This article proposes an end-to-end learnt lossy image compression approach, which is built on top of the deep nerual network (DNN)-based variational auto-encoder (VAE) structure with Non-Local ...Attention optimization and Improved Context modeling (NLAIC). Our NLAIC 1) embeds non-local network operations as non-linear transforms in both main and hyper coders for deriving respective latent features and hyperpriors by exploiting both local and global correlations, 2) applies attention mechanism to generate implicit masks that are used to weigh the features for adaptive bit allocation, and 3) implements the improved conditional entropy modeling of latent features using joint 3D convolutional neural network (CNN)-based autoregressive contexts and hyperpriors. Towards the practical application, additional enhancements are also introduced to speed up the computational processing (e.g., parallel 3D CNN-based context prediction), decrease the memory consumption (e.g., sparse non-local processing) and reduce the implementation complexity (e.g., a unified model for variable rates without re-training). The proposed model outperforms existing learnt and conventional (e.g., BPG, JPEG2000, JPEG) image compression methods, on both Kodak and Tecnick datasets with the state-of-the-art compression efficiency, for both PSNR and MS-SSIM quality measurements. We have made all materials publicly accessible at https://njuvision.github.io/NIC for reproducible research.
SUMMARY
Porosity characterization is of profound significance for seismic inversion and hydrocarbon prediction. Although semi-supervised learning (SSL) based methods have been used to boost ...prediction accuracy and lateral continuity of supervised learning (SL) inverted subsurface properties, their variations are relatively limited since the relationships between the data and the parameter model are straightforward in most reported cases. To further figure out their essential differences, we proposed the SSL-based network (SSLBN) for reservoir porosity prediction using seismic and well log data with disparate complexity and quality, and compared it with the SL-based network (SLBN). The SSLBN comprises a data-driven inverse model named decoder and a data-driven forward model named encoder based on the bidirectional-gated recurrent units. The architecture of the SLBN is the same as the encoder. Trained by several seismic-to-well pairs and numerous unlabelled seismic logs, the SSLBN learns the physical process from input single-trace observed seismic log to the intermediate porosity log, and the inverted porosity to the output generated seismic log. We first prepare the porosity model with biased or unbiased labels, the convolution model (CM) and reverse time migration (RTM) based synthetic seismic data, and then implement SL- and SSL-based statistical tests. The synthetic data examples demonstrate that the SSLBN has significant preponderance over the SLBN in the scenes of the RTM imaged seismic data and biased porosity labels. Compared with the SLBN, the physical regularization of the data misfit in the SSLBN improves estimation accuracy and reduces prediction uncertainty of porosity. Finally, statistical tests on a braided river deposited field data example illustrate that the SSLBN can generate more geologically trustworthy porosity models and indicate the oil layers of high porosity sandstone reservoirs.
With technological development of multi sensors, UAV (unmanned aerial vehicle) can identify and locate key targets in essential monitoring areas or geological disaster-prone areas by taking video ...sequence images, and precise positioning of the video sequence images is constantly a matter of great concern. In recent years, precise positioning of aerial images has been widely studied. But it is still a challenge to simultaneously realize precise, robust and dynamic positioning of UAV's patrolling video sequence images in real time. In order to solve this problem, a visual positioning model for patrolling video sequence images based on DOM rectification is proposed, including a robust block-matching algorithm and a precise polynomial-rectifying algorithm. First, the robust block-matching algorithm is used to obtain the best matching area for UAV's video sequence image on DOM (Digital Orthophoto Map), a pre-acquired digital orthophoto map covering the whole UAV's patrolling region. Second, the precise polynomial-rectifying algorithm is used to calculate accurate rectification parameters of mapping UAV's video sequence image to the best matching area obtained above, and then real time positioning of UAV's patrolling video sequence images can be realized. Finally, the above two algorithms are analyzed and verified by three practical experiments, and results indicate that even if spatial resolution, surface specific features, illumination condition and topographic relief are significantly different between DOM and UAV's patrolling video sequence images, proposed algorithms can still steadily realize positioning of each UAV's patrolling video sequence image with about 2.5 m level accuracy in 1 s. To some extent, this study has improved precise positioning effects of UAV's patrolling video sequence images in real time, and the proposed mathematical model can be directly incorporated into UAV's patrolling system without any hardware overhead.
Display omitted
•Microtopography controls carbon accumulation and nutrient release.•Soil organic matter content and carbon:nitrogen ratio were negatively correlated with surface elevation.•High ...ammonium concentrations are released from organic rich low-lying soils when flushed with brackish water.•Local geomorphological depressions are hotspots of carbon turnover and nutrient release in coastal peatland.
