Drone Images Research Articles

Physics-Driven Image Dehazing from the Perspective of Unmanned Aerial Vehicles

Drone vision is widely used in change detection, disaster response, and military reconnaissance due to its wide field of view and flexibility. However, under haze and thin cloud conditions, image quality is usually degraded due to atmospheric scattering. This results in issues like color distortion, reduced contrast, and lower clarity, which negatively impact the performance of subsequent advanced visual tasks. To improve the quality of unmanned aerial vehicle (UAV) images, we propose a dehazing method based on calibration of the atmospheric scattering model. We designed two specialized neural network structures to estimate the two unknown parameters in the atmospheric scattering model: the atmospheric light intensity A and medium transmission t. However, calculation errors always occur in both processes for estimating the two unknown parameters. The error accumulation for atmospheric light and medium transmission will cause the deviation in color fidelity and brightness. Therefore, we designed an encoder-decoder structure for irradiance guidance, which not only eliminates error accumulation but also enhances the detail in the restored image, achieving higher-quality dehazing results. Quantitative and qualitative evaluations indicate that our dehazing method outperforms existing techniques, effectively eliminating haze from drone images and significantly enhancing image clarity and quality in hazy conditions. Specifically, the compared experiment on the R100 dataset demonstrates that the proposed method improved the peak signal-to-noise ratio (PSNR) and structure similarity index measure (SSIM) metrics by 6.9 dB and 0.08 over the second-best method, respectively. On the N100 dataset, the method improved the PSNR and SSIM metrics by 8.7 dB and 0.05 over the second-best method, respectively.

Efficient Ship Detection System for Maritime Surveillance using Deep Learning Approach

The usage of drones in maritime surveillance is a very effective means to observe ships. As the ship detection is vital for applications such as port monitoring, cross-border surveillance, Moreover, it is not only vital for maritime surveillance but also environmental conservation. Despite its effectiveness, the drone-captured images present itself with its own challenges, specifically when leveraging on a custom dataset which is tailored for specific application. To address the issues, we propose a novel ship detection approach called YOLOv9 with Adan optimizer (YOLOv9-Adan) which is more accurate to increase the efficiency of maritime surveillance. The YOLOv9-Adan model integrates the robust object detection capabilities of the adaptive learning capabilities of the Adan (ADAptive Nesterov momentum algorithm) optimiser, trained on the drone image dataset comprising 3200 images of maritime scenes and ship types in drone view. Our model is trained on a drone-image dataset comprising 3200 images of maritime scenes and ship types in drone views collected from various sources. The experimental results show that our approach using the YOLOv9-Adan model achieves 65.5% mAP, which exceeds the mAP of YOLOv9 by 4.3%. Additionally, This article also provides a comparative analysis of our model YOLOv9-Adan with other existing models in literature with consistently surpassing existing approaches

IoT Based Precision Farming

This project focuses on leveraging drone images of the pests equipped with advanced sensors for pest detection in crops, combined with methods for image processing to identify diseases. The ultimate goal is to enhance crop health and productivity through timely and targeted pesticide application. Image processing techniques are used to detect signs of diseases and pests in the captured images. The use of machine learning CNN algorithm enhances the system’s ability to accurately classify and diagnose crop heath issues. Upon detection of pests, the IOT platform triggers a response mechanism to deploy a precision pesticide spraying system. This ensures targeted and localized treatment, reducing the overall use of pesticides and minimizing environmental impact. This project involves capturing images of pests using a camera, followed by processing these images to extract key features using various image processing techniques. The extracted features are analyzed using algorithms, primarily Convolutional Neural Networks (CNNs), to detect variations in color and other dominant characteristics in the images. By comparing these features across samples, the system can identify pests and plant diseases more efficiently. This approach aims to provide a quicker and more cost-effective solution for pest detection and disease management.

