Enhanced Small‐Object Detection in UAV Images Using Modified YOLOv5 Model

Less effective information is obtained by the object detection network, due to the small size of the detection object in the entire image, the complex background, and the dense object in unmanned aerial vehicle (UAV) images. In response to the difficulties encountered, a small object detection method in UAV images is proposed as an improved YOLOv5-based algorithm in this paper. First, the space-to-depth(SPD) conv module is introduced into the basic feature extraction network, to improve significant loss of image information during downsampling. Then, various attention mechanisms are added, to intensify the acquisition of regions of interest in UAV images. Finally, the multiscale detection module is improved, to enhance the network's ability to detect small objects in UAV images. By conducting experiments on the VisDrone-DET2019 dataset, the test results of the established model show. The improved algorithm achieved a Mean Average Precision (mAP) of 41.8%, which is 7.8% better than the baseline network. In addition, the detection performance is better than most current mainstream target detection algorithms and is of some practical value.

Accurate object detection in unmanned aerial vehicle (UAV) images plays a crucial role in fields such as aviation, transportation, and agriculture. However, UAV images often contain a high proportion of small objects, and the limited resources of UAV platforms make it challenging for existing small object detection algorithms to balance detection performance and resource consumption. To address these issues, a lightweight small object detection algorithm called SRP-YOLO is proposed. Compared to the YOLOv8 network, the following improvements are made: first, a high-resolution detection head is designed; second, a receptive field attention convolution module tailored for small UAV objects is introduced to replace the standard convolution module, enhancing the detection capability for small objects and significantly improving their feature representation; Third, partial convolution is combined with C2f to replace some of the standard convolutions, fully leveraging the high-resolution information from shallow features. Finally, A normalized Gaussian Wasserstein Distance (NWD) metric is introduced to reduce the sensitivity of Intersection over Union (IoU) to minor positional deviations of small objects. On VisDrone-DET2019, SRP-YOLO reduces parameters while improving mAP 0.5 by 7.9%. Generalization tests on TinypersonV2 and RAIVD confirm significant gains in small object detection.

To improve the efficiency and effectiveness of mosaicking unmanned aerial vehicle (UAV) images, we propose in this paper a rapid mosaicking method based on scale-invariant feature transform (SIFT) for mosaicking UAV images used for crop growth monitoring. The proposed method dynamically sets the appropriate contrast threshold in the difference of Gaussian (DOG) scale-space according to the contrast characteristics of UAV images used for crop growth monitoring. Therefore, this method adjusts and optimizes the number of matched feature point pairs in UAV images and increases the mosaicking efficiency. Meanwhile, based on the relative location relationship of UAV images used for crop growth monitoring, the random sample consensus (RANSAC) algorithm is integrated to eliminate the influence of mismatched point pairs in UAV images on mosaicking and to keep the accuracy and quality of mosaicking. Mosaicking experiments were conducted by setting three types of UAV images in crop growth monitoring: visible, near-infrared, and thermal infrared. The results indicate that compared to the standard SIFT algorithm and frequently used commercial mosaicking software, the method proposed here significantly improves the applicability, efficiency, and accuracy of mosaicking UAV images in crop growth monitoring. In comparison with image mosaicking based on the standard SIFT algorithm, the time efficiency of the proposed method is higher by 30%, and its structural similarity index of mosaicking accuracy is about 0.9. Meanwhile, the approach successfully mosaics low-resolution UAV images used for crop growth monitoring and improves the applicability of the SIFT algorithm, providing a technical reference for UAV application used for crop growth and phenotypic monitoring.

With special consideration for complex scenes and densely distributed small objects, this frequently leads to serious false and missed detections for unmanned aerial vehicle (UAV) images in small object detection scenarios. Consequently, we propose a UAV image small object detection algorithm, termed SMA-YOLO. Firstly, a parameter-free simple slicing convolution (SSC) module is integrated in the backbone network to slice the feature maps and enhance the features so as to effectively retain the features of small objects. Subsequently, to enhance the information exchange between upper and lower layers, we design a special multi-cross-scale feature pyramid network (M-FPN). The C2f-Hierarchical-Phantom Convolution (C2f-HPC) module in the network effectively reduces information loss by fine-grained multi-scale feature fusion. Ultimately, adaptive spatial feature fusion detection Head (ASFFDHead) introduces an additional P2 detection head to enhance the resolution of feature maps to better locate small objects. Moreover, the ASFF mechanism is employed to optimize the detection process by filtering out information conflicts during multi-scale feature fusion, thereby significantly optimizing small object detection capability. Using YOLOv8n as the baseline, SMA-YOLO is evaluated on the VisDrone2019 dataset, achieving a 7.4% improvement in mAP@0.5 and a 13.3% reduction in model parameters, and we also verified its generalization ability on VAUDT and RSOD datasets, which demonstrates the effectiveness of our approach.

