Search Results (32)

Monocular depth estimation (MDE) is a critical computer vision task that enhances environmental perception in fields such as autonomous driving and robot navigation. In recent years, deep learning-based MDE methods have achieved notable progress in these fields. However, achieving robust monocular depth estimation in low-altitude forest environments remains challenging, particularly in scenes with dense and cluttered foliage, which complicates applications in environmental monitoring, agriculture, and search and rescue operations. This paper presents a comprehensive evaluation of state-of-the-art deep learning-based MDE methods on low-altitude forest datasets. The evaluated models include both self-supervised and supervised approaches, employing different network structures such as convolutional neural networks (CNNs) and Vision Transformers (ViTs). We assessed the generalization of these approaches across diverse low-altitude scenarios, specifically focusing on forested environments. A systematic set of evaluation criteria is employed, comprising traditional image-based global statistical metrics as well as geometry-aware metrics, to provide a more comprehensive evaluation of depth estimation performance. The results indicate that most Transformer-based models, such as DepthAnything and Metric3D, outperform traditional CNN-based models in complex forest environments by capturing detailed tree structures and depth discontinuities. Conversely, CNN-based models like MiDas and Adabins struggle with handling depth discontinuities and complex occlusions, yielding less detailed predictions. On the Mid-Air dataset, the Transformer-based DepthAnything demonstrates a 54.2% improvement in RMSE for the global error metric compared to the CNN-based Adabins. On the LOBDM dataset, the CNN-based MiDas has the depth edge completeness error of 93.361, while the Transformer-based Metric3D demonstrates the significantly lower error of only 5.494. These findings highlight the potential of Transformer-based approaches for monocular depth estimation in low-altitude forest environments, with implications for high-throughput plant phenotyping, environmental monitoring, and other forest-specific applications. Full article

Background: Emphysematous pyelonephritis (EPN) is an infectious disease of the renal system caused by gas-producing microorganisms harboring the kidneys. Patients with diabetes mellitus (DM), an endocrine disease with hyperglycemia, are particularly susceptible to the EPN as their immune system is compromised in fighting against infections. Case Description: We present a case of a 50-year-old female with a history of chronic diabetes and persistent hypertension. She presented with symptoms of pyrexia and flank pain. Following findings from ultrasound, she was advised to undergo computed tomographic (CT) scans that reveal air-filled hypodense areas at the upper and mid pole of the right kidney and in the renal pelvis of the right ureter, which confirms the class I EPN in the patient. Urine culture identifies Escherica coli as the causative agent for EPN. The patient was managed with third-generation antibiotics over two weeks, leading to full recovery without surgical intervention. Discussion: The availability of CT imaging makes early diagnosis and reduces mortality associated with EPN. Conservative medical management should be the initial treatment strategy for EPN. However, severe cases require immediate therapeutic action. In our case, the patient was treated with antibiotic therapy and recovered. Conclusions: CT scan seems to be the optimal diagnosis in patients with acute emphysematous pyelonephritis. Patients with EPN class I respond well to medical treatment with excellent outcomes. Full article

When implementing outside-the-window (OTW) visuals in aviation tactical simulators, maintaining terrain image color consistency is critical for enhancing pilot immersion and focus. However, due to various environmental factors, inconsistent image colors in terrain can cause visual confusion and diminish realism. To address these issues, a color correction technique based on a Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP) is proposed. The proposed WGAN-GP model utilizes multi-scale feature extraction and Wasserstein distance to effectively measure and adjust the color distribution difference between the input image and the reference image. This approach can preserve the texture and structural characteristics of the image while maintaining color consistency. In particular, by converting Bands 2, 3, and 4 of the BigEarthNet-S2 dataset into RGB images as the reference image and preprocessing the reference image to serve as the input image, it is demonstrated that the proposed WGAN-GP model can handle large-scale remote sensing images containing various lighting conditions and color differences. The experimental results showed that the proposed WGAN-GP model outperformed traditional methods, such as histogram matching and color transfer, and was effective in reflecting the style of the reference image to the target image while maintaining the structural elements of the target image during the training process. Quantitative analysis demonstrated that the mid-stage model achieved a PSNR of 28.93 dB and an SSIM of 0.7116, which significantly outperforms traditional methods. Furthermore, the LPIPS score was reduced to 0.3978, indicating improved perceptual similarity. This approach can contribute to improving the visual elements of the simulator to enhance pilot immersion and has the potential to significantly reduce time and costs compared to the manual methods currently used by the Republic of Korea Air Force. Full article

