Most LiDAR odometry and SLAM systems construct maps in point clouds, which are discrete and sparse when zoomed in, making them not directly suitable for navigation. Mesh maps represent a dense and continuous map format with low memory consumption, which can approximate complex structures with simple elements, attracting significant attention of researchers in recent years. However, most existing methods operate under a static environment assumption. In effect, moving objects cause ghosting, degrading the quality of meshing. To address these issues, we propose a plug-and-play meshing module adapting to dynamic environments, which can easily integrate with various LiDAR odometry to generally improve the pose estimation accuracy of odometry. In our meshing module, a novel two-stage coarse-to-fine dynamic removal method is designed to effectively filter dynamic objects, generating consistent, accurate, and dense mesh maps. To the best of our knowledge, this is the first mesh construction method with explicit dynamic removal. Additionally, sliding window-based keyframe aggregation and adaptive downsampling strategies are used to ensure the uniformity of point cloud, benefiting for Gaussian process in mesh construction. We evaluate the localization and mapping accuracy on six publicly available datasets. Extensive experiments demonstrate the superiority of our method compared with the state-of-the-art algorithms. The code and introduction video are publicly available at https://yaepiii.github.io/CAD-Mesher/.

Place recognition is an important component for autonomous robot navigation. Many existing LiDAR-based place recognition methods encode the structural information of 3D LiDAR data into 2D image representations. However, most of these intermediates only exploit the projection in a single view, ignoring a great amount of useful information. In this paper, a compact fusion-view image representation of LiDAR point cloud is proposed to extract important structural information from different views. Our proposed method generates such fusion-view images using the corresponding geometric information among points highlighting the edges of objects. It then extracts texture features encoding the shapes and layouts of scene elements into global descriptors, where regional features are designed to adapt local discrepancies caused by seasonal changes. Extensive experiments on the Oxford RobotCar, NCLT, UTBM datasets and our cross-season dataset validate the proposed method and demonstrate its superior generalization performance under different LiDAR sensors and season shifts. Moreover, our proposed method can operate online with a single CPU, making it suitable for resource-limited real robot platforms.

In this study, we address the challenge of traversability analysis for autonomous vehicles in diverse environments, leveraging LiDAR sensors. We propose the Transformer-Voxel-Bird’s eye view (BEV) Network (TVBNet), a novel dual-branch framework designed to increase the accuracy and versatility of such analyses in both urban and off-road conditions. TVBNet first preprocesses raw point cloud data through voxelization and the generation of a BEV. It incorporates a Transformer network with a rotational attention mechanism to aggregate features from multiple point cloud frames, capturing long-range correlations both within and between point clouds. Additionally, a Swin Transformer extracts the relative positional relationships in the BEV projection, facilitating a comprehensive understanding of the scene. The fusion of data from both branches via a multisource feature fusion module, which employs a context aggregation mechanism based on a residual structure, allows for robust local to global contextual understanding. This approach not only improves the extraction of correlation features between 2D BEV and 3D voxel data but also demonstrates superior performance on the challenging off-road dataset RELLIS-3D and the urban dataset SemanticKITTI.

Automatically detecting human mental workload to prevent mental diseases is highly important. With the development of information technology, remote detection of mental workload is expected. The development of artificial intelligence and Internet of Things technology will also enable the identification of mental workload remotely based on human physiological signals. In this article, a method based on the spatial and time-frequency domains of electroencephalography (EEG) signals is proposed to improve the classification accuracy of mental workload. Moreover, a hybrid deep learning model is presented. First, the spatial domain features of different brain regions are proposed. Simultaneously, EEG time-frequency domain information is obtained based on wavelet transform. The spatial and time-frequency domain features are input into two types of deep learning models for mental workload classification. To validate the performance of the proposed method, the Simultaneous Task EEG Workload public database is used. Compared with the existing methods, the proposed approach shows higher classification accuracy. It provides a novel means of assessing mental workload.

Automatically predicting cardiovascular and cerebrovascular events (CCEs) is a key technology that can prevent deaths and disabilities. Herein, we propose predicting CCE occurrences based on heart rate variability (HRV) analysis and a deep belief network (DBN). The proposed prediction algorithm uses eight novel HRV signal features, which are calculated based on the following steps. First, the instantaneous amplitude (IA), instantaneous frequency (IF), and instantaneous phase (IP) are calculated for the HRV signals. Second, the high-order cumulant is estimated for the HRV and its IA, IF, and IP. Third, a high-order singular entropy is calculated to measure the fluctuation in signals. Fourth, eight novel features are obtained and processed using a DBN classifier designed for CCE prediction. The DBN classification method, with the novel HRV features, outperformed existing methods in terms of accuracy. Thus, the scheme proposed herein provided a novel direction for predicting CCEs.

