Unmanned Aerial Vehicle Images Research Articles

Object detection model with efficient feature extraction and asymptotic feature fusion for unmanned aerial vehicle image

Object detection model with efficient feature extraction and asymptotic feature fusion for unmanned aerial vehicle image

Tree Species Classification from UAV Canopy Images with Deep Learning Models

Forests play a critical role in the provision of ecosystem services, and understanding their compositions, especially tree species, is essential for effective ecosystem management and conservation. However, identifying tree species is challenging and time-consuming. Recently, unmanned aerial vehicles (UAVs) equipped with various sensors have emerged as a promising technology for species identification due to their relatively low cost and high spatial and temporal resolutions. Moreover, the advancement of various deep learning models makes remote sensing based species identification more a reality. However, three questions remain to be answered: first, which of the state-of-the-art models performs best for this task; second, which is the optimal season for tree species classification in a temperate forest; and third, whether a model trained in one season can be effectively transferred to another season. To address these questions, we focus on tree species classification by using five state-of-the-art deep learning models on UAV-based RGB images, and we explored the model transferability between seasons. Utilizing UAV images taken in the summer and fall, we captured 8799 crown images of eight species. We trained five models using summer and fall images and compared their performance on the same dataset. All models achieved high performances in species classification, with the best performance on summer images, with an average F1-score was 0.96. For the fall images, Vision Transformer (ViT), EfficientNetB0, and YOLOv5 achieved F1-scores greater than 0.9, outperforming both ResNet18 and DenseNet. On average, across the two seasons, ViT achieved the best accuracy. This study demonstrates the capability of deep learning models in forest inventory, particularly for tree species classification. While the choice of certain models may not significantly affect performance when using summer images, the advanced models prove to be a better choice for fall images. Given the limited transferability from one season to another, further research is required to overcome the challenge associated with transferability across seasons.

Detection of tea leaf blight in UAV remote sensing images by integrating super-resolution and detection networks.

Tea leaf blight (TLB) is a common disease of tea plants and is widely distributed in tea gardens. Although the use of unmanned aerial vehicle (UAV) remote sensing can help to achieve a wider scale for TLB detection, the blurring of UAV images, overlapping of tea leaves, and small size of TLB spots pose significant challenges to the task of detection. This study proposes a method of detecting TLB in UAV remote sensing images by integrating super-resolution (SR) and detection networks. We use an SR network called SERB-Swin2sr to reconstruct the detailed features of UAV images and solve the problem of detail loss caused by the blurring in UAV images. In SERB-Swin2sr, a squeeze-and-excitation ResNet block (SERB) is introduced to enhance the models' ability to extract the target details in the images, and the convolution stem replaces the convolution block in order to increase the convergence rate and stability of the network. A detection network called SDDA-YOLO is applied to achieve precise detection of TLB in UAV remote sensing images. In SDDA-YOLO, a shuffle dual-dimensional attention (SDDA) module is introduced to enhance the feature fusion capability of the network, and an Xsmall-scale detection layer is used to enhance the detection ability of small lesions. Experimental results show that the proposed method is superior to current detection methods. Compared with a baseline YOLOv8 model, the precision, mAP@0.5, and mAP@0.5:0.95 of the proposed method are improved by 4.2%, 1.6%, and 1.8%, and the size of our model is only 4.6MB.

Differentiating Growth Patterns in Winter Wheat Cultivars via Unmanned Aerial Vehicle Imaging

Wheat is one of the most widely grown cereal crops, serving as a key factor in sustaining the nutritional and food balance in numerous countries. The use of non-contact methods for wheat monitoring allows for the rapid diagnosis of vegetation density, crop growth, and the presence of weeds and diseases in the investigated fields. This study aims to assess the potential for differentiating growth patterns in winter wheat cultivars by examining them with two unmanned aerial vehicles (UAVs), the Mavic 2 Pro and Phantom 4 Pro, equipped with a multispectral camera from the MAPIR™ brand. Based on an experimental study conducted in the Southern Dobruja region (Bulgaria), vegetation reflectance indices, such as the Normalized-Difference Vegetation Index (NDVI), Soil-Adjusted Vegetation Index (SAVI), and Enhanced Vegetation Index 2 (EVI2), were generated, and a database was created to track their changing trends. The obtained results showed that the values of the NDVI, EVI2, and SAVI can be used to predict the productive potential of wheat, but only after accounting for the meteorological conditions of the respective growing season. The proposed methodology provides accurate results in small areas, with a resolution of 0.40 cm/pixel when flying at an altitude of 12 m and 2.3 cm/pixel when flying at an altitude of 100 m. The achieved precision in small and ultra-small agricultural areas, at a width of 1.2 m, will help wheat breeders conduct precise diagnostics of individual wheat varieties.

