Search

Cloud cover is a significant factor affecting the effectiveness of satellite-based Earth observations. Existing cloud detection algorithms primarily rely on imaging data from satellite sensors in the visible to near-infrared spectral range, making it challenging to achieve day-and-night cloud monitoring. Convolutional neural networks have shown outstanding performance in previous cloud detection algorithms due to their robust ability to extract local information. However, their inherent inductive bias limits their capacity to learn long-range semantic information. To address these challenges, we proposed SwinCloud, a U-shaped semantic segmentation network based on an enhanced Swin Transformer for cloud detection in the thermal infrared spectral range. Specifically, we augment the Swin Transformer's window attention module with a CNN-based parallel pathway to effectively model global-local information. We employ a feature fusion module before the final upsampling module in the decoder to better integrate low-level spatial information and high-level semantic information. On the Landsat-8 cloud detection dataset, our model outperforms state-of-the-art methods. When transferred to the SDGSAT-TIS cloud detection dataset, the mIOU of experiment results reaches 69.9%, demonstrating the strong transferability of SwinCloud across different sensors. We also applied SwinCloud to cloud detection in the visible bands of Landsat-8. The results demonstrated SwinCloud's generalization capability across different bands.

With the rapid development of the aviation industry, the impact of contrails on global warming is becoming increasingly significant. However, due to limitations in sensor technology and data sources, precise, all-weather observations of contrails remain a significant challenge. To address this, this study utilizes thermal imagery data from the SDGSAT-1 to classify newly formed contrails based on their length, creating SDGCONTR, the first three-band thermal infrared contrail dataset with a 30-meter spatial resolution. Based on this dataset, we propose an effective contrail image classification method: the SDGContrail Images Classification Method. This method builds on the existing Multi-axis Vision Transformer deep learning model. In the preprocessing stage, it employs an innovative contrail enhancement method by creating masks based on contrail brightness temperature and brightness temperature ratio percentiles in thermal infrared images. Our method achieved optimal classification performance compared to other methods, with the F1 Score of 95.2%. Using this model, we further analyzed the geographic and diurnal distribution patterns of long and short contrails, validating the mechanism of contrail formation. Our method not only enhances the ability to conduct large-scale, all-weather, and detailed contrail observations, but also sets a new standard for environmental monitoring and aviation safety.

High-precision geometric registration has been widely used in remote-sensing applications, such as image fusion. In practice, it is found that the geometric deformation (GD) of the bidirectional whisk-broom images, namely forward- and backward-scanning images, is not evenly distributed. Additionally, the deformation in some special areas exhibits significant nonlinear characteristics, which ultimately makes the traditional linear homography matrix inoperative in describing the GD of the whole image. Meanwhile, it is difficult to establish an elastic deformation model for the whole image due to the large swath and complex GD. To solve the problems above, we proposed a thin-plate-spline-based registration method for correcting the nonlinear deformation of large-size bidirectional whisk-broom images through scene classification. First, a classification method based on histogram Gaussian distribution fitting is proposed to divide the image into sea and land blocks of the same size. And the adjacent blocks must have overlapping areas. Second, the thin-plate-spline transformation is carried out to correct the local nonlinear deformation of the land blocks. Finally, the registered blocks are stitched by a gradual image fusion method to obtain the fully registered images. Our proposed method can correct the local nonlinear GD of bidirectional whisk-broom images of sustainable development science satellite 1 thermal infrared spectrometer (TIS), and the registration accuracy is better than 0.3 pixels, which is of great significance for the practical application of TIS thermal infrared image.

SDGSAT-1, the inaugural Earth science satellite commissioned by the Chinese Academy of Sciences, successfully lifted off from China's Taiyuan Satellite Launch Center on 5 November 2021. Cross calibration has emerged as an indispensable methodology, providing an effective means for quality assessment, stability monitoring, and uncertainty analysis. This is achieved through the meticulous selection of an appropriate reference instrument. In our investigation, we conducted a comprehensive evaluation of the absolute radiometric calibration accuracy of the thermal infrared spectrometer (TIS), a pivotal component aboard SDGSAT-1. Our evaluation centered on a meticulous comparative analysis, where we juxtaposed the brightness temperature (BT) values recorded by TIS with the BT values derived from TIRS-2, computed following the spectral alignment with TIS. Our findings reveal that the current absolute BT bias for the B2 channel of TIS is less than 0.52 K. Meanwhile, the B3 channel exhibits a slightly larger fluctuation in absolute BT bias, yet remains within acceptable parameters, registering at less than 1 K. Notably, B3 emerges as the more dependable option for accurately gauging the temperature of the target region. Specifically, when the BT of B3 surpasses 290 K, the BT bias remains consistently below 0.3 K. In light of the approximately 40-min time difference in the water surface data utilized in our study, we conducted simulations to assess the channel BT of TIS under the scenario of a 0.25 °C water temperature variation. Our simulations conclusively demonstrate that the corresponding change in BT does not exceed 0.25 K.