Fracture length data for geothermal applications

Fracture lengths govern permeability and are unknowns in geothermal assessment. Along their lengths, fracture widths vary due to growth by linkage. Under the influence of diagenesis, narrow widths seal, breaking porosity continuity and reducing open length. The largest range of widths and thus susceptibility to fill occurs where fractures are linked by narrow segments. Outcrops of a geothermal target, Cambrian Potsdam quartz arenite, contain opening-mode fractures having lengths spanning five orders of magnitude from 0.082 mm to 17.9 m. Combined lengths measured at a range of scales can be described by power laws, but at a given image resolution, lengths are best fit by exponential functions. Owing to preferential sealing of small fractures, open fractures follow exponential functions, but values depend on rules for designating fractures as continuous. En échelon segments, offset 10 mm, are connected by narrow fractures or microfractures (hard linked) not evident on outcrop 1-m-elevation LiDAR or 30-m-height drone images. A rule that identifies where narrow but probably connected segments yields lengths that are meaningful for flow simulation. Depending on diagenesis, continuity rules can halve or double average and maximum length values. Length values from outcrop for geothermal applications should be adjusted based on wellsite-specific diagenesis information.

THE USE OF UNMANNED AIRCRAFT IN THE PRODUCTION PROCESS OF DRILLING AND BLASTING AT SURFACE MINES

Surface mines that have solid rocks in their geological structure require considerable dedication to the work processes of drilling and blasting. Drilling and blasting, as one of the first work processes in surface mines - quarries, affects all subsequent work processes (loading, transport and crushing) as a result. By improving the effects of drilling and blasting, KPI results are significantly improved on all subsequent work processes, as well as on the reduced impact of mining on the environment. The paper shows the use of UAVs (unmanned aerial vehicles) for the purpose of optimizing the results of drilling and blasting in the Anhovo quarry (carbonate mineral - R. Slovenia), whose entire annual production of 1.5×106 t goes to the cement industry (Alpacem cement d.d. Slovenia). The equipment used in the aforementioned quarry is a handheld Leica GNSS RTK device, a DJI Mavic 3E drone with RTK positioning, a Sandvik Ranger DX800i drilling rig with RTK positioning, as well as BlastMetrix software for processing data obtained from drone images and projecting minefields.

Two-dimensional hyperchaos-based encryption and compression algorithm for agricultural UAV-captured planar images.

This study presents an approach that integrates compressed sensing technology with two-dimensional hyperchaotic coupled Fourier oscillator systems (2D-HCFOS) to address the challenge of slow encryption speeds in agricultural unmanned aerial vehicles (UAVs). The primary challenge in enhancing encryption speed lies in the limited capacity inherent in traditional chaotic-based systems and the computational complexity of their processes. The 2D-HCFOS utilizes a complex two-dimensional hybrid chaotic system, which significantly enhances the security of agricultural UAV image data. Notably, the image encryption process is performed on a personal computer connected to the drone, ensuring efficient processing. By integrating advanced Fourier series and nonlinear coupled oscillators, the model surpasses existing chaotic-based methods, improving both the pseudo-randomness and robustness of encryption. Additionally, incorporating Bonouille functions into the discrete cosine transform (DCT) domain results in a sparser measurement matrix, which is essential for efficient encryption on personal computers. The effectiveness of 2D-HCFOS in securely encrypting agricultural drone images has been rigorously validated through simulations and analytical evaluations using sophisticated row, rotation, and matrix encryption techniques. The improved security performance is further verified by comparative analysis. Compared with other models, the Lyapunov index of 2D-HCFOS is 15.1039, and the sample entropy is 2.4987, indicating that it possesses superior chaotic performance and encryption reliability.

