Cloud detection is one of the key technologies in the field of remote sensing. Although extensive deep learning-based cloud detection methods achieve good performance, their detection results in confusing areas such as cloud boundaries and thin clouds are often not satisfactory due to the potential inter-class similarity and intra-class inconsistency of objects. To this end, we propose a Boundary-Aware Bilateral Fusion network (BABFNet), which effectively enhances cloud detection in confusing areas by introducing a boundary prediction branch as an auxiliary. To avoid the loss of details, the boundary prediction branch is designed to run at full resolution with a shallow architecture, while some Semantic Enhancement Modules (SEMs) are used to supplement high-level semantic information by introducing multi-level encoder features of the cloud detection branch. This feature sharing in turn drives the cloud detection branch to focus more on cloud boundaries during training. At the end of the network, a Bilateral Fusion Module (BFM) is added for information complementarity between features from these two branches. The features from the cloud detection branch provide multi-scale features to the boundary prediction branch for more accurate boundary prediction, while the features from the boundary prediction branch further serve as prior knowledge to help the cloud detection branch aggregate contextual information. To verify the effectiveness of the proposed method, we select four different networks as cloud detection branches and conduct comparative experiments on two public datasets, GF-1 WFV and MODIS. The experimental results show that the proposed method significantly enhances cloud detection in confusing areas.

It has become a consensus to improve the measurement accuracy of microelectromechanical systems (MEMS) inertial measurement unit (IMU) through IMU array composed of multiple low-cost IMUs. In order to meet the accuracy requirements of the inertial navigation MEMS IMU array, it can be divided into two steps: the first step is to calibrate a single IMU in the array, and the second step is to isomorphically fuse the gyroscopes and accelerometers of multiple IMUs. This paper discusses an innovative calibration and fusion technology based on the accompanying test method and weighted fusion technology, which optimizes the two steps to improve the accuracy of the MEMS IMU array. The calibration is more efficient and does not require complex precision calibration equipment. We verified this method as a useful scheme by designing a principle prototype and experimental platform.

Image classification is one of the mainstream tasks of computer vision. However, the most existing methods use labels of the same granularity level for training. This leads to ignoring the hierarchy that may help to differentiate different visual objects better. Embedding hierarchical information into the convolutional neural networks (CNNs) can effectively regulate the semantic space and thus reduce the ambiguity of prediction. To this end, a multi‐task learning framework, named as Hierarchical Bilinear Convolutional Neural Network (HB‐CNN), is developed by seamlessly integrating CNNs with multi‐task learning over the hierarchical visual concept structures. Specifically, the labels with a tree structure are used as the supervision to hierarchically train multiple branch networks. In this way, the model can not only learn additional information (e.g. context information) as the coarse‐level category features, but also focus the learned fine‐level category features on the object properties. To smoothly pass hierarchical conceptual information and encourage feature reuse, a connectivity pattern is proposed to connect features at different levels. Furthermore, a bilinear module is embedded to generalise various orderless texture feature descriptors so that our model can capture more discriminative features. The proposed method is extensively evaluated on the CIFAR‐10, CIFAR‐100, and ‘Orchid’ Plant image sets. The experimental results show the effectiveness and superiority of our method.

With the continuous improvement of high-resolution remote-sensing image-acquisition technologies, image quality and resolution are constantly improved, which greatly promotes the development of object detection in large-scale remote sensing images. However, due to the memory constraints of Graphics Processing Units (GPUs), directly inputting large-scale remotesensing images into detectors poses significant challenges. Downsampling these images to sizes suitable for detectors often leads to the loss of critical details. Current mainstream object detection methods for large-scale images typically divide large images into patches and detect the objects of interest in all patches. These methods allocate computational resources uniformly on each patch, regardless of the presence of objects or variations in their density and size. This not only disrupts the delicate balance between accuracy and speed but also results in inefficient utilization of computational resources. To address these challenges, we propose a Situation-Aware Network (SANet) for object detection in large-scale remote sensing images. SANet introduces a Situation-Aware (SA) module designed to filter out invalid information in large-scale images, thereby reducing computational complexity. Moreover, it integrates a RegionAdaptive (RA) module, which dynamically adjusts the level of granularity for object regions and allocates them to appropriate detectors, thus minimizing resource wastage. Our method has been tested on challenging datasets such as DOTA-v1.0, DOTAv1.5, DOTA-v2.0, and TT100K, demonstrating consistent advantages in both speed and accuracy compared to existing object detection methodologies.