Shadow Region Research Articles

Image shadow removal based on pseudo-illuminated region and log domain transformation

ABSTRACT The presence of shadows in images can significantly degrade image quality. To address the challenges, this study utilizes pseudo-illumination regions and logarithmic domain transformations. Firstly, the input shadow image and its corresponding shadow region are segmented and clustered into multiple image blocks. Then, a comprehensive consideration of distance, color, and texture features is employed to construct pseudo illumination regions. Subsequently, the designed logarithmic domain transformation function is utilized for shadow removal. Finally, color correction and optimization of shadow edges complete the overall process. The feasibility of this approach has been preliminarily verified through experimental analysis.

Frequency‐Aware Facial Image Shadow Removal through Skin Color and Texture Learning

AbstractExisting facial image shadow removal methods predominantly rely on pre‐extracted facial features. However, these methods often fail to capitalize on the full potential of these features, resorting to simplified utilization. Furthermore, they tend to overlook the importance of low‐frequency information during the extraction of prior features, which can be easily compromised by noises. In our work, we propose a frequency‐aware shadow removal network (FSRNet) for facial image shadow removal, which utilizes the skin color and texture information in the face to help recover illumination in shadow regions. Our FSRNet uses a frequency‐domain image decomposition network to extract the low‐frequency skin color map and high‐frequency texture map from the face images, and applies a color‐texture guided shadow removal network to produce final shadow removal result. Concretely, the designed fourier sparse attention block (FSABlock) can transform images from the spatial domain to the frequency domain and help the network focus on the key information. We also introduce a skin color fusion module (CFModule) and a texture fusion module (TFModule) to enhance the understanding and utilization of color and texture features, promoting high‐quality result without color distortion and detail blurring. Extensive experiments demonstrate the superiority of the proposed method. The code is available at https://github.com/laoxie521/FSRNet.

Research on Precise Attitude Measurement Technology for Satellite Extension Booms Based on the Star Tracker

This paper reports the successful application of a self-developed, miniaturized, low-power nano-star tracker for precise attitude measurement of a 5-m-long satellite extension boom. Such extension booms are widely used in space science missions to extend and support payloads like magnetometers. The nano-star tracker, based on a CMOS image sensor, weighs 150 g (including the baffle), has a total power consumption of approximately 0.85 W, and achieves a pointing accuracy of about 5 arcseconds. It is paired with a low-cost, commercial lens and utilizes automated calibration techniques for measurement correction of the collected data. This system has been successfully applied to the precise attitude measurement of the 5-m magnetometer boom on the Chinese Advanced Space Technology Demonstration Satellite (SATech-01). Analysis of the in-orbit measurement data shows that within shadowed regions, the extension boom remains stable relative to the satellite, with a standard deviation of 30′′ (1σ). The average Euler angles for the “X-Y-Z” rotation sequence from the extension boom to the satellite are [−89.49°, 0.08°, 90.11°]. In the transition zone from shadow to sunlight, influenced by vibrations and thermal factors during satellite attitude adjustments, the maximum angular fluctuation of the extension boom relative to the satellite is approximately ±2°. These data and the accuracy of the measurements can effectively correct magnetic field vector measurements.

An Inverse recursive algorithm to retrieve the shape of the inaccessible dielectric objects

A regularized electromagnetic iterative inverse algorithm is formulated and implemented to reconstruct the shape of 2D dielectric objects using the far-field pattern of the scattered field data. To achieve this, an integral operator that maps the unknown boundary of the object onto the far-field pattern of the scattered field is defined and solved for the unknown boundary. The addressed inverse problem has an ill-posed nature and inherits nonlinearity. To overcome these, the proposed solution is linearized via Newton and regularized by Tikhonov in the sense of least squares. Besides, the dominance of the shadow region in the inverse-imaging process is exceeded by considering the superposition of multi-incoming plane waves, leading to less computational cost and a very fast inversion process. Comprehensive numerical analyses are carried out to ascertain the algorithm's feasibility, revealing that it is very efficient and promising.