Rainfall water collection and irrigation via stone bud and karren on karst rocky desertification slopes: Application and benefit analysis

Lack of surface water is a key factor leading to agricultural drought in karst regions and exacerbates the karst rocky desertification, which is characterized by high rate of exposed bedrock. With the characteristics of impermeability, runoff collection, and wide distribution, stone bud and karren (SBK), a unique karst landscape type composed of exposed bedrock and gullies, has a great potential to solve water shortage in agricultural irrigation, especially in the rain-rich karst regions of Guizhou Province, China. Using unmanned aerial vehicle images to identify SBK and measured rainfall–runoff data for validation, we applied SBK as a basic unit to harvest rainfall water on rocky desertification slopes and analyzed the benefits of rainfall water collection and irrigation via SBK. The results showed the following: (1) Karrens on rocky desertification slopes could be identified using a method coupling terrain opening difference and depression filling, and the simulated runoff on SBK could achieved the average NSE and R2 of 0.82 and 0.86, with RMSE and RSR less than 0.03 m3 and 0.51, respectively. (2) The exposed bedrock utilization rate and rainfall water collection coefficient of SBK were 0.62 and 0.55, respectively, such that the area ratio of exposed bedrock to the cultivated land meeting irrigation water demand reached 1:3.1, 1:3.7 and 1:7.0 under 90 %, 80 % and 50 % possibilities of irrigation. (3) Implementing SBK for rainfall water harvesting and irrigation could efficiently avoid crop failure in the karst regions where crops faced drought within only one week after rainfall, and was at an affordable cost to local farmers with limited environmental damage. The use of SBK to harvest rainfall water could also alleviate surface water shortages in karst regions caused by problems such as uneven temporal distribution of rainfall and engineering water shortages. Consequently, SBK demonstrate a strong application potential to alleviate agricultural drought in karst rocky desertification regions.

Unmanned aerial vehicle-based aerial survey of mines in Shanxi Province based on image data

Abstract Accurately monitoring the change of mine area in the mining process is beneficial to mine safety management. This paper briefly introduces the collection of remote sensing images by unmanned aerial vehicles (UAVs) and its application in measuring surface mining subsidence and surrounding vegetation in the mining area. A case study was carried out in some mining areas of Nanshan Mountain, Yuncheng City, Shanxi Province. The surface mining subsidence value and vegetation-related parameters were measured by comparing the digital elevation model and multi-spectral images collected on May 12 and June 12, 2023. The validity experiment verified that the UAV image data could be used to measure the mining subsidence and vegetation parameters. Moreover, it was found that mining underground coal could lead to significant ground subsidence and pollute the surrounding environment, reducing vegetation. The innovation of this article lies in using UAV-collected remote sensing images instead of manually collecting ground elevation data and vegetation distribution data, providing effective references for safe mining in mining areas.

Estimation of Soil Salinity by Combining Spectral and Texture Information from UAV Multispectral Images in the Tarim River Basin, China

