Aircraft Images Research Articles

Multi-reference source aircraft classification network based on unknown class prototype estimation

Abstract Gathering complete aircraft types for close-set tasks is challenging and costly in fine-grained aircraft classification, resulting in encountering unknown class aircraft images in real-world models. To address this problem, we propose a network called a multi-reference source fine-grained aircraft classification network (MRSN) to explicitly and implicitly distinguish known class boundaries and embed unknown classes into the sample space. Specifically, we propose an embedding module (UCPE) to synthesize an unknown class prototype to facilitate the modeling of unknown class data distribution. Besides, we introduce a concatenated multi-reference source module (CMS) that combines distance measurement and probability estimation to classify whether testing images are of unknown class or not. Our proposed MRSN innovatively extends the open-set learning paradigm to the fine-grained classification of geospatial objects rather than simple scene classification. Besides, it shows excellent performance in our constructed dataset, which demonstrates the effectiveness and progressiveness of our method.

Imaging with an inverse-designed 50 mm-diameter f/1 MWIR flat lens with enhanced field of view and depth of focus.

We utilize inverse design and grayscale optical lithography to create a flat lens with a diameter and focal length of 50 mm, operating in the mid-wavelength infrared (MWIR) band. This lens demonstrates an extended depth of focus (DOF ≥±100μm), a field of view (FOV ≥20°), and an angular resolution of 300μrad. We characterize the lens's performance and use it as the primary optic in a hybrid refractive-diffractive telescope, which increases the angular resolution to 160μrad. Using this telescope, we perform video imaging of aircraft and vehicles. Our experiments were constrained by the higher f-number of the focal plane array. Nonetheless, through rigorous simulations, we demonstrate that the inverse-designed flat lens surpasses the performance of a conventional Fresnel zone plate (FZP) in DOF and in FOV, even under these limitations. The flat lens, weighing approximately 20g, is significantly lighter than its refractive counterparts, confirming the feasibility of high-resolution, lightweight MWIR imaging systems.

Willingness to Fly on Urban Air Mobility Aircraft in Canada

Urban air mobility (UAM) is projected to be a major part of the future transportation landscape. It has the potential to reduce trip times and is viewed as a low-emission option. However, little is known about the public’s willingness to fly in these vehicles. In this study, a survey was conducted to assess willingness to fly on an air taxi. The survey of 1022 Canadians found that 39.4% were willing to fly after being shown an image of a specific UAM concept aircraft and given a short description about UAM, with another 29.1% stating that they would consider it. Factors leading to a higher level of support included gender, age, career, previous flight behavior, and support for other novel technologies. These results indicate that there is a desire for short-distance air travel, and the recommended next steps include assessing infrastructure needs and investigating regulatory and certification requirements for the introduction of UAM.

Semi-supervised few-shot class-incremental learning based on dynamic topology evolution

Incremental learning of new classes is crucial to developing real-world artificial intelligence systems. However, in some cases, we can only use a limited number of images to train the incremental model due to the difficulty of obtaining them. Few-shot class-incremental learning (FSCIL) is a new paradigm of incremental learning, which continually learns new concepts from a few examples. This setup can quickly lead to catastrophic forgetting and overfitting problems, severely affecting model performance. To address these problems, we propose a dynamic topological evolution (DTE) model that unifies knowledge of old and new categories. In this work, we focus on a novel semi-supervised few-shot class-incremental learning (SSFSCIL) to incrementally learn new categories with a limited number of labeled samples and a large number of unlabeled samples. In detail, we use the Voronoi polygon criterion to divide the feature space and construct a Hebbian graph to represent topological relationships between classes. Then, the corresponding k-nearest neighbor subgraph is constructed for the new incremental task, and label propagation of graph nodes is carried out on the unlabeled set to generate the corresponding pseudo labels. Finally, the new category knowledge is evolved into the actual category knowledge by integrating the Hebbian graph and the k-nearest neighbor subgraph obtained in the new incremental task. Extensive experiments on the mini-ImageNet and CIFAR100 datasets show that the accuracy of the DTE outperforms the baseline and improves by 1.41% and 1.13% on the last incremental task. Furthermore, our methodology achieved a 35.6% time improvement compared to full-supervised methods on the CIFAR100 dataset. In addition to the experiments on the two datasets mentioned above, the validation of our proposed DTE model extends to the engineering application domain of the RESICS-45 aircraft image scene classification dataset.

