The sea surface temperature (SST) is an important parameter of the energy balance on Earth’s surface. SST prediction crucial to marine production, protection, and climate prediction. However, current model still has low precision poor stability. In this study, a medium long-term designed basis gated recurrent unit (GRU) neural network algorithm. This captures time regularity by using GRU layer outputs predicted results through fully connected layer. Bohai Sea, which characterized large...

Monitoring of offshore aquaculture zones is important to marine ecological environment protection and maritime safety security. Remote sensing technology has the advantages large-area simultaneous observation strong timeliness, which provide normalized monitoring zones. Aiming at problems weak generalization ability low recognition rate in signal environments traditional target algorithm, this paper proposes a method for automatic extraction fish cage floating raft based on semantic...

Accurate navigation is crucial in the construction of intelligent orchards, and need for vehicle accuracy becomes even more important as production refined. However, traditional methods based on global satellite system (GNSS) 2D light detection ranging (LiDAR) can be unreliable complex scenarios with little sensory information due to tree canopy occlusion. To solve these issues, this paper proposes a 3D LiDAR-based method trellis orchards. With use LiDAR simultaneous localization mapping...

Urban waterlogging is mainly caused by garbage clogging the sewer manhole covers. If amount of at a cover can be detected, together with an early warning signal when large enough, it will great significance in preventing urban from occurring. Based on YOLOX algorithm, this paper accomplishes identifying covers and building flood control system that automatically recognize monitor accumulation garbage. This also display statistical results send information. During identification, lead to...

Hyperspectral remote sensing images with high spatial resolution (H <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> imagery) have an abundant spatial-spectral information, holding tremendous potential for fine-grained monitoring and classification. However, challenges such as heterogeneity, severe intra-class spectral variability poor signal-to-noise ratio especially in unmanned aerial vehicle (UAV) hyperspectral imagery constrain hinder...

Miniaturized and rapid blood coagulation assay technologies are critical in many clinical settings. In this paper, we present a ZnO film bulk acoustic resonator for the kinetic analysis of human coagulation. The operated thickness shear resonance mode at 1.4 GHz. When contacted liquid environment, viscous loading effect was considered as additional resistance inductance equivalent circuits, resulting linear relationship with slope approximately -217 kHz/cP between viscosity frequency...

ABSTRACT Research in moving object tracking has shown significant progress towards application recent years. However, single sensors suffer from illumination variations for the vision sensor and low directional resolution radar. In this paper, we propose a method that fuses radar data. First, false objects are filtered out. Second, an adaptive background subtraction is used to detect candidate regions images. Finally, determined when effective regions.

Wetland ecosystems are essential to the preservation of biodiversity. Plateau wetland vital components ecosystems, characterized by diverse types and fragmented land distribution. In extraction plateau wetlands, there such issues as inaccuracy in classification, inadequately fine categories, difficulty extracting vegetated wetlands. The aim this study was establish a new classification framework for detailed information about Dianchi Basin used area. Using Gaofen-2 (GF-2) imagery from...

The classification of remote sensing images with high spatial resolution requires considerable training samples, but the process sample making is slow and laborious. How to guarantee accuracy supervised under condition limited samples an urgent problem be solved in field classification. For addressing this problem, we propose improved conditional generative adversarial network edge feature (ECGAN) augment data for article. On basis network, factors interclass boundaries intraclass edges are...

Timely monitoring of marine aquaculture has considerable significance for ecological protection and maritime safety security. Considering that supervised learning needs to rely on a large number training samples the characteristics intensive regular distribution laver zone, in this paper, an inaccurate classification model based fully convolutional neural network conditional random filed (FCN-CRF) is designed study zone Lianyungang, Jiangsu Province. The proposed can extract calculate area...

Hyperspectral images with very high resolution (VHR-HSI) have become considerably valuable due to their abundant spectral and spatial details. Classification of hyperspectral (HSIs) is a basic important procedure for diverse applications. However, low interclass variability intraclass in VHR-HSI, shadows, pedestrians, signal-to-noise ratio increase the fuzziness different categories. To address known challenges VHR-HSI classification, an effective classification method based on...

Mariculture is crucial in environmental monitoring and safety assurance of marine environments. Certain mariculture areas are often partially or completely submerged water, which causes the target signal to be extremely weak difficult detect. A method recognition classification based on convolutional neural network called semantic segmentation can fully consider space spectrum context information. Therefore, this study proposes a extraction basis multisource feature fusion, such as nNDWI G/R...

Optical remote sensing is an effective means of water depth measurement, but the current approach mainstream bathymetric retrieval requires a large amount onsite measurement data. Such data are hard to obtain from places where underwater terrains complicated and unsteady, sea areas affected by issues with rights conflicts interest. In recent years, emergence airborne light detection ranging (LiDAR) provided new technical for field survey. this study, inversion was carried out around island...

Feature extraction (FE), an important preprocessing step in hyperspectral image (HSI) classification, has received growing attention the remote sensing community. In recent years, FE ability of deep learning (DL) methods been widely recognized. However, most DL models focus on training networks with strong nonlinear mapping ability. They fail to explore intrinsic manifold structure HSI, and their performance depends large size labeled set. To address above problems, a novel approach, termed...

Miniaturization, flexibility and high coding capacity design are very important while considering the development of chipless tags. In this paper, a flexible high-encoding radio frequency identification (RFID) tag with compact structure is designed using Polyimide, material high-temperature resistance as substrate. The works in band 1.5∼10.5 GHz, density 4.348 bits/cm2 obtained four alternating nested improved U-shaped resonance structures. Frequency shift realized by changing structural...