In recent years, there has been extensive research and application of unsupervised monocular depth estimation methods for intelligent vehicles. However, a major limitation of most existing approaches is their inability to predict absolute depth values in physical units, as they generally suffer from the scale problem. Furthermore, most research efforts have focused on ground vehicles, neglecting the potential application of these methods to unmanned aerial vehicles (UAVs). To address these gaps, this paper proposes a novel absolute depth estimation method specifically designed for flight scenes using a monocular vision sensor, in which a geometry-based scale recovery algorithm serves as a post-processing stage of relative depth estimation results with scale consistency. By exploiting the feature correspondence between successive images and using the pose data provided by equipped navigation sensors, the scale factor between relative and absolute scales is calculated according to a multi-view geometry model, and then absolute depth maps are generated by pixel-wise multiplication of relative depth maps with the scale factor. As a result, the unsupervised monocular depth estimation technology is extended from relative depth estimation in semi-structured scenes to absolute depth estimation in unstructured scenes. Experiments on the publicly available Mid-Air dataset and customized data demonstrate the effectiveness of our method in different cases and settings, as well as its robustness to navigation sensor noise. The proposed method only requires UAVs to be equipped with monocular camera and common navigation sensors, and the obtained absolute depth information can be directly used for downstream tasks, which is significant for this kind of vehicle that has rarely been explored in previous depth estimation studies. Full article

With the increasing demand for unmanned aerial vehicles (UAVs), the number of UAVs in the airspace and the risk of mid-air collisions caused by UAVs are increasing. Therefore, detect and avoid (DAA) technology for UAVs has become a crucial element for mid-air collision avoidance. This study presents a collision avoidance approach for UAVs equipped with a monocular camera to detect small fixed-wing intruders. The proposed system can detect any size of UAV over a long range. The development process consists of three phases: long-distance object detection, object region estimation, and collision risk assessment and collision avoidance. For long-distance object detection, an optical flow-based background subtraction method is utilized to detect an intruder far away from the host. A mask region-based convolutional neural network (Mask R-CNN) model is trained to estimate the region of the intruder in the image. Finally, the collision risk assessment adopts the area expansion rate and bearing angle of the intruder in the images to conduct mid-air collision avoidance based on visual flight rules (VFRs) and conflict areas. The proposed collision avoidance approach is verified by both simulations and experiments. The results show that the system can successfully detect different sizes of fixed-wing intruders, estimate their regions, and assess the risk of collision at least 10 s in advance before the expected collision would happen. Full article

Evapotranspiration (ET) is an important channel for water transport and energy conversion in land–air systems, and the spatial quantification of actual ET is crucial for water resource management and scheduling in arid areas. Using the Surface Energy Balance Algorithm for Land (SEBAL) model and satellite images, this study determined the actual ET during the growing season of 2020 in the Shiyang River Basin of northwest China and investigated the driving mechanism of ET using a principal component regression. The results showed that the ET obtained using the Penman-Monteith equation exhibited a good correlation with the ET estimated using SEBAL (R² = 0.85). Additionally, SEBAL overestimated ET to some extent compared to the Moderate-Resolution Imaging Spectroradiometer (MODIS) ET (MOD16) product. The daily ET (ET_d) in the Shiyang River Basin showed a single-peak variation during the growing season, with the maximum value occurring around mid-July. Spatially, the ET gradually increased from northeast to southwest with the variation in the land use/land cover (LULC) type. Among the six LULC types, ET_d was higher for woodland, water body, and grassland, all exceeding 5.0 mm/d; farmland and built-up land had ET_d close to 3.9 mm/d; and barren land had the lowest ET_d of below 2.5 mm/d. Furthermore, the standardized regression coefficients indicated that the Normalized Difference Vegetation Index (NDVI) is the main driving factor influencing ET. Overall, the SEBAL model has the potential to estimate spatially actual ET, and the study results provide a scientific basis for water resource accounting and hydrological analysis in arid areas. Full article