SYNTHESIS MODELS OF LAYOUT SCHEME IN THE TASK OF FORMING AN AIRCRAFT IMAGE

SYNTHESIS MODELS OF LAYOUT SCHEME IN THE TASK OF FORMING AN AIRCRAFT IMAGE

Quality Analysis of Unmanned Aerial Vehicle Images Using a Resolution Target

Unmanned aerial vehicle (UAV) photogrammetry is an emerging means of acquiring high-precision rapid spatial information and data because it is cost-effective and highly efficient. However, securing uniform quality in the results of UAV photogrammetry is difficult due to the use of low-cost navigation devices, non-surveying cameras, and rapid changes in shooting locations depending on the aircraft’s behavior. In addition, no specific procedures or guidelines exist for performing quantitative quality tests or certification methods on UAV images. Additionally, test tools for UAV image quality assessment only use the ground sample distance (GSD), often resulting in a reduced image quality compared with that of manned aircraft images. In this study, we performed a modulation transfer function (MTF) analysis using a slanted edge target and a GSD analysis to confirm the necessity of MTF analysis in UAV image quality assessments. In this study, we aimed to address this issue by conducting a modulation transfer function (MTF) analysis using a slanted edge target and a ground sample distance (GSD) analysis. This was carried out to confirm the necessity of MTF analysis in evaluating UAV image quality. Furthermore, we analyzed the impact of flight height and mounted sensors on image quality at different study sites.

Mathematical modeling of optical images of high-speed aircraft in the IR range

Mathematical modeling of optical images of high-speed aircraft in the IR range

Autonomous Identification of Bridge Concrete Cracks Using Unmanned Aircraft Images and Improved Lightweight Deep Convolutional Networks

The identification of the development of structural defects is an important part of bridge structure damage diagnosis, and cracks are considered the most typical and highly dangerous structural disease. However, existing deep learning‐based methods are mostly aimed at the scene of concrete cracks, while they rarely focus on designing network architectures to improve the vision‐based model performance from the perspective of unmanned aircraft system (UAS) inspection, which leads to a lack of specificity. Because of this, this study proposes a novel lightweight deep convolutional neural network‐based crack pixel‐level segmentation network for UAS‐based inspection scenes. Firstly, the classical encoder‐decoder architecture UNET is utilized as the base model for bridge structural crack identification, and the hourglass‐shaped depthwise separable convolution is introduced to replace the traditional convolutional operation in the UNET model to reduce model parameters. Then, a kind of lightweight and efficient channel attention module is used to improve model feature fuzzy ability and segmentation accuracy. We conducted a series of experiments on bridge structural crack detection tasks by utilizing a long‐span bridge as the research item. The experimental results show that the constructed method achieves an effective balance between reasoning accuracy and efficiency with the value of 97.62% precision, 97.23% recall, 97.42% accuracy, and 93.25% IOU on the bridge concrete crack datasets, which are significantly higher than those of other state‐of‐the‐art baseline methods. It can be inferred that the application of hourglass‐shaped depth‐separable volumes can actively reduce basic model parameters. Moreover, the lightweight and efficient attention modules can achieve local cross‐channel interaction without dimensionality reduction and improve the network segmentation performance.

Morphometric Changes between Peroba and Redonda Beaches, Icapuí-Ceará Region, Brazil, Applying High Resolution Remotely Piloted Aircraft Images from 2020 to 2021

Beaches are regions of the coastal zone where the effects of interaction between marine, terrestrial and atmospheric ecosystems are often seen. Natural and anthropic agents act together in coastal processes, which may result in erosion or accretion. If the erosion is intense and continuous, it will represent a severe problem, especially in densely occupied and economically significant beaches, such as the beaches of Peroba and Redonda that have registered erosive events. Thus, this work aimed to analyze the variation of the coastline and the volume of beach sediments through a historical series of one year employing a remotely piloted aircraft, conjugated with the analysis of sediments collected in fields. At Peroba beach, the shoreline was stable, having varied by about 0,87 m/year, and the volume indicated a positive sediment balance of about 18.815 m3. At Redonda beach, the coastline degraded by about 1,27 m/year. The volume indicated a negative sediment balance of -1.230 m3. The beaches were mainly characterized by sediments of very fine and fine sand fractions. The volume variation indicated similar trends to the shoreline, with more significant deposition in the west and erosion in the east.