Water ice particles detected by SELENE's Spectral Profiler at lunar shadowed regions in various local times and latitudes

Processes of water (OH and H2O) migration on the Moon remain unclear, prompting active research. Understanding lunar water migration requires investigation of the trapping and diffusion properties of water at various latitudes and local times. This study analyzed visible to near-infrared spectral data obtained by the Spectral Profiler (SP) onboard SELENE for shadowed regions at various local times and latitudes, not limited to the polar permanently shadowed regions. We assessed SP data for shadowed regions in 60 areas, each spanning a 10° × 10° latitude–longitude grid. Of the 1,061,907 analyzed shadowed-region data, 41,385 at various latitudes exhibited significant absorption in the 1.25 and 1.5 µm bands, indicating water ice particles. Data with the two absorption features suggest the presence of a water ice frost layer covering the lunar surface or suspended water ice particles above the lunar surface, at various latitude shadowed regions. Our spectral simulations have quantified the ice particles as being 0.1–1 µm in diameter, with a column density of 10–4–10–3 kg/m2. The spectral parameters for band absorption at the 1.5 µm band show symmetry between morning and evening sides, which is potentially attributed to the absence of variations in ice grain size and quantity. The 1.5 µm band absorption shows an increasing trend toward terminator regions, indicating variation in the water ice distribution and likely reflecting temperature conditions for water retention. The latitudinal trend of ice grain size and quantity remains uncertain because of the observed noise levels. Observations of water ice particles in shadowed regions at various latitudes and local times can provide new constraints on trapping and diffusion processes of lunar water migration.

An improved analytical method for rarefied aerothermodynamics on a complex geometry in free-molecular flows

A numerical method for high altitude aerothermodynamic analysis code for Very Low Earth Orbit (VLEO) satellite and spacecraft, named as HACS, is presented in this article. HACS is a computational tool that addresses the challenges by integrating the free-molecular gas kinetic model with novel particle sampling method and robust particle tracing algorithm to predict the shadow region on surfaces. To ensure the reliability and accuracy of the HACS, a rigorous verification and validation process has been systematically conducted by comparing the aerothermodynamic properties of forces, moments, and heat flux with the direct simulation Monte-Carlo (DSMC) results. The HACS is applied to the realistic geometries of VLEO satellite, the Apollo re-entry module, and spacecraft including fin and wing shapes. Results demonstrate that HACS provides the aerothermodynamic properties within the time frame of 60 seconds for the complex geometries and the accuracy of the HACS is guaranteed above 120 km altitude. The maximum relative error of the aerothermodynamic properties is less than 5% compared to the DSMC results at the lowest altitude of about 120 km, and has smaller relative errors as the altitude increases.

Reconfigurable intelligent surface assisted by computer vision for increasing coverage area in mobile communication systems

The paper describes a reflective intelligent surface (RIS) assisted by computer vision (CV). A RIS can reflect the transmitter’s signal into shadow regions with low signal strength thus increasing the coverage area of a communication system. CV assists the RIS in searching for and tracking potential users. Information about user’s position is used to direct the RIS beam towards the user. Operation of CV algorithms and their integration in the RIS control loop is demonstrated. Increase in the received signal power of 15–20 dB in non-line of sight (NLOS) situations due to RIS operation is achieved. The system has a low cost and power consumption.

Two-stage deep image restoration network with application to single image shadow removal

In this paper, we introduce a two-stage deep learning-based image restoration network and its application to remove shadow information from a single image, named by ESCNet (Encoder-decoder based Shadow removal with Colorization Network). Most existed single image-based shadow removal methods may suffer from that the shadow contains multiple regions of different colors or rich image details. To tackle with the problems, our key idea is to first remove shadow(s) from an image followed by repainting the shadow-removed region(s) in this image. To accomplish this, we present a deep two-stage network, cascading a shadow removal network (SRN) and a colorization network (CN). The presented encoder-decoder-based SRN with fusion of global and local feature information is used to remove the shadow(s) in the grayscale domain of the input image while recovering the image details for the shadow-removed region(s). Then the proposed CN aims at repainting the removed shadow region(s) via re-colorization. The proposed deep model has been well trained and well evaluated on the two well-known public datasets, i.e., ISTD (Image Shadow Triplets Dataset) and SRD (Shadow Removal Dataset). Experimental results have shown that the proposed method outperforms the compared state-of-the-art (SOTA) shadow removal approaches quantitatively and qualitatively.