Launched in November 2021, the sustainable development science satellite-1 carries the thermal infrared spectrometer (TIS) as a pivotal payload. Achieving precise radiometric calibration for geophysical parameter accuracy is crucial. Leveraging the frequent cross-calibration capability of the geostationary satellite FY-4A/advanced geostationary radiation imager (AGRI), this study addresses observational geometry challenges. The proposed radiance correction model for FY-4A/AGRI effectively reduces radiance discrepancies caused by differences in satellite zenith angles between TIS and AGRI. After integrating the corrected data into nadir-observed statistics, the long-term brightness temperature (BT) bias of TIS and AGRI was monitored. The results show that the daytime and nighttime BT bias exhibit similar trends, with the B2 band mainly showing negative bias. Overall, the BT bias for the B2 and B3 bands is within 1 K. The standard deviation of the BT bias for approximately 2250 sample points was calculated, revealing the values of 0.22 K and 0.40 K during the daytime, and 0.31 K and 0.57 K at night. The introduced model error remains below 0.31 K. Regarding the uncertainty of the BT bias introduced by spectral matching, when the spectral broadening is 1.15 times the wavelength interval, the uncertainty is less than 0.29 K for B2 and less than 0.38 K for B3. This work enhances our understanding of radiometric calibration challenges, providing insights into satellite-based geophysical parameter accuracy.

A thermal infrared spectrometer is an important sensor onboard the SDGSAT-1 dedicated to serving the 2030 Agenda for Sustainable Development of the United Nations. Field campaigns were conducted from January to December 2022 to evaluate its radiometric calibrations. A radiance-based calibration method was used to derive the spectral radiance of the lake surface around the satellite overpass using a boat-deployed TIR spectrometer or radiometer. The radiometric calibration conditions were quantified by examining the temperature bias between the satellite-based brightness temperature and the predicted at-sensor brightness temperature. Our results indicated that the average brightness temperature bias reached 0.772 K, −1.032 K, and −0.300 K, with RMSEs of 1.00, 1.10, and 0.43 K in bands 1, 2 and 3, respectively. Eighty percent of the field experiments achieved a high calibration accuracy with temperature biases of less than 1 K, and the calibration accuracy of band 3 was better than that of other bands. Atmospheric conditions and surface characterization play important roles in determining the calibration accuracy through surface observations and atmospheric transfer simulations. With the increasing need for SDGSAT-1 products, periodic field experiments are needed to monitor and improve the operational radiometric calibration of satellite instruments to create high-resolution and reasonable scientific datasets.

The effective discrimination of targets from backgrounds in environments characterized by a low signal-to-clutter ratio (SCR) is paramount for the advancement of infrared small target detection (IRSTD). In this work, we propose a novel detection framework predicated on low-rank sparse decomposition (LRSD), incorporating an improved logarithmic norm and a mechanism for sparse structure refinement, herein referred to as the improved logarithmic norm and sparse structure refinement (ILN-SSR). The ILN-SSR framework more precisely characterizes the sparse properties of both the background and the target, enabling a more effective distinction between the target and its background. Initially, our approach entails the utilization of an improved logarithmic norm to precisely estimate the low-rank attributes of the infrared image background. This is followed by the employment of a linear sparse regularization term alongside a target-traits-based sparse regularization term aimed at meticulously identifying targets within sparse regions and refining the sparse structure. Subsequently, we combine these components into the ILN-SSR framework, which formulates IRSTD as an optimization problem. The resolution of this framework is achieved through the implementation of the alternating direction method of multipliers (ADMM). The efficacy of the proposed framework is corroborated through the analysis of six image sequences. Comprehensive experimental assessments affirmed the framework’s substantial robustness in navigating various complex backgrounds.