MFRENet: efficient detection of drone image based on multiscale feature aggregation and receptive field expanded

MFRENet: efficient detection of drone image based on multiscale feature aggregation and receptive field expanded

Assessment of Dataset Scalability for Classification of Black Sigatoka in Banana Crops Using UAV-Based Multispectral Images and Deep Learning Techniques

This paper presents an evaluation of different convolutional neural network (CNN) architectures using false-colour images obtained by multispectral sensors on drones for the detection of Black Sigatoka in banana crops. The objective is to use drones to improve the accuracy and efficiency of Black Sigatoka detection to reduce its impact on banana production and improve the sustainable management of banana crops, one of the most produced, traded, and important fruits for food security consumed worldwide. This study aims to improve the precision and accuracy in analysing the images and detecting the presence of the disease using deep learning algorithms. Moreover, we are using drones, multispectral images, and different CNNs, supported by transfer learning, to enhance and scale up the current approach using RGB images obtained by conventional cameras and even smartphone cameras, available in open datasets. The innovation of this study, compared to existing technologies for disease detection in crops, lies in the advantages offered by using drones for image acquisition of crops, in this case, constructing and testing our own datasets, which allows us to save time and resources in the identification of crop diseases in a highly scalable manner. The CNNs used are a type of artificial neural network widely utilised for machine training; they contain several specialised layers interconnected with each other in which the initial layers can detect lines and curves, and gradually become specialised until reaching deeper layers that recognise complex shapes. We use multispectral sensors to create false-colour images around the red colour spectra to distinguish infected leaves. Relevant results of this study include the construction of a dataset with 505 original drone images. By subdividing and converting them into false-colour images using the UAV’s multispectral sensors, we obtained 2706 objects of diseased leaves, 3102 objects of healthy leaves, and an additional 1192 objects of non-leaves to train classification algorithms. Additionally, 3640 labels of Black Sigatoka were generated by phytopathology experts, ideal for training algorithms to detect this disease in banana crops. In classification, we achieved a performance of 86.5% using false-colour images with red, red edge, and near-infrared composition through MobileNetV2 for three classes (healthy leaves, diseased leaves, and non-leaf extras). We obtained better results in identifying Black Sigatoka disease in banana crops using the classification approach with MobileNetV2 as well as our own datasets.

The Early Prediction of Kimchi Cabbage Heights Using Drone Imagery and the Long Short-Term Memory (LSTM) Model

Accurate and timely crop growth prediction is crucial for efficient farm management and food security, particularly given challenges like labor shortages and climate change. This study presents a novel method for the early prediction of Kimchi cabbage heights using drone imagery and a long short-term memory (LSTM) model. High-resolution drone images were used to generate a canopy height model (CHM) for estimating plant heights at various growth stages. Missing height data were interpolated using a logistic growth curve, and an LSTM model was trained on this time series data to predict the final height at harvest well before the actual harvest date. The model trained on data from 44 days after planting (DAPs) demonstrated the highest accuracy (R2 = 0.83, MAE = 2.48 cm, and RMSE = 3.26 cm). Color-coded maps visualizing the predicted Kimchi cabbage heights revealed distinct growth patterns between different soil types, highlighting the model’s potential for site-specific management. Considering the trade-off between accuracy and prediction timing, the model trained on DAP 36 data (MAE = 2.77 cm) was deemed most suitable for practical applications, enabling timely interventions in cultivation management. This research demonstrates the feasibility and effectiveness of integrating drone imagery, logistic growth curves, and LSTM models for the early and accurate prediction of Kimchi cabbage heights, facilitating data-driven decision-making in precision agriculture for improved crop management and yield optimization.

A novel domain independent scene text localizer

Text localization across multiple domains is crucial for applications like autonomous driving and tracking marathon runners. This work introduces DIPCYT, a novel model that utilizes Domain Independent Partial Convolution and a Yolov5-based Transformer for text localization in scene images from various domains, including natural scenes, underwater, and drone images. Each domain presents unique challenges: underwater images suffer from poor quality and degradation, drone images suffer from tiny text and loss of shapes, and scene images suffer from arbitrarily oriented, shaped text. Additionally, license plates in drone images may not provide rich semantic information compared to other text types due to loss of contextual information between characters. To tackle these challenges, DIPCYT employs new partial convolution layers within Yolov5 and integrates Transformer detection heads with a novel Fourier Positional Convolutional Block Attention Module (FPCBAM). This approach leverages common text properties across domains, such as contextual (global) and spatial (local) relationships. Experimental results demonstrate that DIPCYT outperforms existing methods, achieving F-scores of 0.90, 0.90, 0.77, 0.85, 0.85, and 0.88 on Total-Text, ICDAR 2015, ICDAR 2019 MLT, CTW1500, Drone, and Underwater datasets, respectively.