Uncrewed aerial vehicle radiometric calibration: A comparison of autoexposure and fixed‐exposure images

AbstractRemote sensing with uncrewed aerial vehicles (UAVs) is increasingly being used in agriculture to provide data on the physical characteristics of plants under field conditions. Data accuracy is critical for decision making with a high degree of confidence. In this work, we compared two multispectral camera calibration methods for image data collected with a UAV: (1) an autoexposure method that relies on a single calibration panel and a post hoc calibration, and (2) a fixed‐exposure system that uses three in‐field gray calibration panels using the empirical line calibration method. Both methods were compared to reflectance data from (a) four ground calibration targets measured with a spectroradiometer and (b) a single manned aircraft image calibrated with commercial calibration tarps. In a band‐by‐band comparison, the autoexposure method produced almost twice as much radiometric error on average compared with fixed exposure. Because remote sensing data are commonly converted to spectral indices, the calibration methods were also evaluated by calculating the visible atmospherically resistant index (VARI) and comparing the resulting data to the manned aircraft image. Similarly, the autoexposure method in this case produced twice the error of the fixed‐exposure method. The effect of the error was considered in a production agriculture context by simulating a remote sensing‐based prescription map for pesticide application in a cotton (Gossypium) field and calculating the number of mislabeled management zones. The simulation showed that the autoexposure method would be more costly to the farm because of its higher error, roughly $8.00/ha based on the assumptions made.

Optical performance evaluation of an infrared system of a hypersonic vehicle in an aero-thermal environment.

At hypersonic velocities, the turbulent flow field generated by an aircraft, along with its temperature distribution, leads to significant aerodynamic optical effects that severely impede the performance of internal optical systems. This study proposes a method for analyzing the temporal characteristics of imaging degradation in a detector window infrared imaging system under different field angles of hypersonic velocity. Based on heat transfer theory, a method for solving the transient temperature field in the optical window of a high-speed aircraft is derived and established, considering unsteady thermal conduction-radiation coupling. Additionally, an optical window radiation tracing method is introduced, which directly determines the initial direction vector of light reaching the detector. This method reduces the workload of radiation transmission, significantly enhancing the efficiency of radiation calculations. The time characteristics of image degradation caused by aero-optical effects in high-speed aircraft are analyzed using metrics such as peak signal-to-noise ratio, wave aberration, and point diffusion function. The results demonstrate that as working time increases and the viewing angle widens, the impact of aero-optics on the aircraft imaging system becomes more severe. Moreover, compared to the aerodynamic light transmission effect, the aerodynamic thermal radiation effect has a more detrimental influence on imaging quality.

The use of cylindrical wing propellers in aviation technology

In the program documents, which describe the national development goals of the Russian Federation until 2030 with a forecast until 2025, it is stated that one of the main state goals is to expand public access to safe and high-quality transport services, increase spatial connectivity and transport accessibility of territories and mobility of the population, etc. To solve the state tasks, it is proposed to use new aviation equipment generations based on cylindrical wing propellers. The article discusses the results of research and development work on several dimensions of cylindrical wing propellers. The results obtained allow us to lay the foundation for the theory and practice of operation of cylindrical wing propellers. The understanding of the role of the geometric relations of the rotor elements and the number of blades is laid. The limits of regulation of the angles of attack and deflection of the thrust vector are determined. Strength requirements have been formed for the design and rigidity of the parts of the cylindrical-shaped impeller. The solution of kinematic mechanisms for controlling the angles of attack of the blades and flaps, and the thrust vector in general, etc. was formed. A conceptual image of a two-seat aircraft using cylindrical wing propellers is proposed. Based on the conducted bench tests of a cylindrical-type winged propulsor, tactical and technical requirements for a two-seat local aircraft using these propellers were formed. Conclusions were drawn about the need for additional research and development work to study the influence of several working aircraft rotors on each other. It is stated that it is necessary to optimize the overall dimensions of the cylindrical rotors themselves, etc. It is proposed to conduct further research to determine structural solutions for promising aircraft of greater payload and capacity.

Electromagnetic Metasurface with Object Information and Position Illusion in Radar Vision at Microwave Frequencies

AbstractCamouflage skills are important for understanding the relationship between predators and prey in ecological systems. Mimicry is a common strategy used by creatures to alter their appearance in order to blend in with their surroundings and avoid detection. This paper unveils a previously unknown camouflage strategy that creates false object information and positions of an object in the visual range of radar sensor vision. A dispersion‐modulated illusionary metasurface to realize focal spot coding in a 3‐dimensional space to fit the radar imaging process is proposed. As the radar image indicates, an aircraft image is created by a flat metasurface plane with good polarization adaptation ability. Design, simulation, and experimental results validate the study. This camouflage strategy indicates the complexity of the food chain and can be used by both predators and prey, providing accessible insights for maintaining biological diversity. The metasurface‐based system can be potentially utilized in future radar markers for marine traffic control systems.