Stripe noise is present in the on-orbit images captured by the Sustainable Development Goals Satellite-1 (SDGSAT-1) Thermal Infrared Spectrometer (TIS). Removing these stripes lays the crucial groundwork for subsequent remote sensing data applications. This paper proposes a high-performance method for removing stripe noise in SDGSAT-1 TIS large dynamic range imaging. Based on the fixation properties of the stripe noise, we utilize the a priori information to enhance the gradient variation around the stripe region. This step successfully separates the background from the noise and speeds up the noise extraction. We extract local thresholds from the central difference matrix using an adaptive threshold extraction method based on the Maximum Interclass Variance (OTSU). To maximize the retention of image details, we combine structural information about stripe positions with a weighted filter kernel based on the L2 norm to predict noise digital number (DN). The proposed method extensively experiments with simulated and on-orbit data in different scenarios. It has the best overall performance in qualitative evaluation, quantitative evaluation, processing speed, and robustness.

Two-terminal artificial optoelectronic devices in which photoexcited carriers and ions are coupled constitute a built-in artificial neural network that can sense, memory and process simultaneously.

The quality of hyperspectral images seriously impedes subsequent high-level vision tasks such as image segmentation, image encoding, and target detection. However, the frequency, spectral, and spatial properties of the hyperspectral noise pictures are not utilized fully by existing image denoising algorithms. To address this issue, a novel convolutional network based on united Octave and attention mechanism (UOANet) is proposed to extract the frequency-spectral-spatial-feature for denoising the actual noise of HSIs. In particular, the negative residual mapping embedded in Unet is proposed for multiscale abstract representation and two modules are designed for modeling global noisy HSI features in the frequency-spectral-spatial domain. First, with the use of residual Octave convolution module, our model can focus on the intrinsic properties of HSI noise distribution for desirable noise removal. Next, a parallel spatial-spectral attention module is used to fully utilize the rich spectrum data and the various spatial data of each band in HSI, which improves the richness of HSI details after denoising. Experimental results on both synthetic and real HSIs demonstrate the validity and superiority of UOANet compared with the state-of-the-arts under various noise settings.

Ground Control Points (GCPs) are of great significance for applications involving the registration and fusion of heterologous remote sensing images (RSIs). However, utilizing low-level information rather than deep features, traditional methods based on intensity and local image features turn out to be unsuitable for heterologous RSIs because of the large nonlinear radiation difference (NRD), inconsistent resolutions, and geometric distortions. Additionally, the limitations of current heterologous datasets and existing deep-learning-based methods make it difficult to obtain enough precision GCPs from different kinds of heterologous RSIs, especially for thermal infrared (TIR) images that present low spatial resolution and poor contrast. In this paper, to address the problems above, we propose a convolutional neural network-based (CNN-based) layer-adaptive GCPs extraction method for TIR RSIs. Particularly, the constructed feature extraction network is comprised of basic and layer-adaptive modules. The former is used to achieve the coarse extraction, and the latter is designed to obtain high-accuracy GCPs by adaptively updating the layers in the module to capture the fine communal homogenous features of the heterologous RSIs until the GCP precision meets the preset threshold. Experimental results evaluated on TIR images of SDGSAT-1 TIS and near infrared (NIR), short wave infrared (SWIR), and panchromatic (PAN) images of Landsat-8 OLI show that the matching root-mean-square error (RMSE) of TIS images with SWIR and NIR images could reach 0.8 pixels and an even much higher accuracy of 0.1 pixels could be reached between TIS and PAN images, which performs better than those of the traditional methods, such as SIFT, RIFT, and the CNN-based method like D2-Net.

