Landslide susceptibility assessment (LSA) based on machine learning methods has been widely used in landslide geological hazard management and research. However, the problem of sample imbalance in landslide susceptibility assessment, where landslide samples tend to be much smaller than non-landslide samples, is often overlooked. This problem is often one of the important factors affecting the performance of landslide susceptibility models. In this paper, we take the Wanzhou district of Chongqing city as an example, where the total number of data sets is more than 580,000 and the ratio of positive to negative samples is 1:19. We oversample or undersample the unbalanced landslide samples to make them balanced, and then compare the performance of machine learning models with different sampling strategies. Three classic machine learning algorithms, logistic regression, random forest and LightGBM, are used for LSA modeling. The results show that the model trained directly using the unbalanced sample dataset performs the worst, showing an extremely low recall rate, indicating that its predictive ability for landslide samples is extremely low and cannot be applied in practice. Compared with the original dataset, the sample set optimized through certain methods has demonstrated improved predictive performance across various classifiers, manifested in the improvement of AUC value and recall rate. The best model was the random forest model using over-sampling (O_RF) (AUC = 0.932).

Object detection is one of the most widespread applications for numerous Unmanned Aerial Vehicle (UAV) tasks. Due to the shooting angle and flying height of the UAV, compared with general scenarios, small objects account for a large proportion of aerial images, and common object detectors are not extremely effective in aerial images. Moreover, since the computing resources of UAV platforms are generally limited, the deployment of common detectors with a large number of parameters on UAV platforms is difficult. This paper proposes a lightweight object detector YOLO-UAVlite for aerial images. Firstly, the spatial attention module and coordinate attention module are modified and combined to form a novel Spatial-Coordinate Self-Attention (SCSA) module, which integrates spatial, location, and channel information to enhance object representation. On this basis, we construct a lightweight backbone, named SCSAshufflenet, which combines the Enhanced ShuffleNet (ES) network with the proposed SCSA module to improve feature extraction and reduce model size. Secondly, we propose an improved feature pyramid model, namely Slim-BiFPN, where we construct new lightweight convolutional blocks to reduce the information loss during the feature map fusion process while reducing the model weights. Finally, the localization loss function is modified to increase the bounding box regression rate while improving the localization accuracy. Extensive experiments conducted on the VisDrone-DET2021 dataset indicate that, compared with the YOLOv5-N baseline, the proposed YOLO-UAVlite reduces the number of parameters by 25.8% and achieves gains of 10.9% in mAP0.50. Compared with other lightweight detectors, both the mAP and the number of parameters are improved.

Recently, object detection has become a research hotspot in computer vision, which often detects regular images with small viewing angles. In order to obtain a field of view without blind spots, fisheye cameras, which have distortions and discontinuities, have come into use. The fisheye camera, which has a wide viewing angle, and an unmanned aerial vehicle equipped with a fisheye camera are used to obtain a field of view without blind spots. However, distorted and discontinuous objects appear in the captured fisheye images due to the unique viewing angle of fisheye cameras. It poses a significant challenge to some existing object detectors. To solve this problem, this paper proposes a PGDS-YOLOv8s model to solve the issue of detecting distorted and discontinuous objects in fisheye images. First, two novel downsampling modules are proposed. Among them, the Max Pooling and Ghost’s Downsampling (MPGD) module effectively extracts the essential feature information of distorted and discontinuous objects. The Average Pooling and Ghost’s Downsampling (APGD) module acquires rich global features and reduces the feature loss of distorted and discontinuous objects. In addition, the proposed C2fs module uses Squeeze-and-Excitation (SE) blocks to model the interdependence of the channels to acquire richer gradient flow information about the features. The C2fs module provides a better understanding of the contextual information in fisheye images. Subsequently, an SE block is added after the Spatial Pyramid Pooling Fast (SPPF), thus improving the model’s ability to capture features of distorted, discontinuous objects. Moreover, the UAV-360 dataset is created for object detection in fisheye images. Finally, experiments show that the proposed PGDS-YOLOv8s model on the VOC-360 dataset improves mAP@0.5 by 19.8% and mAP@0.5:0.95 by 27.5% compared to the original YOLOv8s model. In addition, the improved model on the UAV-360 dataset achieves 89.0% for mAP@0.5 and 60.5% for mAP@0.5:0.95. Furthermore, on the MS-COCO 2017 dataset, the PGDS-YOLOv8s model improved AP by 1.4%, AP${}_{50}$ by 1.7%, and AP${}_{75}$ by 1.2% compared with the original YOLOv8s model.