Suppressed possibilistic fuzzy c-means clustering based on shadow sets for noisy data with imbalanced sizes

Possibilistic fuzzy c-means clustering (PFCM) algorithm generates fewer overlapping clustering centers than the possibilistic c-means clustering (PCM) algorithm and possesses better noise immunity than fuzzy c-means clustering (FCM) algorithm. However, with the increasing noise intensity and number of clusters, PFCM still faces the problem of getting partially overlapping clustering centers or mislocated centers in noise regions. Moreover, the sample-size imbalance increasingly intensifies the difficulty of positioning centers of small clusters. To solve the above problems, a suppressed possibilistic fuzzy c-means clustering algorithm based on shadow sets (S-SPFCM) is proposed by introducing the shadow set theory and the "suppressed competitive learning" strategy. Firstly, KL divergence is introduced in the objective function of PFCM to increase the anti-noise robustness of fuzzy memberships against long-range noise and outliers. Secondly, to reduce the number of overlapping centers caused by possibilistic memberships, the shadow set theory is introduced to divide each class adaptively into three regions (core, shadow, and exclusion regions) by an uncertainty balance method. The suppressed competitive learning method is extended by modifying the memberships of points within the three regions, thus artificially guiding the iterative track of clustering centers. Meanwhile, to further reduce the influence of imbalanced sizes, a scheme to reset mislocated centers in the core and shadow regions is also designed. In addition, to improve the segmentation performance of S-SPFCM for noisy images, a suppressed possibilistic fuzzy c-means clustering algorithm based on shadow sets and local information (SL-SPFCM) is also proposed. The SL-SPFCM first improves the Euclidean distance by a distance filtering scheme. Then SL-SPFCM introduces the local median membership of each pixel into the KL divergence of the objective function. Finally, experiments on synthetic datasets and color images which are characteristic of imbalanced sizes and noise injection demonstrate the proposed S-SPFCM and SL-SPFCM algorithms achieve smaller center deviations and higher clustering accuracy compared with several state-of-the-art clustering algorithms.

CRFormer: A cross-region transformer for shadow removal

Image shadow removal is a fundamental task in computer vision, which aims to restore damaged signals caused by shadows, thereby improving image quality and scene understanding. Recently, transformers have demonstrated strong capabilities in various applications by capturing global pixel interactions, a capability highly desirable for shadow removal. However, applying transformers to promote shadow removal is non-trivial for the following two reasons: 1) The patchify operation is not suitable for shadow removal due to irregular shadow shapes; 2) Shadow removal only requires one-way interaction from the non-shadow region to the shadow region instead of the common two-way interactions among all pixels in the image. In this paper, we propose a novel Cross-Region transFormer (CRFormer) for shadow removal which differs from existing transformers by only considering the pixel interactions from the non-shadow region to the shadow region without splitting images into patches. This is achieved by a carefully designed region-aware cross-attention mechanism that aggregates the recovered shadow region features, conditioned on the non-shadow region features. Extensive experiments on the ISTD, AISTD, SRD, and Video Shadow Removal datasets demonstrate the superiority of our method compared to other state-of-the-art methods.

Double-Exposure Algorithm: A Powerful Approach to Address the Accuracy Issues of Fractional Vegetation Extraction under Shadow Conditions

When recording the vegetation distribution with a camera, shadows can form due to factors like camera angle and direct sunlight. These shadows result in the loss of pixel information and texture details, significantly reducing the accuracy of fractional vegetation coverage (FVC) extraction. To address this issue, this study proposes an efficient double-exposure algorithm. The method reconstructs the pixel information in shadow areas by fusing normal-exposure and overexposed images. This approach overcomes the limitations of the camera’s dynamic range in capturing pixel information in shadowed regions. The study evaluates images with five levels of overexposure combined with five vegetation extraction indices. The aim is to determine the best-performing double-exposure combination under shadow conditions and the most suitable vegetation index. Experimental results reveal that the R² value between the best vegetation index and the FVC calculated from the fused double-exposure images and the ground truth FVC increases from 0.750 to 0.969. The root mean square error (RMSE) reduces from 0.146 to 0.046, and the intersection over union (IOU) increases from 0.856 to 0.943. These results demonstrate the excellent vegetation extraction capability of the double-exposure algorithm under shadow conditions, offering a straightforward and effective solution to low accuracy of FVC in shadowed areas.