Texture features have been consistently overlooked in digital soil mapping, especially in soil salinization mapping. This study aims to clarify how to leverage texture information for monitoring soil salinization through remote sensing techniques. We propose a novel method for estimating soil salinity content (SSC) that combines spectral and texture information from unmanned aerial vehicle (UAV) images. Reflectance, spectral index, and one-dimensional (OD) texture features were extracted from UAV images. Building on the one-dimensional texture features, we constructed two-dimensional (TD) and three-dimensional (THD) texture indices. The technique of Recursive Feature Elimination (RFE) was used for feature selection. Models for soil salinity estimation were built using three distinct methodologies: Random Forest (RF), Partial Least Squares Regression (PLSR), and Convolutional Neural Network (CNN). Spatial distribution maps of soil salinity were then generated for each model. The effectiveness of the proposed method is confirmed through the utilization of 240 surface soil samples gathered from an arid region in northwest China, specifically in Xinjiang, characterized by sparse vegetation. Among all texture indices, TDTeI1 has the highest correlation with SSC (|r| = 0.86). After adding multidimensional texture information, the R2 of the RF model increased from 0.76 to 0.90, with an improvement of 18%. Among the three models, the RF model outperforms PLSR and CNN. The RF model, which combines spectral and texture information (SOTT), achieves an R2 of 0.90, RMSE of 5.13 g kg−1, and RPD of 3.12. Texture information contributes 44.8% to the soil salinity prediction, with the contributions of TD and THD texture indices of 19.3% and 20.2%, respectively. This study confirms the great potential of introducing texture information for monitoring soil salinity in arid and semi-arid regions.

Non-contact Structural Displacement Measurement using UAV and DIC with Non-coplanar Fixed Region of Interest

The use of unmanned aerial vehicles (UAV) for displacement measurement using Digital Image Correlation (DIC) has been popular. A fixed region of interest (FROI) is often used to calibrate the UAV camera. Existing techniques require the FROI to be coplanar with the target plane, such that a corner of the FROI can be taken directly as the origin of the global coordinates. However, this condition does not exist when the target plane is in motion, and this has been a significant challenge in the correction of the UAV images. This paper proposed an algorithm to estimate the posture of the UAV camera for solving this issue. Numerical simulation shows that the proposed algorithm has high computation accuracy and stability, and it is robust to the amplitude of motion of the target image or camera. Field experiments confirm the feasibility of the proposed algorithm for images with non-coplanar conditions with a maximum error of less than 2.3 mm and a peak error below 5 %. In comparison to the traditional coplanar correction method, the CPS method demonstrates lower errors, making it better suited for cases where the FROI and the target plane are not coplanar.

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.

Small object change detection in UAV imagery via a Siamese network enhanced with temporal mutual attention and contextual features: A case study concerning solar water heaters

Small object change detection (SOCD) based on high-spatial resolution (HSR) images is of significant practical value in applications such as the investigation of illegal urban construction, but little research is currently available. This study proposes an SOCD model called TMACNet based on a multitask network architecture. The model modifies the YOLOv8 network into a Siamese network and adds structures, including a feature difference branch (FDB), temporal mutual attention layer (TMAL) and contextual attention module (CAM), to merge differential and contextual features from different phases for the accurate extraction and analysis of small objects and their changes. To verify the proposed method, an SOCD dataset called YZDS is created based on unmanned aerial vehicle (UAV) images of small-scale solar water heaters on rooftops. The experimental results show that TMACNet exhibits strong resistance to image registration errors and building height displacement and prevents error propagation from object detection to change detection originating from overlay-based change detection. TMACNet also provides an enhanced approach to small object detection from the perspective of multitemporal information fusion. In the change detection task, TMACNet exhibits notable F1 improvements exceeding 5.96% in comparison with alternative change detection methods. In the object detection task, TMACNet outperforms the single-temporal object detection models, increasing accuracy with an approximately 1–3% improvement in the AP metric while simplifying the technical process.

Soil moisture content estimation of drip-irrigated citrus orchard based on UAV images and machine learning algorithm in Southwest China