The Harmony satellite mission was recently approved as the next European Space Agency (ESA) Earth Explorer 10. The mission science objectives cover several applications related to solid earth, the cryosphere, upper-ocean dynamics and air–sea interactions. The mission consists of a constellation of two satellites, flying with the Copernicus Sentinel 1 (C or D) spacecraft, each hosting a C-band receive-only radar and a thermal infrared (TIR) payload. From an ocean dynamics/air–sea interaction perspective, the mission will provide the unique opportunity to observe simultaneously the signature of submesoscale upper-ocean processes via synthetic aperture radar and TIR imagery. The TIR imager is based on microbolometer technology and its acquisitions will rely on four channels: three narrow-band channels yielding observations at a ≃1 km spatial sampling distance (SSD) and a panchromatic (PAN, 8–12 $\mathsf{\mu }$m) channel characterized by a ≃300 m SSD. Our study investigates the potential of Harmony in retrieving spatial features related to sea surface temperature (SST) gradients from the high-resolution PAN channel, relying on top-of-atmosphere (TOA) observations. Compared to a standard SST gradient retrieval, our approach does not require atmospheric correction, thus avoiding uncertainties due to inter-channel co-registration and radiometric consistency, with the possibility of exploiting the higher resolution of the PAN channel. The investigations were carried out simulating the future Harmony TOA radiances (TARs), as well as relying on existing state-of-the-art level 1 satellite products. Our approach enables the correct description of SST features at the sea surface avoiding the generation of spurious features due to atmospheric correction and/or instrumental issues. In addition, analyses based on existing satellite products suggest that the clear-sky TOA observations, in a typical mid-latitude scene, allow the reconstruction of up to 85% of the gradient magnitudes found at the sea-surface level. The methodology is less efficient in tropical areas, suffering from smoothing effects due to the high concentrations of water vapor. Full article

Background: This study aimed at identifying errors encountered in orthopantomography (OPG) in post-traumatic patients caused by limitations in performing a correct technique. Methods: A retrospective observational study was performed. Diagnosis, exposure/processing mistakes, positioning-related errors, and bimaxillary immobilization were evaluated. Results: Thirty panoramic radiographs with mandible fractures were examined. Twelve error types were encountered: errors in exposure or processing, air radiolucency in the palatoglossal space, errors in the alignment of the Frankfort horizontal plane: head in flexion, with a joyful expression or head extended, with a somber appearance, errors towards the mid-sagittal plane (lateral head inclination, deviation, or rotation), errors caused by the non-use of the bite-block or inappropriate position on the device, errors caused by positioning outside the focal plane, artifacts/shadow images produced by post-operative metal plates, and bimaxillary immobilization errors. The number of errors per radiograph ranged from two to a maximum of five. The most dominant ones were inappropriate alignment in the focal plane and lateral rotation of the head in over 70% of cases. Lateral deviation and palatoglossal air were present in more than 50% of images. Conclusions: In trauma cases, technical difficulties in obtaining a proper OPG image are common and often insurmountable, limiting the diagnosis. Full article