Robust and rapid image dense matching is the key to large-scale three-dimensional (3D) reconstruction for multiple Unmanned Aerial Vehicle (UAV) images. However, the following problems must be addressed: (1) the amount of UAV image data is very large, but ordinary computer memory is limited; (2) the patch-based multi-view stereo-matching algorithm (PMVS) does not work well for narrow-baseline cases, and its computing efficiency is relatively low, and thus, it is difficult to meet the UAV photogrammetry’s requirements of convenience and speed. This paper proposes an Image-grouping and Self-Adaptive Patch-based Multi-View Stereo-matching algorithm (IG-SAPMVS) for multiple UAV imagery. First, multiple UAV images were grouped reasonably by a certain grouping strategy. Second, image dense matching was performed in each group and included three processes. (1) Initial feature-matching consists of two steps: The first was feature point detection and matching, which made some improvements to PMVS, according to the characteristics of UAV imagery. The second was edge point detection and matching, which aimed to control matching propagation during the expansion process; (2) The second process was matching propagation based on the self-adaptive patch. Initial patches were built that were centered by the obtained 3D seed points, and these were repeatedly expanded. The patches were prevented from crossing the discontinuous terrain by using the edge constraint, and the extent size and shape of the patches could automatically adapt to the terrain relief; (3) The third process was filtering the erroneous matching points. Taken the overlap problem between each group of 3D dense point clouds into account, the matching results were merged into a whole. Experiments conducted on three sets of typical UAV images with different texture features demonstrate that the proposed algorithm can address a large amount of UAV image data almost without computer memory restrictions, and the processing efficiency is significantly better than that of the PMVS algorithm and the matching accuracy is equal to that of the state-of-the-art PMVS algorithm.

Recent advances in structure from motion (SfM) and dense matching algorithms enable surface reconstruction from unmanned aerial vehicle (UAV) images with high spatial resolution, allowing for new insights into earth surface processes. However, accuracy issues are inherent in parallel‐axes UAV image configurations. In this study, the quality of digital elevation models (DEMs) is assessed using images from a simulated UAV flight. Five different SfM tools and three different cameras are compared. If ground control points (GCPs) are not integrated into the adjustment process with parallel‐axes image configurations, significant dome‐effect systematic errors are observed, which can be reduced based on calibration parameters retrieved from a testfield captured with convergent images immediately before or after the UAV flight. A comparison between DEMs of a soil surface generated from UAV images and terrestrial laser‐scanning data show that natural surfaces can be very accurately reconstructed from UAV images, even when GCPs are missing and simple geometric camera models are considered.

Small object detection is a challenging computer vision problem due to their low feature representation in the images and factors such as occlusions and noise. In images captured from a camera mounted on an unmanned aerial vehicle (UAV), objects are usually acquired in small sizes depending on the UAV flight altitude. The state-of-the-art object detectors often have lower detection accuracy with small objects. New approaches of combining features at multi-levels in the network helps in improving the object detection performance. In this paper, we propose a multi-scale approach of low-level feature combinations with deconvolutional modules on a single shot multibox detection (SSD) object detector to improve the small object detection in images acquired from a UAV. The proposed SSD based architecture is evaluated on UAV datasets to compare its performance with the state-of-the-art detectors.

Semantic segmentation of weed and crop images is a key component and prerequisite for automated weed management. For weeds in unmanned aerial vehicle (UAV) images, which are usually characterized by small size and easily confused with crops at early growth stages, existing semantic segmentation models have difficulties to extract sufficiently fine features. This leads to their limited performance in weed and crop segmentation of UAV images. We proposed a fine-grained feature-guided UNet, named FG-UNet, for weed and crop segmentation in UAV images. Specifically, there are two branches in FG-UNet, namely the fine-grained feature branch and the UNet branch. In the fine-grained feature branch, a fine feature-aware (FFA) module was designed to mine fine features in order to enhance the model's ability to segment small objects. In the UNet branch, we used an encoder-decoder structure to realize high-level semantic feature extraction in images. In addition, a contextual feature fusion (CFF) module was designed for the fusion of the fine features and high-level semantic features, thus enhancing the feature discrimination capability of the model. The experimental results showed that our proposed FG-UNet, achieved state-of-the-art performance compared to other semantic segmentation models, with mean intersection over union (MIOU) and mean pixel accuracy (MPA) of 88.06% and 92.37%, respectively. The proposed method in this study lays a solid foundation for accurate detection and intelligent management of weeds. It will have a positive impact on the development of smart agriculture. © 2024 Society of Chemical Industry.

