Non-contact and active vegetation or plant parameters extraction using hyperspectral information is a prospective research direction among the remote sensing community. Hyperspectral LiDAR (HSL) an instrument capable of acquiring spectral spatial actively, which could mitigate environmental illumination influence on collection. However, HSL usually has limited resolution coverage, vital for parameter extraction. In this paper, to broaden range increase resolution, Acousto-optical Tunable...

This paper offers a solution to challenge navigation in the indoor environment by making use of existing infrastructure. Estimating pedestrian trajectory using dead reckoning (PDR) and WiFi is very popular technique. However, cumulative errors mismatching are major problems PDR fingerprint matching, respectively. heading used as state transition equation, step length matching results observation equations. A federated particle filter (FPF) based on principle information sharing proposed...

LiDAR has become a vital sensor for autonomous driving scene understanding. To meet the accuracy and speed of point clouds semantic segmentation, an efficient model ACPNet is proposed in this paper. In feature extraction stage, backbone constructed with asymmetric convolutions, so skeleton square convolution kernel enhanced, which leads to greater robustness target rotation. Moreover, contextual enhancement module designed extract richer features. During training, global scaling translation...

Fusing LiDAR point cloud and camera image for 3D object detection in autonomous driving has emerged as a captivating research avenue. The core challenge of multimodal fusion is how to seamlessly fuse with 2D image. Although current approaches exhibit promising results, they often rely solely on at either the data level, feature or there still room improvement utilization information. We present an advanced effective framework called EPAWFusion fusing both level level. model consists three...

Thermoelectric cooler has dynamic thermoelectric performance under complex environment. A model of is derived using small-signal linearization method. It shows that the one zero and two poles. The shown to vary with different operating condition. Based on average linear a cooler, temperature control system designed for cold end adaptive NN-PID algorithm. step response tests show controller satisfying steady performance. In room environment, time cooling down 10□ around 70s error very small....

There will always exist a compromise between the network bandwidth and Web contents in mobile Internet era. images need to be compressed efficiently by servers rendered fleetly clients. are several alternatives for compression of images, from long-existing coding standard JPEG emerging solution WebP. This paper presents comparative analysis on both lossless lossy Images based designed objective experiments, employing performance metrics including Compression Ratio (CR), Structural Similarity...

Purpose At present, smartphones are embedded with accelerometers, gyroscopes, magnetometers and WiFi sensors. Most researchers have delved into the use of these sensors for localization. However, there still many problems in reducing fingerprint mismatching fusing positioning data. The purpose this paper is to improve accuracy by designing a weighted fusion algorithm. Design/methodology/approach For problem magnetic caused singularity fingerprint, derivative Euclidean distance uses adjacent...

Abstract The use of dead reckoning and fingerprint matching for navigation is a widespread technical method. However, mismatching low fusion accuracy are prevalent issues in indoor systems. This work presents an improved dynamic time warping chicken particle filter to handle these two challenges. To generate the Horizontal Vertical (HV) fingerprint, pitch roll employed instead original intensity extract horizontal vertical components magnetic field fingerprint. Derivative employs HV its...

Abstract Based on the analysis of accurate estimation method state charge (SOC) lithium battery for electric vehicles, aiming at shortcomings back propagation (BP) neural network model, an algorithm based Improved Particle Swarm Optimization (IPSO) is proposed to optimize parameters BP network. In this algorithm, particle swarm optimization optimized by introducing shrinkage factor limit speed, so as determine initial Finally, model established using data set published NASA PCoE, and...

With the application of random sampling method in down-sampling point clouds data, processing speed has been greatly improved. However, utilization semantic information is still insufficient. To address this problem, we propose a cloud segmentation network called MFFRand (Multi-Scale Feature Fusion Based on RandLA-Net). RandLA-Net, multi-scale feature fusion module developed, which stacked by encoder-decoders with different depths. The maps extracted are continuously concatenated and fused....

WiFi fingerprint-based positioning is a method for indoor localization with the advent of widespread deployment and Internet Things. However, single fingerprint has problems mismatch, unstable signal strength limited accuracy. Aiming to address these issues, this paper proposes fusion algorithm combining pedestrian dead reckoning (PDR). Firstly, particle swarm optimization (PSO) model utilized optimize weighted k-nearest neighbors (WKNN) in part. Additionally, artemisinin (AO) used filter...

Existing 3D object detection frameworks in sensor-based applications heavily rely on large-scale annotated data to achieve optimal performance. However, obtaining such annotations from sensor data—like LiDAR or image sensors—is both time-consuming and costly. Semi-supervised learning offers an efficient solution this challenge holds significant potential for sensor-driven artificial intelligence (AI) applications. While it reduces the need labeled data, semi-supervised still depends a small...

In order to solve the problem of autonomous vehicles driving safely on road while searching for optimal paths avoid traffic congestion and obstruction, this paper establishes an path planning model based image feature extraction. The consists four parts: selection key turning points, interpolation between definition evaluation function, global search. First, map information provided by network, detection algorithm was proposed, points route were selected stored in open table. Then,...

To mitigate the safety risks and economic losses caused by wheel damage, this paper proposes an interval valued fuzzy inference-based sound analysis method for damage detection. Firstly, sets are defined to represent various levels of severity. A similarity calculation is then designed, based on sets, assess level components. Moreover, OWA operator employed assign higher weights key features while reducing influence noise or redundant features. Finally, a double-threshold inference approach...

In the fields of agriculture and forestry, Normalized Difference Vegetation Index (NDVI) is a critical indicator for assessing physiological state plants. Traditional imaging sensors can only collect two-dimensional vegetation distribution data, while dual-wavelength LiDAR technology offers capability to capture vertical information, which essential forest structure recovery precision management. However, existing systems face challenges in detecting echoes at two wavelengths, typically...

In camera-based bird’s-eye view (BEV) 3D object detection, non-maximum suppression (NMS) plays a crucial role. However, traditional NMS methods become ineffective in BEV scenarios where the predicted bounding boxes of small instances often have no overlapping areas. To address this issue, paper proposes intersection over union (IoU) computation method based on relative position and absolute spatial information, referred to as B-IoU. Additionally, circular search method, called B-Grouping, is...

Single-vehicle light detection and ranging (LiDAR) has limitations in capturing comprehensive environmental information. The advancement of vehicle-to-infrastructure (V2I) collaboration presents a potent solution to this challenge. During the collaboration, point cloud registration precisely aligns data from various LiDARs, effectively mitigating constraints associated with collection by single-vehicle LiDAR. Registration furnishes autonomous vehicles more dependable understanding....