Photovoltaic plants are helping to reduce CO₂ emissions, but the energy performance of photovoltaic systems must remain high throughout their operational life. Supervision and monitoring are mandatory for large photovoltaic plants because failures can cause high power losses due to the large number of photovoltaic modules. Infrared analysis is effective and reliable in detecting anomalies or failures in photovoltaic modules, but it is time-consuming and expensive when the infrared inspection of large photovoltaic plants is manual. Nowadays, the diffusion of unmanned aerial vehicles equipped with infrared cameras can support the fast supervision of photovoltaic plants. Nevertheless, the use of drones is regulated by international and national rules; consequently, it is not always possible to use a drone, or its utilization is limited based on geographic areas and/or authorizations. Moreover, infrared analysis requires additional requirements when done by drone, because the mutual position between the photovoltaic modules and the infrared camera affects the goodness of the infrared acquisition. This article discusses these critical issues, directs the reader to official, national, and geographic maps for drones, and suggests technical solutions for some specific issues not considered in the technical specification for the outdoor infrared thermography of photovoltaic modules. In particular, the paper proposes a systematic procedure for the legal and effective infrared inspection of photovoltaic modules by means of a drone and proposes improvements for some issues not discussed in the international rules: the correction of infrared images with respect to the view angle, the impact of a mid-wave and long-wave infrared sensor on the acquired image, and the impact of air transmittance. Full article

We studied whether TOF reduces error propagation from attenuation correction to PET image reconstruction in PET/MR neuroimaging, by using imperfect attenuation maps in a clinical PET/MR system with 525 ps timing resolution. Ten subjects who had undergone ¹⁸F-FDG PET neuroimaging were included. Attenuation maps using a single value (0.100 cm⁻¹) with and without air, and a 3-class attenuation map with soft tissue (0.096 cm⁻¹), air and bone (0.151 cm⁻¹) were used. CT-based attenuation correction was used as a reference. Volume-of-interest (VOI) analysis was conducted. Mean bias and standard deviation across the brain was studied. Regional correlations and concordance were evaluated. Statistical testing was conducted. Average bias and standard deviation were slightly reduced in the majority (23–26 out of 35) of the VOI with TOF. Bias was reduced near the cortex, nasal sinuses, and in the mid-brain with TOF. Bland–Altman and regression analysis showed small improvements with TOF. However, the overall effect of TOF to quantitative accuracy was small (3% at maximum) and significant only for two attenuation maps out of three at 525 ps timing resolution. In conclusion, TOF might reduce the quantitative errors due to attenuation correction in PET/MR neuroimaging, but this effect needs to be further investigated on systems with better timing resolution. Full article

Detection of small moving objects in long range infrared (IR) videos is challenging due to background clutter, air turbulence, and small target size. In this paper, we present two unsupervised, modular, and flexible frameworks to detect small moving targets. The key idea was inspired by change detection (CD) algorithms where frame differences can help detect motions. Our frameworks consist of change detection, small target detection, and some post-processing algorithms such as image denoising and dilation. Extensive experiments using actual long range mid-wave infrared (MWIR) videos with target distances beyond 3500 m from the camera demonstrated that one approach, using Local Intensity Gradient (LIG) only once in the workflow, performed better than the other, which used LIG in two places, in a 3500 m video, but slightly worse in 4000 m and 5000 m videos. Moreover, we also investigated the use of synthetic bands for target detection and observed promising results for 4000 m and 5000 m videos. Finally, a comparative study with two conventional methods demonstrated that our proposed scheme has comparable performance. Full article