High-accuracy sea surface temperature (SST) retrieval near nuclear power plants (NPPs) is one of the most significant indicators for evaluating marine ecological environment quality, monitoring the real-time situation of thermal discharge, and supporting planning decisions. However, complex computations, the inaccessible real-time vertical profile of the atmosphere, and the uncertainty of atmospheric profile data increase the error of SST retrieval. Additionally, influenced by their low spatial resolution, the widely used AVHRR/MODIS remote sensing images (RSIs) are unable to retrieve the detailed distribution of SST in small scale regions such as coastal NPPs. In this paper, we propose a simplified split-window-based temperature retrieval method (the SW method) suitable for SDGSAT-1 30 m thermal infrared spectrometer (TIS) RSIs. Specially, this method only needs atmospheric transmittance and surface emissivity by counteracting the average atmospheric temperature to monitor the thermal discharge of offshore NPPs. First, the geometric and radiometric calibrated thermal infrared and multi-spectral cloudless data of the target regions are selected to obtain the corresponding apparent radiance of the RSIs. Second, in accordance with the red and near-infrared (NIR) bands of multi-spectral RSIs, the surface emissivity is calculated to distinguish water from land. Next, we determine the atmospheric profile parameters from the weather conditions of the target region at the imaging time. Finally, according to the theory of surface-atmosphere radiative transfer, the SST of target regions is retrieved with the proposed SW method, and the results are compared with those of the conventional radiative transfer equation (RTE), mono-window (MW), and the nonlinear sea surface temperature (NLSST) algorithms. The experimental results indicate that the SST retrieved from the split-window algorithms (i.e., SW and NLSST) are generally higher than those of the single-channel algorithms (i.e., RTE and MW). The SST difference between the SW algorithm and the NLSST algorithm is within 0.5 °C. In addition, SDGSAT-1 can monitor the seasonal detailed variation of the thermal discharge near coastal NPPs. This article is the first to attempt to quantitative small-scale SST retrieval based on thermal infrared and multi-spectral images obtained from the SDGSAT-1 TIS and a multispectral imager (MII), and therefore, provide an effective reference for marine environment monitoring.

Infrared dim and small target detection is widely used in military and civil fields. Traditional methods in that application rely on the local contrast between the target and background for single-frame detection. On the other hand, those algorithms depend on the motion model with fixed parameters for multi-frame association. For the great similarity of gray value and the dynamic changes of motion model parameters in the condition of low SNR and strong clutter, those methods possess weak robustness, low detection probability, and high false alarm rate. In this paper, an infrared video sequences encoding and decoding model based on Bidirectional Convolutional Long Short-Term Memory structure (Bi-Conv-LSTM) and 3D Convolutional structure (3D-Conv) is proposed, addressing the problem of high similarity and dynamic changes of parameters. For solving the problem of dynamic change in parameters, Bi-Conv-LSTM structure is used to learn the motion model of targets. And for the problem of low local contrast, 3D-Conv structure is adopted to extend receptive field in the time dimension. In order to improve the precision of detection, the Decoding part is divided into two different full connections with distinctive active function. Simulation results show that the trajectory detection accuracy of the proposed model is more than 90% under the condition of low SNR and maneuvering motion, which is better than traditional method of 80% in DB-TBD 20% in others. Real data experiment to illustrate that that our proposed method can detect small infrared targets of a low false alarm rate and high detection probability.

Semantic segmentation (SS) has been widely applied for cloud detection (CD) in remote sensing images (RSIs) with high spatial and spectral resolution because of its effective pixel-level feature extraction structure. However, the typical model of lightweight SS, namely the fully convolutional network (FCN) with only seven layers, has difficulty in extracting high-level features, and the heavy pyramid scene parsing network (PSPNet) with complicated calculations is not practical in real-time CD, let alone on-orbit CD. So, in view of the problems above, we propose a compact attention mechanism cloud detection network (AM-CDN) based on the modified FCN to refine and fuse the multi-scale features for on-orbit CD. Specifically, taking the FCN as the baseline, our model increases the numbers of hidden layers and adds the residual connections between the input and output to eliminate the network degradation and extract the advanced context feature maps effectively. To expand the receptive field without losing the spatial information, the ordinary convolutions in FCN are replaced by the dilated convolution in AM-CDN. And inspired by the selective kernels of human vision, we introduce the convolutional attention mechanism (AM) into the encoder to adaptively adjust the receptive field to highlight the key texture features. According to experimental results using Landsat-8 infrared RSIs, the accuracy of the proposed CD method is 95.31%, which is 10.17% higher than that of FCN. And the calculation complexity of AM-CDN is only 7.63% of that of PSPNet.

Feature representation is the key to the hyperspectral images (HSI) inpainting. Existing works mainly focus on using spectral and temporal auxiliary images to inpainting the corrupted region, which were proved to be low robust for all bands missing and high requirements for image acquisition. In this work, we propose an end-to-end inpainting framework for HSI based on convolutional neural networks, which does not require auxiliary images and makes full use of both spectral characteristics and spatial information. For spectral characteristics, a channel attention mechanism is proposed to reduce the redundancy of hyperspectral channels and model the correlation between channels. For spatial information, a local discriminative network is able to cope with the structural continuity of the corrupted regions, and a gradient consistency loss function is proposed to maintain the texture consistency of HSIs. Experimental results in the Airborne Visual Infrared Imaging Spectrometer Indians Pines public dataset and Feicheng Hyperspectral datasets show that our proposed method can provide competitive results compared with state-of-the-art methods.