Soil moisture content (SMC), as a pivotal component in the energy and matter exchange processes within the soil-plant-atmosphere continuum, plays a crucial role in surface water dynamics, energy fluxes, and carbon cycling within ecosystems. The development of remote sensing technology has offered new perspectives for monitoring soil moisture at regional scales. Unmanned aerial vehicles (UAV) equip with multispectral have distinct advantages for vegetation monitoring, including rapidity and cost-effectiveness, which has superior applicability and practicality. Therefore, in a 5a "Daya" late-maturing citrus orchard, the vegetation index (VI) and texture feature (TF) information of citrus canopy based on UAV multi-spectral images were extracted, and soil and plant analyzer development (SPAD) of citrus was collected. These different data sources were integrated into the framework of the random forest algorithm (RF) and genetic algorithm-optimized random forest (GA-RF) to evaluate the accuracy of surface SMC (SSMC) estimation in citrus orchard. The Biswas model was utilized to simulate the root zone SMC (RSMC). The spatiotemporal variations of SMC in citrus orchard were analyzed, and the potential of low-cost sensor-equipped drones in rapidly acquiring spatial and temporal distribution information of SMC at a large regional scale was explored. The results indicated that the GA-RF models outperformed the RF models in estimating citrus orchard SMC (with R2 ranging from 0.502 to 0.949 and RMSE ranging from 0.552 % to 3.166 % for GA-RF, compared to R2 ranging from 0.430 to 0.936 and RMSE ranging from 0.587 % to 3.449 % for the RF). The GA-RF models using VI+SPAD as inputs exhibited the best performance for SMC at depths of 5 cm, 10 cm, 20 cm and 40 cm (SMC5, SMC10, SMC20 and SMC40) across citrus growth stages (R2 ranging from 0.793 to 0.949 at 5 cm, R2 ranging from 0.702 to 0.938 at 10 cm, R2 ranging from 0.714 to 0.927 at 20 cm). In bud bust to flowering, young fruit and fruit maturation stages (stage Ⅰ, ⅠⅠ and ⅠⅤ), all models demonstrated good accuracy in estimating SMC at depth of 10 cm (R2 ranging from 0.567 to 0.908 in stage Ⅰ, with R2 ranging from 0.681 to 0.916 in stage ⅠⅠ and R2 ranging from 0.579 to 0.938 in stage ⅠⅤ). In fruit expansion stage (stage III), the models performed best in predicting SMC5 (R2 ranging from 0.698 to 0.861). The Biswas model was constructed to simulate SMC40 by utilizing the inverted SMC10 and SMC20, thereby generating spatiotemporal distribution maps of SMC at different depths in citrus orchard. The SSMC was susceptible to environmental factors, exhibiting significant spatiotemporal heterogeneity. In summary, this study illustrated that the integration of multiple data sources into GA-RF enhanced the estimation performance of SMC at different growth stages of late-maturing citrus orchard in the Southwest China. Additionally, it enabled the rapid and efficient monitoring of spatiotemporal variations in SMC, providing an effective method and practical foundation for precision irrigation and improved water use efficiency.

Unifying Building Instance Extraction and Recognition in UAV Images

Building instance extraction and recognition (BEAR) extracts and further recognizes building instances in unmanned aerial vehicle (UAV) images, holds with paramount importance in urban understanding applications. To address this challenge, we propose a unified network, BEAR-Former. Given the difficulty of building instance recognition due to the small area and multiple instances in UAV images, we developed a novel multi-view learning method, Cross-Mixer. This method constructs a cross-regional branch and an intra-regional branch to, respectively, extract the global context dependencies and local spatial structural details of buildings. In the cross-regional branch, we cleverly employed cross-attention and polar coordinate relative position encoding to learn more discriminative features. To solve the BEAR problem end to end, we designed a channel group and fusion module (CGFM) as a shared encoder. The CGFM includes a channel group encoder layer to independently extract features and a channel fusion module to dig out the complementary information for multiple tasks. Additionally, an RoI enhancement strategy was designed to improve model performance. Finally, we introduced a new metric, Recall@(K, iou), to evaluate the performance of the BEAR task. Experimental results demonstrate the effectiveness of our method.

Improved maize leaf area index inversion combining plant height corrected resampling size and random forest model using UAV images at fine scale

ContextAccurate monitoring of leaf area index (LAI) is conducive to timely and targeted management measures. Unmanned aerial vehicle (UAV) remote sensing provides an important way for non-destructive monitoring of crop leaf area index. ObjectiveIn this study, visible light (RGB) and multispectral remote sensing data from the UAV and ground-measured LAI data from the Plant Canopy Analyzer LAI-2200 C were used to conduct inversion of maize LAI on a fine scale. MethodsTo address the problem of spatial scale mismatch between the spatial resolution of UAV images and the ground-measured LAI, the scale difference between UAV image data and ground-measured data was reduced by removing the outermost ring data measured by the LAI-2200 C instrument, calculating the spatial resolution of the UAV images after resampling based on the height of the plant, and the resampling method based on the circle. Finally, through the above method to resample the UAV images, we extract the vegetation index and canopy height features as the input variables of the random forest model to build the maize LAI inversion model in vegetative stages and reproductive stages respectively, which is referred to as the Vis_H+RF method. Results and conclusionsThe Vis_H+RF method of Tongliao experimental station has an R2 of 0.96 in the vegetative stages and a R2 of 0.61 in the reproductive stages, both of which perform well and have certain migration capabilities. SignificanceThe LAI inversion model constructed based on the method in this study is basically consistent with the actual situation and can provide data support for maize growth monitoring.