LRF-SRNet: Large-Scale Super-Resolution Network for Estimating Aircraft Pose on the Airport Surface.

The introduction of various deep neural network architectures has greatly advanced aircraft pose estimation using high-resolution images. However, realistic airport surface monitors typically take low-resolution (LR) images, and the results of the aircraft pose estimation are far from being accurate enough to be considered acceptable because of long-range capture. To fill this gap, we propose a brand-new, end-to-end low-resolution aircraft pose estimate network (LRF-SRNet) to address the problem of estimating the pose of poor-quality airport surface surveillance aircraft images. The method successfully combines the pose estimation method with the super-resolution (SR) technique. Specifically, to reconstruct high-resolution aircraft images, a super-resolution network (SRNet) is created. In addition, an essential component termed the large receptive field block (LRF block) helps estimate the aircraft's pose. By broadening the neural network's receptive field, it enables the perception of the aircraft's structure. Experimental results demonstrate that, on the airport surface surveillance dataset, our method performs significantly better than the most widely used baseline methods, with AP exceeding Baseline and HRNet by 3.1% and 4.5%.

Identification and Counting of Coffee Trees Based on Convolutional Neural Network Applied to RGB Images Obtained by RPA

Computer vision algorithms for counting plants are an indispensable alternative in managing coffee growing. This research aimed to develop an algorithm for automatic counting of coffee plants and to determine the best age to carry out monitoring of plants using remotely piloted aircraft (RPA) images. This algorithm was based on a convolutional neural network (CNN) system and Open Source Computer Vision Library (OpenCV). The analyses were carried out in coffee-growing areas at the development stages three, six, and twelve months after planting. After obtaining images, the dataset was organized and inserted into a You Only Look Once (YOLOv3) neural network. The training stage was undertaken using 7458 plants aged three, six, and twelve months, reaching stability in the iterations between 3000 and 4000 it. Plant detection within twelve months was not possible due to crown unification. A counting accuracy of 86.5% was achieved with plants at three months of development. The plants’ characteristics at this age may have influenced the reduction in accuracy, and the low uniformity of the canopy may have made it challenging for the neural network to define a pattern. In plantations with six months of development, 96.8% accuracy was obtained for counting plants automatically. This analysis enables the development of an algorithm for automated counting of coffee plants using RGB images obtained by remotely piloted aircraft and machine learning applications.

Rotating blade propellers in prospective aviation technology

The study proposes the use of a new generation of aircraft based on the use of cylindrical rotating blade propellers as a means of transportation. The results of research and development work carried out on several dimensions of cylindrical rotating blade propellers are considered. The results obtained allow us to lay down a certain basis for the theory and practice of the operation of cylindrical rotating blade propellers. In the process of scientific research, an understanding of the role of geometric ratios of rotor elements and the number of blades is obtained. The limits of the angle of attack and thrust vector deflection are defined. The requirements for structural strength and stiffness of the rotating blade propeller parts are formed. The solution of the kinematic mechanisms for controlling the angles of attack of the blades and flaps, and the thrust vector in general, etc., was formed. A conceptual image of a two-seat aircraft using cylindrical rotating blade propellers is proposed. Bench tests of the cylindrical rotating blade propellers allowed the formation of tactical and technical requirements for the two-seat aircraft. It is suggested that further research be considered to determine design solutions for promising aircraft of greater payload and capacity.

Research on Aircraft image recognition based on transfer learning and improved YOLOv5 model

Effective differentiation of aircraft types using images is important for providing military combat information as well as civilian aircraft operations. Aircraft image recognition has many difficulties such as large variations of target scale, complex backgrounds, and difficult data set acquisition, which lead to the low recognition accuracy of existing models. To address the problem of low recognition accuracy caused by the above difficulties. This paper proposes the improved YOLOv5 model for the recognition of aircraft images. First, this paper designs the CSPResNet50dCA network as the backbone of the YOLOv5 model to enhance the feature extraction capability for small target aircraft in images. By introducing the coordinate attention mechanism and the CSP structure, the feature-focusing capability and the computing speed of the model are enhanced. Afterward, we use data enhancement to expand the data set and transfer learning to improve the generalization ability and convergence speed of the model, so as to improve its robustness. The experimental results show that the improved YOLOv5 model has significantly improved the recognition accuracy of aircraft targets， and significantly enhanced feature extraction ability for small target aircraft with good generalization ability.