Thermal infrared imagery plays an important role in a variety of fields, such as surface temperature inversion and urban heat island effect analysis, but the spatial resolution has severely restricted the potential for further applications. Data fusion is defined as data combination using multiple sensors, and fused information often has better results than when the sensors are used alone. Since multi-resolution analysis is considered an effective method of image fusion, we propose an MTF-GLP-TAM model to combine thermal infrared (30 m) and multispectral (10 m) information of SDGSAT-1. Firstly, the most relevant multispectral bands to the thermal infrared bands are found. Secondly, to obtain better performance, the high-resolution multispectral bands are histogram-matched with each thermal infrared band. Finally, the spatial details of the multispectral bands are injected into the thermal infrared bands with an MTF Gaussian filter and an additive injection model. Despite the lack of spectral overlap between thermal infrared and multispectral bands, the fused image improves the spatial resolution while maintaining the thermal infrared spectral properties as shown by subjective and objective experimental analyses.

The development of infrared remote sensing technology improves the ability of night target observation, and thermal imaging systems (TIS) play a key role in the military field. Ship detection using thermal infrared (TI) remote sensing images (RSIs) has aroused great interest for fishery supervision, port management, and maritime safety. However, due to the high secrecy level of infrared data, thermal infrared ship datasets are lacking. In this paper, a new three-bands thermal infrared ship dataset (TISD) is proposed to evaluate all-day ship target detection algorithms. All images are from SDGSAT-1 satellite TIS three bands RSIs of the real world. Based on the TISD, we use the state-of-the-art algorithm as a baseline to do the following. (1) Common ship detection methods and existing ship datasets from synthetic aperture radar, visible, and infrared images are elementarily summarized. (2) The proposed standard deviation of single band, correlation coefficient of combined bands, and optimum index factor features of three-bands datasets are analyzed, respectively. Combined with the above theoretical analysis, the influence of the bands’ information input on the detection accuracy of a neural network model is explored. (3) We construct a lightweight network based on Yolov5 to reduce the number of floating-point operations, which is beneficial to reduce the inference time. (4) By utilizing up-sampling and registration pre-processing methods, TI images are fused with glimmer RSIs to verify the detection accuracy at night. In practice, the proposed datasets are expected to promote the research and application of all-day ship detection.

Atmospheric refraction is one of the most significant factors that affect the geolocation accuracy of high-resolution remote sensing images. However, most of the current atmospheric refraction correction methods based on empirical data neglect the spatiotemporal variation of pressure, temperature, and humidity of the atmosphere, inevitably resulting in poor geometric positioning accuracy. Therefore, in terms of the problems mentioned above, this study proposed a spatiotemporal atmospheric refraction correction method (SARCM) based on global measured data to avoid the uncertainty of traditional empirical models. Initially, the atmosphere was stratified into 42 layers according to their pressure property, and each layer was divided into 1,042,560 grid cells with intervals of 0.25 longitude and 0.25 latitude. Then, the atmospheric refractive index of each grid in the imaging region was accurately calculated using the high-precision Ciddor formula, and the result was interpolated using three splines. Subsequently, according to the rigorous geometric positioning model, the line-of-sight of each pixel and the viewing zenith angle outside the atmosphere in WGS84 were derived to provide input for atmospheric refraction correction. Finally, the coordinates of the ground control points were corrected with the calculated atmospheric refractive index and Snell’s law. The experimental results showed that the proposed SARCM could effectively improve the positioning accuracy of the image with a large viewing zenith angle, and especially, the improvement percentage for a viewing zenith angle of 34.2426° in the x-direction was 99.5%. Moreover, the atmospheric refraction correction result of the SARCM was better than that of the primary state-of-the-art methods.