Simulation of horizontal refraction of continental slope in the southern continental slope of the South China Sea

Abstract Sound waves are significantly influenced by boundaries during their propagation in certain environmental conditions. The extent of this impact is related to the complexity of the boundary, such as in the case of slopes and seamounts. In these areas, sound waves may deviate from their original paths, resulting in three-dimensional effects. Recent experiments and simulations have demonstrated that three-dimensional effects occur when sound waves propagate over seamounts in the South China Sea, leading to larger acoustic shadow regions. However, there are limited studies on 3D acoustic propagation based on measured slope topography or database topography. In this study, the topography of the South China Sea area database is selected, and the measured sound velocity profile and seabed acoustic parameters in the South China Sea are taken into considered. The FOR3D sound field calculation model is used to calculate the N×2D and 3D results. The simulation results show that in the southern continental slope region of the South China Sea, when the sound source frequency is 50 Hz, the acoustic wave exhibits significant horizontal refraction with a refraction angle of approximately 13°.

A new model for high-accuracy monitoring of water level changes via enhanced water boundary detection and reliability-based weighting averaging

Accurate measurement of water levels is essential for effectively managing reservoirs to proactively mitigate flooding and drought. Nonetheless, the inaccuracies in measurements derived from gauging station and remote sensing images impose constraints to water resource management. In this study, we developed a novel water level estimation model which utilizes solely the altitude of reliable water boundary pixels to improve the accuracy. The enhanced water boundary detection, incorporating preprocessing steps such as image filtering, resampling, and polarization multiplication, was applied to achieve sub-pixel precision in detecting water boundaries. The water boundary pixels located in layover and shadow regions, which could be misidentified due to distortion error, are eliminated based on backward geolocation. Ambiguous water boundaries, potentially indicating land with low intensity, were defined by computing their absolute derivatives, and removed. Finally, to enhance water level precision, the model computed water levels by averaging the altitudes of boundary pixels with weighting factors of local incidence angle, derivatives of detected water boundaries, and altitude distribution. Compared with the previous studies utilizing water boundaries, the proposed model demonstrated outstanding performance in improving the accuracy, up to 1/40th smaller than the spatial resolution of SAR images in Mean Absolute Error (MAE). The validation was executed over the results from >700 Sentinel-1 against the in-situ measurements obtained from multiple reservoirs and streams with significant water level fluctuations in the Korean peninsula. In this process, we found that the water boundaries located on the layover and shadow regions significantly influence the dispersion of altitude of reliable water boundary pixels. This study demonstrates the proposed model relying on remote sensing data without in-situ measurements, which holds potential applicability under situations where in-situ data are unavailable.

Formation of an Extended Equatorial Shadow Zone for Low‐Frequency Saturn Kilometric Radiation

AbstractSaturn Kilometric Radiation (SKR), being the dominant radio emission at Saturn, has been extensively investigated. The low‐frequency extension of SKR is of particular interest due to its strong association with Saturn's magnetospheric dynamics. However, the highly anisotropic beaming of SKR poses challenges for observations. In most cases, the propagation of SKR is assumed to follow straight‐line paths. We explore the propagation characteristics of SKR across different frequencies in this study. An extended equatorial shadow region for low‐frequency SKR is identified, resulting from the merging of the Enceladus plasma torus and the previously known equatorial shadow zone. Ray‐tracing simulations reveal that low‐frequency (100 kHz) SKR is unable to enter the shadow region and is instead reflected toward high latitudes. In contrast, high‐frequency SKR (100 kHz) generally propagates without hindrance. Observations suggest that some low‐frequency SKR can enter the shadow region through reflection by the magnetosheath or leakage from the plasma torus.