Integrating UAV imagery and machine learning via Geographic Object Based Image Analysis (GEOBIA) for enhanced monitoring of Yucca gloriosa in Mediterranean coastal dunes

Effective monitoring and early detection of invasive alien plant species (IAPs) are crucial for mitigating their spread and safeguarding native habitats. Unmanned Aerial Vehicles (UAVs) offer a cost-efficient solution, providing high resolution images. In this study, we aimed to develop a semi-automated methodology using a machine learning algorithm, spatial metrics, and clustering techniques on UAV images to monitor, map, and suggest management measures to counteract Yucca gloriosa, an invasive plant colonizing coastal fixed dunes in central Italy.UAV flights were conducted using two drones: one for the visible spectrum and the other for multispectral bands (Blue, Green, Red, Red Edge, and Near Infrared) along with a Digital Surface Model (DSM). Derived vegetation indices were also utilized. For mapping Y. gloriosa distribution, a Geographic Object Based Image Analysis (GEOBIA) approach was applied to the orthophoto segmentation, followed by a Random Forest algorithm in a training phase, considering three variable combinations (DSM + vegetation indices, DSM + spectral bands, DSM + mixed variables). The most accurate Y. gloriosa map was used to suggest management measures combining the spatial pattern of invaded patches (size, height, isolation level, and aggregation degree) and a mixed clustering approach (hierarchical and partitioning).The results highlighted that the most accurate prediction map was based on the DSM + mixed variables dataset, showing the important role of using a combination of spectral bands and vegetation indices. In all three cases, the DSM emerged as the pivotal variable for discriminating Y. gloriosa from the surrounding environment. Additionally, our results demonstrate the advantages of incorporating vegetation indices in discerning the target invasive alien plant (IAP) from the broader environment, particularly considering its distinctive photosynthesis process and biomass production. From a managerial standpoint, our pilot study indicates that the UAV-based mapping methodology represents an optimal balance between field efforts and costs. This approach allows for the precise identification of containment and removal areas of Y. gloriosa, without compromising the accuracy of the method. The generated prediction maps also hold potential significance for the conservation of coastal dune ecosystems, providing a promising tool for the effective management of invasive species and biodiversity conservation by suggesting management measures for Y. gloriosa.

A proposed plant classification framework for smart agricultural applications using UAV images and artificial intelligence techniques

Utilizing Wireless Sensor Networks (WSNs), Internet of Things (IoTs) sensors, and Unmanned Aerial Vehicles (UAVs), in conjunction with optimization techniques and machine learning algorithms, can present a novel approach to Precision Agricultural (PA) crops. Agriculture has transitioned from traditional legacy systems to incorporate advanced smart technologies. Several complex farming problems are utilized to promote a variety of UAV sensing and Artificial Intelligence (AI) algorithms in PA applications of smart agriculture, especially in developing countries. This paper proposes the design of a conceptual UAV sensing system for crop management using a novel classification framework. UAV-based image datasets can be implemented and expanded to various types of crops. In addition, it can classify various types of rice species and detect weed farms as it consists of a multistage process for identifying and monitoring different crops. The proposed classification framework is evaluated using three different real UAV image datasets compared with Naive Bayes, Decision Tree (DT), Bagging, and Random Forest (RF) techniques. The classification performance metrics obtained are as follows: i) the WeedNet dataset with 100 % F1-score, 100 % recall, 100 % precision, and 100 % accuracy; ii) the Rice Seedling dataset with 99.5 % F1-score, 99.5 % recall, 99.5 % precision, and 99.5 % accuracy; and iii) the Rice Varieties dataset with 97.99 % F1-score, 97.99 % recall, 98.14 % precision, and 97.99 % accuracy. The success rates achieved by the proposed classification framework outperform those of other recent state-of-the-art techniques, demonstrating its efficacy in crop management applications.