SDGSAT-1 was launched in November 2021, and TIS (Thermal infrared sensor) is a major instrument onboard this satellite. An analysis of the radiometric calibration and noise performance of the TIS has been carried out in the thermal vacuum chamber before launch in order to ensure that it meets the requirements. The prelaunch test results show NEdT (noise equivalent temperature difference) is 0.034 K, 0.047 K and 0.076 K@300 K for the three bands, respectively. The maximum fitting residuals are less than 0.5 K at measured blackbody temperatures. In addition, this paper also discusses the dependence between TIS performance and instrument temperature and focal plane array (FPA) temperature. The good radiometric and noise performance of TIS demonstrates it has potential to provide high resolution thermal remote sensing data in urban heat island, and other environmental issues research.

The thermal imaging image of the Sustainable Development Science Satellite (SDGSAT-1) is mainly used for high-resolution observations of the ground width, due to the influence of blind elements and non-uniformity of the detector, and the system is a pendulum sweep imaging mode, resulting in fringed noise in the image. In this paper, a Fringing algorithm based on LRSID (low-rank-based single-image decomposition) algorithm is proposed, which can effectively remove the lateral and vertical fringe noise of the thermal imager and maintain the detail and clarity of the image. First, pretreatment of the obvious light and dark stripes then, based on LLSID algorithm, the vertical direction pinstripes and horizontal stripes are processed; finally, the fringed frequency band of the original image is replaced in the frequency domain with the image frequency domain processed by the LRSID algorithm, and then the Fourier inverse transformation is performed to obtain the final image. Using the method proposed in this paper, the simulated and actual SDGSAT-1 thermal imaging camera remote sensing stripes images are removed, and the visual and quantitative indicators are compared with the processing results of other algorithms, and the results show that the proposed algorithm has the best performance to remove the stripes, which can effectively remove horizontal and vertical fringes at the same time, and retain the detail and clarity of the image.

Abstract Photodetectors built from conventional bulk materials such as silicon, III–V or II–VI compound semiconductors are one of the most ubiquitous types of technology in use today. The past decade has witnessed a dramatic increase in interest in emerging photodetectors based on perovskite materials driven by the growing demands for uncooled, low‐cost, lightweight, and even flexible photodetection technology. Though perovskite has good electrical and optical properties, perovskite‐based photodetectors always suffer from nonideal quantum efficiency and high‐power consumption. Joint manipulation of electrons and photons in perovskite photodetectors is a promising strategy to improve detection efficiency. In this review, electrical and optical characteristics of typical types of perovskite photodetectors are first summarized. Electrical manipulations of electrons in perovskite photodetectors are discussed. Then, artificial photonic nanostructures for photon manipulations are detailed to improve light absorption efficiency. By reviewing the manipulation of electrons and photons in perovskite photodetectors, this review aims to provide strategies to achieve high‐performance photodetectors.

Photovoltaic devices based on 2D materials still suffer from low quantum efficiencies due to interfacial charge recombination and inefficient contacts. Here, the authors design photovoltaic detectors and photodiodes based on MoS2 and doped AsP heterojunction with unilateral depletion region reporting high external quantum efficiency of 71% under zero applied bias.

Garbage detection is important for environmental monitoring in large areas. However, the manual patrol is time-consuming and labor-intensive. This paper proposes a method for monitoring garbage distribution in large areas with airborne hyperspectral data. Since there is no public hyperspectral garbage dataset, a hyperspectral garbage dataset Shandong Suburb Garbage is labeled and published. For garbage detection, a new hyperspectral image (HSI) classification network MSCNN (Multi-Scale Convolutional Neural Network) is proposed to classify the pixels of HSI data and generate binary garbage segmentation map. Unsupervised region proposal generation algorithm Selective Search and None Maximum Suppression (NMS) are used to extract the location and the size of garbage areas based on the garbage segmentation map. The experiment results show that the proposed algorithm has a good performance on garbage detection in large areas. In addition, the MSCNN has achieved better performance in comparison with other HSI classification methods in the public HSI datasets Indian Pines and Pavia University.

The design of the thermal protection system requires high-precision and high-reliability CFD simulation for validation. To accurately predict the hypersonic aerodynamic heating, an overall simulation strategy based on mutual selection is proposed. Foremost, the grid criterion based on the wall cell Reynolds number is developed. Subsequently, the dependence of the turbulence model and the discretization scheme is considered. It is suggested that the appropriate value of wall cell Reynolds number is 1 through careful comparison between one another and with the available experimental data. The excessive number of cells is not recommended due to time-consuming computation. It can be seen from the results that the combination of the AUSM+ discretization scheme and the Spalart-Allmaras turbulence model has the highest accuracy. In this work, the heat flux error of the stagnation point is within 1%, and the overall average relative error is within 10%.