Cattle dung detection in pastures from drone images using YOLOv5

Cattle dung detection in pastures from drone images using YOLOv5

High-resolution mapping of grassland canopy cover in China through the integration of extensive drone imagery and satellite data

Canopy cover is a crucial indicator for assessing grassland health and ecosystem services. However, achieving accurate high-resolution estimates of grassland canopy cover at a large spatial scale remains challenging due to the limited spatial coverage of field measurements and the scale mismatch between field measurements and satellite imagery. In this study, we addressed these challenges by proposing a regression-based approach to estimate large-scale grassland canopy cover, leveraging the integration of drone imagery and multisource remote sensing data. Specifically, over 90,000 10 × 10 m drone image tiles were collected at 1,255 sites across China. All drone image tiles were classified into grass and non-grass pixels to generate ground-truth canopy cover estimates. These estimates were then temporally aligned with satellite imagery-derived features to build a random forest regression model to map the grassland canopy cover distribution of China. Our results revealed that a single classification model can effectively distinguish between grass and non-grass pixels in drone images collected across diverse grassland types and large spatial scales, with multilayer perceptron demonstrating superior classification accuracy compared to Canopeo, support vector machine, random forest, and pyramid scene parsing network. The integration of extensive drone imagery successfully addressed the scale-mismatch issue between traditional ground measurements and satellite imagery, contributing significantly to enhancing mapping accuracy. The national canopy cover map of China generated for the year 2021 exhibited a spatial pattern of increasing canopy cover from northwest to southeast, with an average value of 56 % and a standard deviation of 26 %. Moreover, it demonstrated high accuracy, with a coefficient of determination of 0.89 and a root-mean-squared error of 12.38 %. The resulting high-resolution canopy cover map of China holds great potential in advancing our comprehension of grassland ecosystem processes and advocating for the sustainable management of grassland resources.

GAM4water: An R-based method for extracting wetted areas from remotely-sensed images

We present ‘GAM4water,’ a R-based method to classify wetted and non-wetted (dry) areas using remotely sensed image indices derived from such images. The GAM4water classification algorithm is built around a Generalized Additive Model (GAM) capable of accounting for non-linear responses. GAM4water can use any type of radiometric data, whether from drones, satellites or other platforms, and can be used with data of different spatial resolutions, geographic extents and spatial reference systems. It is a supervised tool that uses pixel information to distinguish between wetted and dry areas within an image set, extract them and produce a rich output that includes a binary raster, polygons of wetted areas, and a classification performance report. We tested the method in two case-studies, one using high resolution drone images and another using satellite images. The tests show that GAM4water can produce highly accurate classifications of wetted and non-wetted areas, and has the additional benefit of being easily customizable and not requiring complex implementation procedures.•This paper introduces the first R based method of wetted area extraction for remotely-sensed images.•The method is based on Generalized Additive Models and is applicable to any remotely-sensed data.

Enhancing oil palm segmentation model with GAN-based augmentation

In digital agriculture, accurate crop detection is fundamental to developing automated systems for efficient plantation management. For oil palm, the main challenge lies in developing robust models that perform well in different environmental conditions. This study addresses the feasibility of using GAN augmentation methods to improve palm detection models. For this purpose, drone images of young palms (< 5 year-old) from eight different estates were collected, annotated, and used to build a baseline detection model based on DETR. StyleGAN2 was trained on the extracted palms and then used to generate a series of synthetic palms, which were then inserted into tiles representing different environments. CycleGAN networks were trained for bidirectional translation between synthetic and real tiles, subsequently utilized to augment the authenticity of synthetic tiles. Both synthetic and real tiles were used to train the GAN-based detection model. The baseline model achieved precision and recall values of 95.8% and 97.2%. The GAN-based model achieved comparable result, with precision and recall values of 98.5% and 98.6%. In the challenge dataset 1 consisting older palms (> 5 year-old), both models also achieved similar accuracies, with baseline model achieving precision and recall of 93.1% and 99.4%, and GAN-based model achieving 95.7% and 99.4%. As for the challenge dataset 2 consisting of storm affected palms, the baseline model achieved precision of 100% but recall was only 13%. The GAN-based model achieved a significantly better result, with a precision and recall values of 98.7% and 95.3%. This result demonstrates that images generated by GANs have the potential to enhance the accuracies of palm detection models.