Shadow-constrained shape-from-shading for pixel-wise 3D surface reconstruction at the lunar south pole

ABSTRACT High-resolution digital elevation models (DEMs) of the lunar surface are crucial for lunar exploration missions and scientific research. The emergence of advanced technologies, such as Shape-from-Shading (SfS), has enabled the generation of pixel-wise high-resolution DEMs from monocular images of the lunar surface. However, SfS encounters significant challenges in locations with limited illumination, such as the lunar south pole, where the surface is largely covered by shadows. Therefore, this paper presents a novel shadow-constrained SfS approach for pixel-wise 3D surface reconstruction at the lunar south pole. The proposed approach uses multiple high-resolution images captured under different illumination conditions and an existing low-resolution DEM as the inputs and generates a high-resolution DEM with the same resolution as that of the input image through hierarchical SfS processing incorporating shadow constraints. Experiments were conducted using actual images collected by the Lunar Reconnaissance Orbiter (LRO) Narrow Angle Camera (NAC) at the lunar south pole. Comparisons with respect to photogrammetric DEMs generated from stereo NAC images show that the DEMs generated using the proposed approach exhibit the smallest root mean square error (RMSE). Moreover, shaded relief images rendered from the DEMs generated using the proposed approach demonstrate the highest similarity to the actual NAC images. Detailed profile comparisons further validate the effectiveness of the shadow constraint in optimizing 3D reconstruction in regions in proximity to shadows and within shadowed regions. The proposed shadow-constrained SfS approach can be used to generate high-resolution DEMs to support future missions to explore the lunar south pole, with applications including landing site evaluation and route planning for lunar probes or astronauts.

MixImages: An Urban Perception AI Method Based on Polarization Multimodalities.

Intelligent urban perception is one of the hot topics. Most previous urban perception models based on semantic segmentation mainly used RGB images as unimodal inputs. However, in natural urban scenes, the interplay of light and shadow often leads to confused RGB features, which diminish the model's perception ability. Multimodal polarization data encompass information dimensions beyond RGB, which can enhance the representation of shadow regions, serving as additional data for assistance. Additionally, in recent years, transformers have achieved outstanding performance in visual tasks, and their large, effective receptive field can provide more discriminative cues for shadow regions. For these reasons, this study proposes a novel semantic segmentation model called MixImages, which can combine polarization data for pixel-level perception. We conducted comprehensive experiments on a polarization dataset of urban scenes. The results showed that the proposed MixImages can achieve an accuracy advantage of 3.43% over the control group model using only RGB images in the unimodal benchmark while gaining a performance improvement of 4.29% in the multimodal benchmark. Additionally, to provide a reference for specific downstream tasks, we also tested the impact of different combinations of polarization types on the overall segmentation accuracy. The proposed MixImages can be a new option for conducting urban scene perception tasks.

Shadow-aware decomposed transformer network for shadow detection and removal

Shadow detection and removal are important yet challenging computer vision tasks. Existing methods simultaneously contend with the brightness and color information of input image while treating different regions of input images equally. We argue that these operations fail to effectively extract the relationship between shadow and non-shadow regions. To relieve these problems, this paper proposes a shadow-aware decomposed transformer network for shadow detection and removal. The network decomposes the input image into brightness and color maps using its bright channel, which are concatenated with the original image as the input data, enabling the network to pay balanced attention to the brightness and color information when calculating the relationship between regions. Additionally, given that the correlation matrix of the transformer measures the relative dependency between regions, the proposed shadow-aware transformer block can extract the relationship between shadow and non-shadow regions more effectively by retaining the specific elements of the correlation matrix. We conduct extensive experiments on three shadow detection benchmark datasets and two shadow removal benchmark datasets. Experimental results show that the proposed method performs favorably against state-of-the-art methods. Codes have been made available at https://github.com/XIAOWANG914/SADT.