Due to sensor instability and atmospheric interference, hyperspectral images (HSIs) often suffer from different kinds of noise which degrade the performance of downstream tasks. Therefore, HSI denoising has become an essential part of HSI preprocessing. Traditional methods tend to tackle one specific type of noise and remove it iteratively, resulting in drawbacks including inefficiency when dealing with mixed noise. Most recently, deep neural network-based models, especially generative adversarial networks, have demonstrated promising performance in generic image denoising. However, in contrast to generic RGB images, HSIs often possess abundant spectral information; thus, it is non-trivial to design a denoising network to effectively explore both spatial and spectral characteristics simultaneously. To address the above issues, in this paper, we propose an end-to-end HSI denoising model via adversarial learning. More specifically, to capture the subtle noise distribution from both spatial and spectral dimensions, we designed a Residual Spatial-Spectral Module (RSSM) and embed it in an UNet-like structure as the generator to obtain clean images. To distinguish the real image from the generated one, we designed a discriminator based on the Multiscale Feature Fusion Module (MFFM) to further improve the quality of the denoising results. The generator was trained with joint loss functions, including reconstruction loss, structural loss and adversarial loss. Moreover, considering the lack of publicly available training data for the HSI denoising task, we collected an additional benchmark dataset denoted as the Shandong Feicheng Denoising (SFD) dataset. We evaluated five types of mixed noise across several datasets in comparative experiments, and comprehensive experimental results on both simulated and real data demonstrate that the proposed model achieves competitive results against state-of-the-art methods. For ablation studies, we investigated the structure of the generator as well as the training process with joint losses and different amounts of training data, further validating the rationality and effectiveness of the proposed method.

The automatic ship detection method for thermal infrared remote sensing images (TIRSIs) is of great significance due to its broad applicability in maritime security, port management, and target searching, especially at night. Most ship detection algorithms utilize manual features to detect visible image blocks which are accurately cut, and they are limited by illumination, clouds, and atmospheric strong waves in practical applications. In this paper, a complete YOLO-based ship detection method (CYSDM) for TIRSIs under complex backgrounds is proposed. In addition, thermal infrared ship datasets were made using the SDGSAT-1 thermal imaging system. First, in order to avoid the loss of texture characteristics during large-scale deep convolution, the TIRSIs with the resolution of 30 m were up-sampled to 10 m via bicubic interpolation method. Then, complete ships with similar characteristics were selected and marked in the middle of the river, the bay, and the sea. To enrich the datasets, the gray value stretching module was also added. Finally, the improved YOLOv5 s model was used to detect the ship candidate area quickly. To reduce intra-class variation, the 4.23–7.53 aspect ratios of ships were manually selected during labeling, and 8–10.5 μm ship datasets were constructed. Test results show that the precision of the CYSDM is 98.68%, which is 9.07% higher than that of the YOLOv5s algorithm. CYSDM provides an effective reference for large-scale, all-day ship detection.

The high-precision characterization of the intra-pixel sensitivity (IPS) for infrared focal plane array (FPA) photodetector is of great significance to high-precision photometry and astrometry in astronomy, as well as target tracking in under-sampled remote sensing images. The discrete sub-pixel response (DSPR) model and fill factor model have been used for IPS characterization in some studies. However, these models are incomplete and lack the description of physical process of charge diffusion and capacitance coupling, leading to the inaccuracy of IPS characterization. In this paper, we propose an improved IPS characterization method based on the diffusion and coupling physical (DCP) model for infrared FPA photodetector, which considering the processes of generation and collection of the charge, can improve the accuracy of IPS characterization. The IPS model can be obtained by convolving the ideal rectangular response function with the charge diffusion function and the capacitive coupling function. Then, the IPS model is convolved with the beam spot profile to obtain the beam spot scanning response model. Finally, we calculate the parameters of IPS by fitting the beam spot scanning response map with the proposed DCP model based on the Trust-Region-Reflective algorithm. Simulated results show that when using a 3 μm beam spot to scan, the error of IPS characterization based on DCP model is 0.63%, which is better than that of DSPR model’s 3.70%. Experimental results show that the fitting error of the beam spot scan response model based on DCP model is 4.29%, which is better than that of DSPR model’s 8.31%.