Monitoring aboveground organs biomass of wheat and maize: A novel model combining ensemble learning and allometric theory

Accurate monitoring of crop organ biomass facilitates optimizing agronomic strategies to maximize yield or economic benefit. Unmanned aerial vehicle (UAV) is extensively employed for aboveground biomass (AGB) monitoring at the farm scale, but previous studies have mostly concentrated on total AGB rather than individual organ biomass. Furthermore, film-mulched crops, a widely used cropping pattern in northwest China, have received less attention for AGB monitoring. We aim to develop a novel model to precisely estimate the AGB of leaf (AGBLeaf), stem (AGBStem), and reproductive organs (AGBR) by UAV for film-mulched wheat and maize. The maize-wheat rotation field experiments with treatments of five nitrogen application amounts and three planting densities were conducted from 2021 to 2023, respectively. Firstly, we constructed allometric models at jointing, heading (tasseling), and grain filling stages by ground sampling data in 2021–2022. Next, the input feature set was obtained by feature selection methods (Lasso and Boruta) using UAV image data, and three traditional methods (partial least squares, ridge regression, and support vector machine) and three ensemble learning models (random forest, extreme gradient boosting, and local cascade ensemble (LCE)) were trained for AGBLeaf inversion based on the physically-based PROSAIL model simulation dataset. Finally, the optimal AGBLeaf inversion hybrid model was coupled with the allometric model to estimate the AGBStem and AGBR in 2022–2023. The results indicated that both wheat and maize organ biomass conformed to the allometric pattern. While feature selection helped reduce computation and complexity, but didn’t improve monitoring accuracy. The normalized root mean square error (NRMSE) of the optimal hybrid model (PROSAIL + Boruta + LCE) on the measured wheat and maize AGBLeaf datasets were 12.72 %–24.93 % and 19.65 %–25.16 %, respectively. After coupling the allometric model, the coefficient of determination (R2) of wheat and maize AGBStem were 0.64–0.85 and 0.63–0.68, and the NRMSE were 15.05 %–25.28 % and 24.10 %–27.06 %, respectively; and the corresponding R2 of AGBR was 0.67–0.76 and 0.72, and the NRMSE were 16.81 %–22.12 % and 21.66 %, for wheat and maize, respectively. Overall, the novel model performed well in film-mulched wheat and maize, providing a cost-effective approach for organ biomass monitoring. In the future, further validation of the model’s transferability is necessary to increase the potential for generalization in production practice.

UAV image object detection based on self-attention guidance and global feature fusion

Unmanned aerial vehicle (UAV) image object detection has garnered considerable attentions in fields such as Intelligent transportation, urban management and agricultural monitoring. However, it suffers from key challenges of the deficiency in multi-scale feature extraction and the inaccuracy when processing complex scenes and small-sized targets in practical applications. To address this challenge, we propose a novel UAV image object detection network based on self-attention guidance and global feature fusion, named SGGF-Net. First, in order to optimizing feature extraction in global perspective and enhancing target localization precision, the global feature extraction module (GFEM) is introduced by exploiting the self-attention mechanism to capture and integrate long-range dependencies within images. Second, a normal distribution-based prior assigner (NDPA) is developed by measuring the resemblance between ground truth and the priors, which improves the precision of target position matching and thus handle the problem of inaccurate localization of small targets. Furthermore, we design an attention-guided ROI pooling module (ARPM) via a deep fusion strategy of multilevel features for optimizing the integration of multi-scale features and improving the quality of feature representation. Finally, experimental results demonstrate the effectiveness of the proposed SGGF-Net approach.