Litchi flowering management is an important link in litchi orchard management. Statistical litchi flowering rate data can provide an important reference for regulating the number of litchi flowers and directly determining the quality and yield of litchi fruit. At present, the statistical work regarding litchi flowering rates requires considerable labour costs. Therefore, this study aims at the statistical litchi flowering rate task, and a combination of unmanned aerial vehicle (UAV) images and computer vision technology is proposed to count the numbers of litchi flower clusters and flushes in a complex natural environment to improve the efficiency of litchi flowering rate estimation. First, RGB images of litchi canopies at the flowering stage are collected by a UAV. After performing image preprocessing, a dataset is established, and two types of objects in the images, namely, flower clusters and flushes, are manually labelled. Second, by comparing the pretraining and testing results obtained when setting different training parameters for the YOLOv4 model, the optimal parameter combination is determined. The YOLOv4 model trained with the optimal combination of parameters tests best on the test set, at which time the mean average precision (mAP) is 87.87%. The detection time required for a single image is 0.043 s. Finally, aiming at the two kinds of targets (flower clusters and flushes) on 8 litchi trees in a real orchard, a model for estimating the numbers of flower clusters and flushes on a single litchi tree is constructed by matching the identified number of targets with the actual number of targets via equation fitting. Then, the data obtained from the manual counting process and the estimation model for the other five litchi trees in the real orchard are statistically analysed. The average error rate for the number of flower clusters is 4.20%, the average error rate for the number of flushes is 2.85%, and the average error for the flowering rate is 1.135%. The experimental results show that the proposed method is effective for estimating the litchi flowering rate and can provide guidance regarding the management of the flowering periods of litchi orchards.

Pine is one of the important forest tree species in China, with a wide distribution and significant economic value, accounting for over 70% of China's artificial forest area. However, due to the rapid spread of pine wilt disease, a large number of pine resources have been damaged, seriously affecting ecological civilization. The most effective method currently is to promptly detect and accurately locate pine trees infected with pine wilt disease (diseased trees). Manual survey and traditional detection method based on satellite remote sensing image have limitations in monitoring complex forest areas. With the advantages of real-time monitoring, low cost, and high spatial resolution, unmanned aerial vehicle (UAV) image has been widely used for agriculture and forestry. Currently, UAV images are mostly used for monitoring diseased trees. To further improve the accuracy of identifying diseased trees, this paper proposes a diseased tree detection method based on improved YOLOv5s network, which combines the YOLOv5s algorithm with Bi-directional Feature Pyramid Network (BiFPN) and Shuffle Attention (SA). In order to demonstrate the role and mutual influence of various modules in the improved YOLOv5s network, the detection results of the improved YOLOv5s algorithm and other algorithms on the same pine wilt disease dataset are compared. The experimental results show that our improved algorithm increased 2.1% in mAP and 6.1% in recall while the quantity of parameter and GFLOPs have only increased by 3.2% and 2.5%, respectively.

The sparse representation error (SRE) exists when the images are represented sparsely. The SRE is particularly large in unmanned aerial vehicles (UAV) images due to the disturbance of the harsh environment or the instability of its flight, which will bring more noise. In the compressed sensing (CS) system, the projected SRE in the compressed measurement will bring a significant challenge to the recovery accuracy of the images. In this work, a new SRE structure is proposed. Following the new structure, a lower sparse representation error (LSRE) is achieved by eliminating groups of sparse representation. With the proposed LSRE modeling, a robust sensing matrix is designed to compress and encrypt the UAV images. Experiments for UAV images are carried out to compare the recovery performance of the proposed algorithm with the existing related algorithms. The results of the proposed algorithm reveal superior recovery accuracy. The new CS framework with the proposed sensing matrix to address the scenario of UAV images with large SRE is dominant.

This paper presents a novel approach to automatically detect and count cars in different aerial images, which can be satellite or unmanned aerial vehicle (UAV) images. Variations in satellite and/or UAV data make it particularly challenging to have a robust method that works properly on a variety of images. A solution based on the bag-of-words (BoW) model is explored in this paper due to its invariance characteristic and highly stable performance in object/scene categorization. Different from categorization tasks, vehicle detection needs to localize the positions of cars in images. To make BoW suitable for this purpose, we extensively improve the methodology in three aspects, namely, by introducing a recently proposed feature representation, i.e., the local steering kernel descriptor, adding spatial structure constraints, and developing an orientation aware scanning mechanism to produce detection with “one-window-one-car” results. Experiments are conducted on various aerial images with large variations, which consist of data from two public databases, e.g., the Overhead Imagery Research Data Set and Vehicle Detection in Aerial Imagery, as well as other satellite and UAV images. The results demonstrate the effectiveness and robustness of the proposed method. Compared with existing techniques, the proposed method is applicable to a wider range of aerial images.