Planet, a commercial company, has achieved a key milestone by launching a large fleet of small satellites (smallsats) that provide high spatial resolution imagery of the entire Earth’s surface on a daily basis with its PlanetScope sensors. Given the potential utility of these data, this study explores the use for fine particulate matter (PM_2.5) air quality applications. However, before these data can be utilized for air quality applications, key features of the data, including geolocation accuracy, calibration quality, and consistency in spectral signatures, need to be addressed. In this study, selected Dove-Classic PlanetScope data is screened for geolocation consistency. The spectral response of the Dove-Classic PlanetScope data is then compared to Moderate Resolution Imaging Spectroradiometer (MODIS) data over different land cover types, and under varying PM_2.5 and mid visible aerosol optical depth (AOD) conditions. The data selected for this study was found to fall within Planet’s reported geolocation accuracy of 10 m (between 3–4 pixels). In a comparison of top of atmosphere (TOA) reflectance over a sample of different land cover types, the difference in reflectance between PlanetScope and MODIS ranged from near-zero (0.0014) to 0.117, with a mean difference in reflectance of 0.046 ± 0.031 across all bands. The reflectance values from PlanetScope were higher than MODIS 78% of the time, although no significant relationship was found between surface PM_2.5 or AOD and TOA reflectance for the cases that were studied. The results indicate that commercial satellite data have the potential to address Earth-environmental issues. Full article

A feasibility study of velocity field measurements using the Particle Image Velocimetry (PIV) method in an axial air turbine model is presented. The wakes past the blades of the rotor wheel were observed using the PIV technique. Data acquisition was synchronized with the shaft rotation; thus, the wakes were phase averaged for statistical analysis. The interaction of the rotor blade wakes with the stator ones was investigated by changing the stator wheel’s angle. The measurement planes were located just behind the rotor blades, covering approximately 3 cm × 3 cm in axial × tangential directions. The spatial correlation function suggests that the resolution used is sufficient for the large-scale flow-patterns only, but not for the small ones. The scales of fluctuations correspond to the shear layer thickness at the mid-span plane but, close to the end-wall, they contain larger structures caused by the secondary flows. The length-scales of the fluctuations under off-design conditions display a dependence on the area of the stator and rotor wakes cross-sections, which, in turn, depend on their angle. The obtained experimental data are to be used for the validation of mathematical simulation results in the future. Full article

The cloud-top-phase climatology over the western North Pacific (WNP) has received little attention. Using 3 years (2017–2019) of cloud-top-phase products from the Advanced Himawari Imager onboard the Japanese Himawari-8 satellite, this study examines the seasonal and diurnal variations in the cloud-top phase over the WNP. Results show that over the low- and mid-latitude maritime regions, ice (water) clouds occur more (less) frequently during boreal winter than summer. Water clouds are more likely to be related to moisture conditions in the lower troposphere than to the underlying sea surface temperature. Owing to the combined effects of moist air mass transport and ocean currents (topography), the WNP region east of Hokkaido (the Sichuan Basin) has a high frequency of water clouds in summer (winter). Furthermore, supercooled water cloud populations have a clear seasonal cycle. The fraction of water clouds that are supercooled appears to be modulated by the near-surface air temperature. A diurnal cycle is seen in ice-cloud populations, which are highest in the late afternoon over both ocean and land except for the Sichuan Basin where summer nocturnal precipitation is typical. The occurrences of continental water clouds peak at noon in summer but early morning (around sunrise) in winter. An increase in the frequency of continental summer water clouds around noon is found to be associated with variations in both the cloud-top elevation of already-existing water clouds and new formations of boundary-layer clouds. Full article

The paper presents an analysis and classification method to evaluate the working condition of angle grinders by means of infrared (IR) thermography and IR image processing. An innovative method called BCAoMID-F (Binarized Common Areas of Maximum Image Differences—Fusion) is proposed in this paper. This method is used to extract features of thermal images of three angle grinders. The computed features are 1-element or 256-element vectors. Feature vectors are the sum of pixels of matrix V or PCA of matrix V or histogram of matrix V. Three different cases of thermal images were considered: healthy angle grinder, angle grinder with 1 blocked air inlet, angle grinder with 2 blocked air inlets. The classification of feature vectors was carried out using two classifiers: Support Vector Machine and Nearest Neighbor. Total recognition efficiency for 3 classes (TR_AG) was in the range of 98.5–100%. The presented technique is efficient for fault diagnosis of electrical devices and electric power tools. Full article