Shadow removal method for high-resolution aerial remote sensing images based on region group matching

Shadow removal is a critical step in preprocessing high-resolution aerial remote sensing images. The presence of shadows reduces the information in high-resolution aerial remote sensing images, lowers the image quality, and seriously interferes with the work of feature classification, object detection, and information extraction. Therefore, this paper proposes a high-resolution aerial remote sensing image shadow removal method based on region group matching. The method starts from both spatial information and color information, and is able to take into account the whole and the local, and recover the detailed features of different features inside the shadows. Firstly, this paper proposes an irregular region color transfer method based on three-dimensional color space, which can effectively restore the original color and texture information of the features in the shadow region. Secondly, considering the random and variable direction of feature distribution in remote sensing images, the direction of texture feature extraction cannot be set in advance. Therefore, in this paper, the texture details of the features are extracted based on the rotationally invariant light-independent texture feature extraction method to exclude the abnormal interference. Then, the shadow and light regions of the image are segmented internally and grouped according to the type of features to avoid subsequent matching anomalies due to the small size of the image blocks. Next, construct an average texture feature vector for each group of image blocks for grouping matching. The matched light groups are then used to guide the shadow groups for local feature information enhancement, which ensures that on top of the overall shadow removal, the detail recovery of different local features is enhanced. Finally, for the boundary effect after shadow removal, a boundary optimization method with dynamically assigned weights is proposed, which can ensure the natural coherence of the resultant image in the boundary part. The experimental results show that the proposed method balances both overall and local shadow removal effects and can effectively achieve high-resolution aerial remote sensing image shadow removal. Compared with existing methods, the proposed method achieves the best quality of shadow removal results and restores more realistic ground features in shadow region.

Robot-LDA Plane Measurements And Smoke Investigations On A 30° Inclined Flat Plate In An Open Wind Channel With And Without Plasma Actuation Effect

Active flow control is a relatively new approach to influence the velocity vector field for plane airfoils or process engineering applications with a plasma source. The goal is to positively effect the point of separation to increase the lift/drag ratio or increase separation efficiency and to reduce pressure drop. In the presented study two approaches, the Robot-LDA measurements and smoke investigations, are chosen to find out whether the plasma source has an effect on the velocity vector field on a 30°inclined flat plate and the separation point. A flow rate of 250 m³/h resulting in a Reynolds number of 100 000 is chosen, since this is typical for already presented results within the field of plasma source investigations. With Robot-LDA in total 825 measurements points of the axial wind channel velocity components are captured twice, with and without activated plasma source. The measurement plane was set 80 mm x 50 mm orthogonal to the flat plate in the middle of the wind channel axis. The plasma source was powered with 14 kVp−p and 5 kHz. Due to stability of the plasma actuator, the LDA measurements were subdivided into eight regions. For the same configurations smoke investigations are performed. A chromium-nickel wire with a diameter of 0.213 mm is heated by an electronic circuit including capacitor and power source to vaporize liquid paraffin. The flow field is captured by a highspeed camera. Robot-LDA measurement results show a free-stream axial wind channel velocity component of about 15 m/s. In the shadow region of the 30°inclined flat plate a recirculation zone is found due to negative velocity components. At a distance up to 12 mm close to the plat plate, wall reflections dominated and to find valid LDA measurement results were not possible. With and without activated plasma source velocity differences between -0.5 m/s and 0.5 m/s are found. The biggest velocity differences are in the region of 15 mm to 45 mm relative to the tip of the inclined flat plate. For post-processing purposes, the 3D wind channel geometry and the Robot-LDA measurement results are loaded in ParaView 5.10.1. Smoke investigations show that the plasma source minimally attach the flow to the flat plate. This effect is higher with smaller flow rates (50 m³/h) than with 250 m³/h. Robot-LDA and smoke measurement results are promising that flow influence due to plasma has potential in various applications in future.

Global Attention-Based DEM: A Planet Surface Digital Elevation Model-Generation Method Combined with a Global Attention Mechanism

Digital elevation models (DEMs), which can provide an accurate description of planetary surface elevation changes, play an important role in scientific tasks such as long-distance path planning, terrain analysis, and planetary surface reconstruction. However, generating high-precision planetary DEMs currently relies on expensive equipment together with complex remote sensing technology, thus increasing the cost and cycle of the task. Therefore, it is crucial to develop a cost-effective technology that can produce high-quality DEMs on the surfaces of planets. In this work, we propose a global attention-based DEM generation network (GADEM) to convert satellite imagery into DEMs. The network uses the global attention mechanism (GAM) together with a multi-order gradient loss function during training to recover precise terrain. The experimental analysis on lunar and Martian datasets not only demonstrated the effectiveness and accuracy of GADEM in bright regions, but also showed its promising reconstruction ability in shadowed regions.