Evaluation of Sugarcane Crop Growth Monitoring Using Vegetation Indices Derived from RGB-Based UAV Images and Machine Learning Models

Crop monitoring with unmanned aerial vehicles (UAVs) has the potential to reduce field monitoring costs while increasing monitoring frequency and improving efficiency. However, the utilization of RGB-based UAV imagery for crop-specific monitoring, especially for sugarcane, remains limited. This work proposes a UAV platform with an RGB camera as a low-cost solution to monitor sugarcane fields, complementing the commonly used multi-spectral methods. This new approach optimizes the RGB vegetation indices for accurate prediction of sugarcane growth, providing many improvements in scalable crop-management methods. The images were captured by a DJI Mavic Pro drone. Four RGB vegetation indices (VIs) (GLI, VARI, GRVI, and MGRVI) and the crop surface model plant height (CSM_PH) were derived from the images. The fractional vegetation cover (FVC) values were compared by image classification. Sugarcane plant height predictions were generated using two machine learning (ML) algorithms—multiple linear regression (MLR) and random forest (RF)—which were compared across five predictor combinations (CSM_PH and four VIs). At the early stage, all VIs showed significantly lower values than later stages (p < 0.05), indicating an initial slow progression of crop growth. MGRVI achieved a classification accuracy of over 94% across all growth phases, outperforming traditional indices. Based on the feature rankings, VARI was the least sensitive parameter, showing the lowest correlation (r < 0.5) and mutual information (MI < 0.4). The results showed that the RF and MLR models provided better predictions for plant height. The best estimation results were observed withthe combination of CSM_PH and GLI utilizing RF model (R2 = 0.90, RMSE = 0.37 m, MAE = 0.27 m, and AIC = 21.93). This study revealed that VIs and the CSM_PH derived from RGB images captured by UAVs could be useful in monitoring sugarcane growth to boost crop productivity.

Pavement damage characteristics in the permafrost regions based on UAV images and airborne LiDAR data

The rapid degradation of “Xing'an-Baikal permafrost” in Northeast China has led to various road engineering problems. Efficient inspection and control of pavement quality are critical for maintaining the structural integrity of roads and driving safety in cold regions. Taking the Jagdaqi-Walagan section (JWS) of the Jagdaqi-Mo'he Highway as the object, based on field investigation, unmanned aerial vehicle images and airborne LiDAR data, combined with geographical information system, this study analyzed the pavement damage characteristics in mid- to high-latitude permafrost regions, including quantification of damage ratio, extraction of pavement cracks, and evaluation of pavement roughness and driving quality. The results showed that the average pavement damage ratio was 8.80 %, significantly higher in isolated permafrost regions. A higher damage rate in the Jagdaqi-Mo'he direction than the opposite, with a more concentrated cracking distribution. The worst pavement roughness and most severe pavement bumping at repetitive repair locations. This study provides an effective method for investigating pavement damages and analyzing their mechanisms, and explores the application potential of visible light images combined with LiDAR data in frozen soil engineering. The results provide a scientific basis for assessing current highway conditions, enabling scientific maintenance, and evaluating the risk of engineering damages.

An enhanced chlorophyll estimation model with a canopy structural trait in maize crops: Use of multi-spectral UAV images and machine learning algorithm

Leaf chlorophyll concentration (LCC) is a key indicator of leaf nitrogen (N) and changes in canopy structure, particularly the leaf area index (LAI), play a significant role in estimating LCC. Spectral prediction models for chlorophyll are a useful tool for timely nutritional management, particularly in precision agriculture. However, the accuracy of LCC estimation is influenced by the LAI. Considering LAI data as input in the spectral prediction model is still inadequate for improving LCC estimation. This study tested the hypothesis that LCC estimate accuracy could be enhanced by using LAI as an input using high-resolution (5 cm) multi-spectral images from an unmanned aerial vehicle (UAV). For this, maize was grown as a test crop under different nutrient management in the hilly ecosystem of Meghalaya. LCC was measured using laboratory destruction methods from ground sampling that coincided with UAV flights. Machine learning algorithms such as random forest (RF), support vector machine (SVM), and kernel ridge regression (KKR) were employed to develop the LCC estimation model, utilizing band reflectance, vegetation indexes, and measured chlorophyll. The model was assessed for its sensitivity to LCC estimation using LAI data. KKR outperformed other two algorithms (RF and SVM) in accuracy of LCC estimation by >11.0 to 19.0 %. The KKR-derived LCC estimation model was significantly improved by the inclusion of LAI (R2 increased from 0.785 to 0.928 and RMSE decreased from 0.065 to 0.053 mg g−1). The model's reliability was proven on multiple UAV flights for maize crops that are healthy and nutrient-stressed. Thus, LCC models derived from multispectral UAV images using KKR algorithms could benefit the adoption of precision agriculture at field scale in mountain ecosystems.