Woody plant encroachment in grassy ecosystems is a widely reported phenomenon associated with negative impacts on ecosystem functions. Most studies of this phenomenon have been carried out in arid and semi-arid grasslands. Therefore, studies in tropical regions, particularly savannas, which are composed of grassland and woodland mosaics, are needed. Our objective was to evaluate the accuracy of woody encroachment classification in the Brazilian Cerrado, a tropical savanna. We acquired dry and wet season unmanned aerial vehicle (UAV) images using RGB and multispectral cameras that were processed by the support vector machine (SVM), decision tree (DT), and random forest (RF) classifiers. We also compared two validation methods: the orthomosaic and in situ methods. We targeted two native woody species: Baccharis retusa and Trembleya parviflora. Identification of these two species was statistically (p < 0.05) most accurate in the wet season RGB images classified by the RF algorithm, with an overall accuracy (OA) of 92.7%. Relating to validation assessments, the in situ method was more susceptible to underfitting scenarios, especially using an RF classifier. The OA was higher in grassland than in woodland formations. Our results show that woody encroachment classification in a tropical savanna is possible using UAV images and field surveys and is suggested to be conducted during the wet season. It is challenging to classify UAV images in highly diverse ecosystems such as the Cerrado; therefore, whenever possible, researchers should use multiple accuracy assessment methods. In the case of using in situ accuracy assessment, we suggest a minimum of 40 training samples per class and to use multiple classifiers (e.g., RF and DT). Our findings contribute to the generation of tools that optimize time and cost for the monitoring and management of woody encroachment in tropical savannas.

Multi-view scene matching refers to the establishment of a mapping relationship between images captured from different perspectives, such as those taken by unmanned aerial vehicles (UAVs) and satellites. This technology is crucial for the geo-localization of UAV views. However, the geometric discrepancies between images from different perspectives, combined with the inherent computational constraints of UAVs, present significant challenges for matching UAV and satellite images. Additionally, the imbalance of positive and negative samples between drone and satellite images during model training can lead to instability. To address these challenges, this study proposes a novel and efficient cross-view geo-localization framework called MSM-Transformer. The framework employs the Dual Attention Vision Transformer (DaViT) as the core architecture for feature extraction, which significantly enhances the modeling capacity for global features and the contextual relevance of adjacent regions. The weight-sharing mechanism in MSM-Transformer effectively reduces model complexity, making it highly suitable for deployment on embedded devices such as UAVs and satellites. Furthermore, the framework introduces a contrastive learning-based Symmetric Decoupled Contrastive Learning (DCL) loss function, which effectively mitigates the issue of sample imbalance between satellite and UAV images. Experimental validation on the University-1652 dataset demonstrates that MSM-Transformer achieves outstanding performance, delivering optimal matching results with a minimal number of parameters.

In recent years, research on Unmanned Aerial Vehicles (UAVs) has developed rapidly. Compared to traditional remote-sensing images, UAV images exhibit complex backgrounds, high resolution, and large differences in object scales. Therefore, UAV object detection is an essential yet challenging task. This paper proposes a multi-scale object detection network, namely YOLO-DroneMS (You Only Look Once for Drone Multi-Scale Object), for UAV images. Targeting the pivotal connection between the backbone and neck, the Large Separable Kernel Attention (LSKA) mechanism is adopted with the Spatial Pyramid Pooling Factor (SPPF), where weighted processing of multi-scale feature maps is performed to focus more on features. And Attentional Scale Sequence Fusion DySample (ASF-DySample) is introduced to perform attention scale sequence fusion and dynamic upsampling to conserve resources. Then, the faster cross-stage partial network bottleneck with two convolutions (named C2f) in the backbone is optimized using the Inverted Residual Mobile Block and Dilated Reparam Block (iRMB-DRB), which balances the advantages of dynamic global modeling and static local information fusion. This optimization effectively increases the model’s receptive field, enhancing its capability for downstream tasks. By replacing the original CIoU with WIoUv3, the model prioritizes anchoring boxes of superior quality, dynamically adjusting weights to enhance detection performance for small objects. Experimental findings on the VisDrone2019 dataset demonstrate that at an Intersection over Union (IoU) of 0.5, YOLO-DroneMS achieves a 3.6% increase in mAP@50 compared to the YOLOv8n model. Moreover, YOLO-DroneMS exhibits improved detection speed, increasing the number of frames per second (FPS) from 78.7 to 83.3. The enhanced model supports diverse target scales and achieves high recognition rates, making it well-suited for drone-based object detection tasks, particularly in scenarios involving multiple object clusters.