Obstructive sleep apnea (OSA) syndrome is a common sleep disorder and a key cause of cardiovascular and cerebrovascular diseases that seriously affect the lives and health of people. The development of Internet of Medical Things (IoMT) has enabled the remote diagnosis of OSA. The physiological signals of human sleep are sent to the cloud or medical facilities through Internet of Things, after which diagnostic models are employed for OSA detection. In order to improve the detection accuracy of OSA, in this study, a novel OSA detection system based on manually generated features and utilizing a parallel heterogeneous deep learning model in the context of IoMT is proposed, and the accuracy of the proposed diagnostic model is investigated. The OSA recognition scheme used in our model is based on short-term heart rate variability (HRV) signals extracted from ECG signals. First, the HRV signals and the linear and nonlinear features of HRV are combined into a one-dimensional (1-D) sequence. Simultaneously, a two-dimensional (2-D) HRV time-frequency spectrum image is obtained. The 1-D data sequences and 2-D images are coded in different branches of the proposed deep learning network for OSA diagnosis. To validate the performance of the proposed scheme, the Physionet Apnea-ECG public database is used. The proposed scheme outperforms the existing methods in terms of accuracy and provides a novel direction for OSA recognition.

This paper presents a tightly coupled pipeline, which efficiently fuses measurements of LiDAR, camera, IMU, encoder, and GNSS to estimate the robot state and build a map even in challenging situations. The depth of visual features is extracted by projecting the LiDAR point cloud and ground plane into image. We select the tracked high-quality visual features and LiDAR features and tightly coupled the pre-integrated values of the IMU and the encoder to optimize the state increment of a robot. We use the estimated relative pose to re-evaluate the matching distance between features in the local window and remove dynamic objects and outliers. In the mapping node, we use refined features and tightly coupled the GNSS measurements, increment factors, and local ground constraints to further refine the robot’s global state by aligning LiDAR features with the global map. Furthermore, the method can detect sensor degradation and automatically reconfigure the optimization process. Based on a six-wheeled ground robot, we perform extensive experiments in both indoor and outdoor environments and demonstrated that the proposed GR-Fusion outperforms state-of-the-art SLAM methods in terms of accuracy and robustness.

In recent years, many excellent SLAM methods based on cameras, especially the camera-IMU fusion (VIO), have emerged, which has greatly improved the accuracy and robustness of SLAM. However, we find through experiments that most of the existing VIO methods perform well on drones or drone datasets, but for ground robots on complex terrain, they cannot continuously provide accurate and robust localization results. Some researchers have proposed methods for ground robots, but most of them have limited applications due to the assumption of plane motion. Therefore, this paper proposes GR-SLAM for the localization of ground robots on complex terrain, which can fuse camera, IMU, and encoder data in a tightly coupled scheme to provide accurate and robust state estimation for robots. First, an odometer increment model is proposed, which can fuse the encoder and IMU data to calculate the robot pose increment on manifold, and calculate the frame constraints through the pre-integrated increment. Then we propose an evaluation algorithm for multi-sensor measurements, which can detect abnormal data and adjust its optimization weight. Finally, we implement a complete factor graph optimization framework based on sliding window, which can tightly couple camera, IMU, and encoder data to perform state estimation. Extensive experiments are conducted based on a real ground robot and the results show that GR-SLAM can provide accurate and robust state estimation for ground robots.

Underwater exploration has become an active research area over the past few decades. The image enhancement is one of the challenges for those computer vision based underwater researches because of the degradation of the images in the underwater environment. The scattering and absorption are the main causes in the underwater environment to make the images decrease their visibility, for example, blurry, low contrast, and reducing visual ranges. To tackle aforementioned problems, this paper presents a novel method for underwater image enhancement inspired by the Retinex framework, which simulates the human visual system. The term Retinex is created by the combinations of “Retina” and “Cortex”. The proposed method, namely LAB-MSR, is achieved by modifying the original Retinex algorithm. It utilizes the combination of the bilateral filter and trilateral filter on the three channels of the image in CIELAB color space according to the characteristics of each channel. With real world data, experiments are carried out to demonstrate both the degradation characteristics of the underwater images in different turbidities, and the competitive performance of the proposed method.

This paper presents a novel method to conduct camera pose estimation though combining Kinect and Perspective-n-points algorithms. Most existing camera pose estimation methods suffer from the errors caused by inevitable outliers between 2D-3D correspondences. To this end, we propose to use a random down sampling process to deal with outliers in this paper. The proposed method is divided into two main steps, which are 2D-3D correspondences generation and pose estimation. The method has been tested in a real project, and the experiment has shown encouraging results compared to the ground truth.

Human face analysis is the basis for many other computer vision tasks, such as camera surveillance, entrance authorization and age estimation. With 3D face models, the vision task based on facial analysis can usually achieve a higher accuracy than the 2D cases since it provides more information with the additional dimension. However, most existing 3D face reconstruction methods suffer from complicated processing and high computation. This paper presents a novel method that simplifies the 3D face reconstruction process with only one shot of Kinect data. The output of the system is a high density of 3D face point cloud with smoother surface. This provides rich details of the human face for other computer vision tasks. Experiments with real world data show promising results using the proposed method.