A Vision-Language Model-Based Traffic Sign Detection Method for High-Resolution Drone Images: A Case Study in Guyuan, China.

As a fundamental element of the transportation system, traffic signs are widely used to guide traffic behaviors. In recent years, drones have emerged as an important tool for monitoring the conditions of traffic signs. However, the existing image processing technique is heavily reliant on image annotations. It is time consuming to build a high-quality dataset with diverse training images and human annotations. In this paper, we introduce the utilization of Vision-language Models (VLMs) in the traffic sign detection task. Without the need for discrete image labels, the rapid deployment is fulfilled by the multi-modal learning and large-scale pretrained networks. First, we compile a keyword dictionary to explain traffic signs. The Chinese national standard is used to suggest the shape and color information. Our program conducts Bootstrapping Language-image Pretraining v2 (BLIPv2) to translate representative images into text descriptions. Second, a Contrastive Language-image Pretraining (CLIP) framework is applied to characterize not only drone images but also text descriptions. Our method utilizes the pretrained encoder network to create visual features and word embeddings. Third, the category of each traffic sign is predicted according to the similarity between drone images and keywords. Cosine distance and softmax function are performed to calculate the class probability distribution. To evaluate the performance, we apply the proposed method in a practical application. The drone images captured from Guyuan, China, are employed to record the conditions of traffic signs. Further experiments include two widely used public datasets. The calculation results indicate that our vision-language model-based method has an acceptable prediction accuracy and low training cost.

Flying foxes optimization with reinforcement learning for vehicle detection in UAV imagery

Intelligent transportation systems (ITS) are globally installed in smart cities, which enable the next generation of ITS depending on the potential integration of autonomous and connected vehicles. Both technologies are being tested widely in various cities across the world. However, these two developing technologies are vital in allowing a fully automatic transportation system; it is necessary to automate other transportation and road components. Unmanned aerial vehicles (UAVs) or drones are utilized for many surveillance applications in the ITS. Detecting on-ground vehicles in drone images is significant for disaster rescue operations, traffic and parking management, and navigating uneven territories. This study presents a flying foxes optimization with deep learning-based vehicle detection and classification model on aerial images (FFODL-VDCAI) technique for ITS application. The main objective of the FFODL-VDCAI technique is to automate and accurately classify vehicles that exist in aerial images. Three primary processes are involved in the presented FFODL-VDCAI technique. Initially, the FFODL-VDCAI approach utilizes YOLO-GD (Ghost-Net and Depthwise convolution) for vehicle detection, where the YOLO-GD uses lightweight Ghost Net in place on the backbone network of YOLO-v4 and interchanges the conventional convolutional with depthwise separable convolutional and pointwise convolutional. Next, the FFO technique is used for hyperparameter tuning the Ghost Net technique. Finally, a deep Q-network (DQN) based reinforcement learning technique is used to classify detected vehicles effectively. A comprehensive simulation analysis of the FFODL-VDCAI methodology is conducted on the UAV image dataset. The performance validation of the FFODL-VDCAI methodology exhibited superior values of 96.15% and 92.03% under PSU and Stanford datasets concerning various aspects.

