sample patches between two temporal images, namely, temporal prediction. By designing the network for the temporal prediction task to imitate the discriminator of generative adversarial networks, the distribution-aware feature representations are automatically captured and the result with powerful robustness can be acquired. Experimental results on real remote sensing data sets show the effectiveness and superiority of our method, improving the detection precision up to 0.94–35.49%.

A Deep CFHT Optical Search for a Counterpart to the Possible Neutron Star–Black Hole Merger GW190814

Abstract We present a wide-field optical imaging search for electromagnetic counterparts to the likely neutron star–black hole (NS–BH) merger GW190814/S190814bv. This compact binary merger was detected through gravitational waves by the LIGO/Virgo interferometers, with masses suggestive of an NS–BH merger. We imaged the LIGO/Virgo localization region using the MegaCam instrument on the Canada–France–Hawaii Telescope (CFHT). We describe our hybrid observing strategy of both tiling and galaxy-targeted observations, as well as our image differencing and transient detection pipeline. Our observing campaign produced some of the deepest multiband images of the region between 1.7 and 8.7 days post-merger, reaching a 5σ depth of g > 22.8 (AB mag) at 1.7 days and i > 23.1 and i > 23.9 at 3.7 and 8.7 days, respectively. These observations cover a mean total integrated probability of 67.0% of the localization region. We find no compelling candidate transient counterparts to this merger in our images, which suggests that the lighter object was tidally disrupted inside of the BH’s innermost stable circular orbit, the transient lies outside of the observed sky footprint, or the lighter object is a low-mass BH. We use 5σ source detection upper limits from our images in the NS–BH interpretation of this merger to constrain the mass of the kilonova ejecta to be M ej ≲ 0. 015M ⊙ for a “blue” ( ) kilonova and M ej ≲ 0. 04M ⊙ for a “red” ( ) kilonova. Our observations emphasize the key role of large-aperture telescopes and wide-field imagers such as CFHT MegaCam in enabling deep searches for electromagnetic counterparts to gravitational-wave events.

Large-Scope PolSAR Image Change Detection Based on Looking-Around-and-Into Mode

A new method based on the Looking-Around-and-Into (LAaI) mode is proposed for the task of change detection in large-scope Polarimetric Synthetic Aperture Radar (PolSAR) image. Specifically, the LAaI mode consists of two processes named Look-Around and Look-Into, which are accomplished by attention proposal network (APN) and recurrent convolutional neural network (CNN) (Recurrent CNN), respectively. The former provides certain subregions efficiently, and the latter detects changes in subregions accurately. In Look-Around, difference image (DI) of whole PolSAR images is calculated first to get global information; then, APN is established to locate the position of interested subregions intentionally by paying special attention to; next interested subregions that contain changed area in high probability are picked out as candidate-regions. Moreover, candidate-regions are sorted in importance descending order so that highly interested regions have priority to be detected. In Look-Into, candidate-regions of different scales are selected at first; then, Recurrent CNN is constructed and employed to deal with multiscale PolSAR subimages so that clearer and finer change detection results are generated. The process is repeated until all candidate-regions are detected. As a whole, the proposed algorithm based on the LAaI mode looks around whole images first to find out the possible position of changes (candidate-regions generation in Look-Around) and then reveal the exact shape of changes in different scales (multiscale change detection in Look-Into). The effect of APN and Recurrent CNN is verified in experiments, and it shows that the proposed method performs well in the task of change detection in the large-scope PolSAR image.