Affected by the vibrations and thermal shocks during launch and the orbit penetration process, the geometric positioning model of the remote sensing cameras measured on the ground will generate a displacement, affecting the geometric accuracy of imagery and requiring recalibration. Conventional methods adopt the ground control points (GCPs) or stars as references for on-orbit geometric calibration. However, inescapable cloud coverage and discontented extraction algorithms make it extremely difficult to collect sufficient high-precision GCPs for modifying the misalignment of the camera, especially for geostationary satellites. Additionally, the number of the observed stars is very likely to be inadequate for calibrating the relative installations of the camera. In terms of the problems above, we propose a novel on-orbit geometric calibration method using the relative motion of stars for geostationary cameras. First, a geometric calibration model is constructed based on the optical system structure. Then, we analyze the relative motion transformation of the observed stars. The stellar trajectory and the auxiliary ephemeris are used to obtain the corresponding object vector for correcting the associated calibration parameters iteratively. Experimental results evaluated on the data of a geostationary experiment satellite demonstrate that the positioning errors corrected by this proposed method can be within ±2.35 pixels. This approach is able to effectively calibrate the camera and improve the positioning accuracy, which avoids the influence of cloud cover and overcomes the great dependence on the number of the observed stars.

Due to the limited varieties and sizes of existing public hyperspectral image (HSI) datasets, the classification accuracies are higher than 99% with convolutional neural networks (CNNs). In this paper, we presented a new HSI dataset named Shandong Feicheng, whose size and pixel quantity are much larger. It also has a larger intra-class variance and a smaller inter-class variance. State-of-the-art methods were compared on it to verify its diversity. Otherwise, to reduce overfitting caused by the imbalance between high dimension and small quantity of labeled HSI data, existing CNNs for HSI classification are relatively shallow and suffer from low capacity of feature learning. To solve this problem, we proposed an HSI classification framework named deep residual spectral spatial setwork (DRSSN). By using shortcut connection structure, which is an asymmetry structure, DRSSN can be deeper to extract features with better discrimination. In addition, to alleviate insufficient training caused by unbalanced sample sizes between easily and hard classified samples, we proposed a novel training loss function named sample balanced loss, which allocated weights to the losses of samples according to their prediction confidence. Experimental results on two popular datasets and our proposed dataset showed that our proposed network could provide competitive results compared with state-of-the-art methods.

Facial recognition systems are often spoofed by presentation attack instruments (PAI), especially by the use of three-dimensional (3D) face masks. However, nonuniform illumination conditions and significant differences in facial appearance will lead to the performance degradation of existing presentation attack detection (PAD) methods. Based on conventional thermal infrared imaging, a PAD method based on the medium wave infrared (MWIR) polarization characteristics of the surface material is proposed in this paper for countering a flexible 3D silicone mask presentation attack. A polarization MWIR imaging system for face spoofing detection is designed and built, taking advantage of the fact that polarization-based MWIR imaging is not restricted by external light sources (including visible light and near-infrared light sources) in spite of facial appearance. A sample database of real face images and 3D face mask images is constructed, and the gradient amplitude feature extraction method, based on MWIR polarization facial images, is designed to better distinguish the skin of a real face from the material used to make a 3D mask. Experimental results show that, compared with conventional thermal infrared imaging, polarization-based MWIR imaging is more suitable for the PAD method of 3D silicone masks and shows a certain robustness in the change of facial temperature.

Portable box volume measurement has always been a popular issue in the intelligent logistic industry. This work presents a portable system for box volume measurement that is based on line-structured light vision and deep learning. This system consists of a novel 2 × 2 laser line grid projector, a sensor, and software modules, with which only two laser-modulated images of boxes are required for volume measurement. For laser-modulated images, a novel end-to-end deep learning model is proposed by using an improved holistically nested edge detection network to extract edges. Furthermore, an automatic one-step calibration method for the line-structured light projector is designed for fast calibration. The experimental results show that the measuring range of our proposed system is 100–1800 mm, with errors less than ±5.0 mm. Theoretical analysis indicates that within the measuring range of the system, the measurement uncertainty of the measuring device is ±0.52 mm to ±4.0 mm, which is consistent with the experimental results. The device size is 140 mm × 35 mm × 35 mm and the weight is 110 g, thus the system is suitable for portable automatic box volume measurement.