Early detection of weed in sugarcane using convolutional neural network

Weed infestation is an essential factor in sugarcane productivity loss. The use of remote sensing data in conjunction with Artificial Intelligence (AI) techniques, can lead the cultivation of sugarcane to a new level in terms of weed control. For this purpose, an algorithm based on Convolutional Neural Networks (CNN) was developed to detect, quantify, and map weeds in sugarcane areas located in the state of Alagoas, Brazil. Images of the PlanetScope satellite were subdivided, separated, trained in different scenarios, classified and georeferenced, producing a map with weed information included. Scenario one of the CNN training and test presented overall accuracy (0,983), and it was used to produce the final mapping of forest areas, sugarcane, and weed infestation. The quantitative analysis of the area (ha) infested by weed indicated a high probability of a negative impact on sugarcane productivity. It is recommended that the adequacy of CNN’s algorithm for Remotely Piloted Aircraft (RPA) images be carried out, aiming at the differentiation between weed species, as well as its application in the detection in areas with different culture crops

Use of RPA Images in the Mapping of the Chlorophyll Index of Coffee Plants

Coffee trading is an important source of income for the Brazilian commercial balance. Chlorophyll (Chl) are pigments responsible for converting radiation into energy; these pigments are closely related to the photosynthetic efficiency of plants, and the evaluation of the nutritional status of the coffee tree. The inversion method can be used for estimating the canopy chlorophyll content (Chlcanopy) using the leaf chlorophyll content (Chlleaf) and the leaf area index (LAI). The application of vegetation indices (VIs) in high spatial resolution images obtained from remotely piloted aircraft (RPA) can assist in the characterization of Chlcanopy in addition to providing vital and fast information for monitoring crops and aiding decision-making. This study aimed to identify which VIs adequately explain the Chl and evaluate the relationships between the VIs obtained from remotely piloted aircraft (RPA) images and the Chlleaf and Chlcanopy in coffee plants during the wet and dry seasons. The experiment was conducted on a Coffea arabica L. plantation in Lavras, Minas Gerais, Brazil. Images were collected on 26 November 2019 (wet), 11 August 2020 (dry), and 26 August 2021 (dry) by a multispectral camera embedded in a quadcopter. Plant height (H), crow diameter (D), and Chlleaf (a, b and total) data were collected in the field by a metre ruler (H and D) and sensor (Chlleaf). The LAI was calculated based on H and D. The Chlcanopy (a, b, and total) was calculated based on Chlleaf and LAI. The image processing was performed in Pix4D software, and postprocessing and calculation of the 21 VIs were performed in QGIS. Statistical analyses (descriptive, statistical tests, Pearson correlation, residuals calculation, and linear regression) were performed using the software R. The VIs from the RPA that best correlates to Chlcanopy in the wet season were the Modified Chlorophyll Absorption Ratio Index 2 (MCARI2RPA), Modified Simple Ratio (MSRRPA) and Simple Ratio (SRRPA). These VIs had high sensitivity and, therefore, were more affected by chlorophyll variability. For the two dry season studied days, there were no patterns in the relationships between Chlleaf, Chlcanopy, and the VIs. It was possible to use the Chl inversion method for the coffee during the wet season.

MS-IAF: Multi-Scale Information Augmentation Framework for Aircraft Detection

Aircrafts have been an important object of study in the field of multi-scale image object detection due to their important strategic role. However, the multi-scale detection of aircrafts and their key parts from remote sensing images can be a challenge, as images often present complex backgrounds and obscured conditions. Most of today’s multi-scale datasets consist of independent objects and lack mixed annotations of aircrafts and their key parts. In this paper, we contribute a multi-scale aircraft dataset (AP-DATA) consisting of 7000 aircraft images that were taken in complex environments and obscured conditions. Our dataset includes mixed annotations of aircrafts and their key parts. We also present a multi-scale information augmentation framework (MS-IAF) to recognize multi-scale aircrafts and their key parts accurately. First, we propose a new deep convolutional module ResNeSt-D as the backbone, which stacks scattered attention in a multi-path manner and makes the receptive field more suitable for the object. Then, based on the combination of Faster R-CNN with ResNeSt-D, we propose a multi-scale feature fusion module called BFPCAR. BFPCAR overcomes the attention imbalance problem of the non-adjacent layers of the FPN module by reducing the loss of information between different layers and including more semantic features during information fusion. Based on AP-DATA, a dataset with three types of features, the average precision (AP) of MS-IAF reached 0.884, i.e., 2.67% higher than that of the original Faster R-CNN. The APs of these two modules were improved by 2.32% and 1.39%, respectively. The robustness of our proposed model was validated using the open sourced RSOD remote sensing image dataset, and the best accuracy was achieved.