Drone and ground-truth data collection, image annotation and machine learning: A protocol for coastal habitat mapping and classification

Aerial drone imaging is an efficient tool for mapping and monitoring of coastal habitats at high spatial and temporal resolution. Specifically, drone imaging allows for time- and cost-efficient mapping covering larger areas than traditional mapping and monitoring techniques, while also providing more detailed information than those from airplanes and satellites, enabling for example to differentiate various types of coastal vegetation. Here, we present a systematic method for shallow water habitat classification based on drone imagery. The method includes:•Collection of drone images and creation of orthomosaics.•Gathering ground-truth data in the field to guide the image annotation and to validate the final map product.•Annotation of drone images into – potentially hierarchical – habitat classes and training of machine learning algorithms for habitat classification.As a case study, we present a field campaign that employed these methods to map a coastal site dominated by seagrass, seaweed and kelp, in addition to sediments and rock. Such detailed but efficient mapping and classification can aid to understand and sustainably manage ecologically and valuable marine ecosystems.

Heating decoys to mimic thermal signatures of live animals for drones

Thermal sensors mounted on drones (unoccupied aircraft systems) are popular and effective tools for monitoring cryptic animal species, although few studies have quantified sampling error of animal counts from thermal images. Using decoys is one effective strategy to quantify bias and count accuracy; however, plastic decoys do not mimic thermal signatures of representative species. Our objective was to produce heat signatures in animal decoys to realistically match thermal images of live animals obtained from a drone-based sensor. We tested commercially available methods to heat plastic decoys of three different size classes, including chemical foot warmers, manually heated water, electric socks, pad, or blanket, and mini and small electric space heaters. We used criteria in two categories, 1) external temperature differences from ambient temperatures (ambient difference) and 2) color bins from a palette in thermal images obtained from a drone near the ground and in the air, to determine if heated decoys adequately matched respective live animals in four body regions. Three methods achieved similar thermal signatures to live animals for three to four body regions in external temperatures and predominantly matched the corresponding yellow color bins in thermal drone images from the ground and in the air. Pigeon decoys were best and most consistently heated with three-foot warmers. Goose and deer decoys were best heated by mini and small space heaters, respectively, in their body cavities, with a heated sock in the head of the goose decoy. The materials and equipment for our best heating methods were relatively inexpensive, commercially available items that provide sustained heat and could be adapted to various shapes and sizes for a wide range of avian and mammalian species. Our heating methods could be used in future studies to quantify bias and validate methodologies for drone surveys of animals with thermal sensors.•We determined optimal heating methods for plastic animal decoys with inexpensive and commercially available equipment to mimic thermal signatures of live animals.•Methods could be used to quantify bias and improve thermal surveys of animals with drones in future studies.

Drone image recognition and intelligent power distribution network equipment fault detection based on the transformer model and transfer learning

Drone image recognition and intelligent power distribution network equipment fault detection based on the transformer model and transfer learning

A Lightweight Insulator Defect Detection Model Based on Drone Images

With the continuous development and construction of new power systems, using drones to inspect the condition of transmission line insulators has become an inevitable trend. To facilitate the deployment of drone hardware equipment, this paper proposes IDD-YOLO (Insulator Defect Detection-YOLO), a lightweight insulator defect detection model. Initially, the backbone network of IDD-YOLO employs GhostNet for feature extraction. However, due to the limited feature extraction capability of GhostNet, we designed a lightweight attention mechanism called LCSA (Lightweight Channel-Spatial Attention), which is combined with GhostNet to capture features more comprehensively. Secondly, the neck network of IDD-YOLO utilizes PANet for feature transformation and introduces GSConv and C3Ghost convolution modules to reduce redundant parameters and lighten the network. The head network employs the YOLO detection head, incorporating the EIOU loss function and Mish activation function to optimize the speed and accuracy of insulator defect detection. Finally, the model is optimized using TensorRT and deployed on the NVIDIA Jetson TX2 NX mobile platform to test the actual inference speed of the model. The experimental results demonstrate that the model exhibits outstanding performance on both the proprietary ID-2024 insulator defect dataset and the public SFID insulator dataset. After optimization with TensorRT, the actual inference speed of the IDD-YOLO model reached 20.83 frames per second (FPS), meeting the demands for accurate and real-time inspection of insulator defects by drones.