Coastal peatlands have been frequently blocked from the sea and artificially drained for agriculture. With an increasing awareness of ecosystem functions, several of these coastal peatlands have been rewetted through dike removal, allowing seawater flooding. In this study, we investigated a recently rewetted peatland on the Baltic Sea coast with the aim to characterize the prevailing soils/sediments with respect to organic matter accumulation and the potential release of nutrients upon seawater flooding. Eighty disturbed soil samples were collected from two depths at different elevations (–0.90 to 0.97 m compared to sea level) and analyzed for soil organic matter (SOM) content and carbon:nitrogen (C:N) ratio. Additionally, nine undisturbed soil cores were collected from three distinct elevation groups and used in leaching experiments with alternating freshwater and Baltic Sea water. The results demonstrated a moderate to strong spatial dependence of surface elevation, SOM content, and C:N ratio. SOM content and C:N ratio were strongly negatively correlated with elevation, indicating that organic matter mineralization was restricted in low-lying areas. The results also showed that the soils at low elevations release more dissolved organic carbon (DOC) and ammonium (NH4+) than soils at high elevations. For soils at low elevations, higher DOC concentrations were observed when flushing with freshwater, whereas higher NH4+ concentrations were found when flushing with brackish water. Recorded NH4+ concentrations in organic-rich peat reached 14.82 ± 9.25 mg L–1, exceeding Baltic seawater (e.g., 0.03 mg L–1) by two orders of magnitude. A potential sea level rise may increase the export of NH4+ from low-lying and rewetted peat soils to the sea, impacting adjacent marine ecosystems. Overall, in coastal peatlands, geochemical processes (e.g., C and N cycling and release) are closely linked to microtopography and related patterns of organic matter content of the soil and sediments.
Maize plant detection was conducted in this study with the goals of target fertilization and reduction of fertilization waste in weed spots and gaps between maize plants. The methods used included ...two types of color featuring and deep learning (DL). The four color indices used were excess green (ExG), excess red (ExR), ExG minus ExR, and the hue value from the HSV (hue, saturation, and value) color space, while the DL methods used were YOLOv3 and YOLOv3_tiny. For practical application, this study focused on performance comparison in detection accuracy, robustness to complex field conditions, and detection speed. Detection accuracy was evaluated by the resulting images, which were divided into three categories: true positive, false positive, and false negative. The robustness evaluation was performed by comparing the average intersection over union of each detection method across different sub–datasets—namely original subset, blur processing subset, increased brightness subset, and reduced brightness subset. The detection speed was evaluated by the indicator of frames per second. Results demonstrated that the DL methods outperformed the color index–based methods in detection accuracy and robustness to complex conditions, while they were inferior to color feature–based methods in detection speed. This research shows the application potential of deep learning technology in maize plant detection. Future efforts are needed to improve the detection speed for practical applications.
LiDAR sensors can provide dependable 3D spatial information at a low frequency (around 10 Hz) and have been widely applied in the field of autonomous driving and unmanned aerial vehicle (UAV). ...However, the camera with a higher frequency (around 20 Hz) has to be decreased so as to match with LiDAR in a multi-sensor system. In this paper, we propose a novel Pseudo-LiDAR interpolation network (PLIN) to increase the frequency of LiDAR sensor data. PLIN can generate temporally and spatially high-quality point cloud sequences to match the high frequency of cameras. To achieve this goal, we design a coarse interpolation stage guided by consecutive sparse depth maps and motion relationship. We also propose a refined interpolation stage guided by the realistic scene. Using this coarse-to-fine cascade structure, our method can progressively perceive multi-modal information and generate accurate intermediate point clouds. To the best of our knowledge, this is the first deep framework for Pseudo-LiDAR point cloud interpolation, which shows appealing applications in navigation systems equipped with LiDAR and cameras. Experimental results demonstrate that PLIN achieves promising performance on the KITTI dataset, significantly outperforming the traditional interpolation method and the state-of-the-art video interpolation technique.
This paper deals with the problem of multi-node transfer alignment estimation of airborne distributed position and orientation system (DPOS), and its aim is to achieve the motion parameters for all ...sub-nodes as high precision as possible by using the main system. The complicated aircraft deformation, especially the wing's flexure, will seriously decrease the accuracy of transfer alignment. Usually, the deformation between the main node and each sub-node is idealized as an independent Markov process. In fact, these deformations are interrelated at a certain moment. To solve the mentioned problem, a multi-node transfer alignment method based on mechanics modeling is proposed in this paper. In this method, the elastic mechanics theory is used to build a unified kinematic equation of the wing to describe the flexible deformations of all sub-nodes, and then the transfer alignment is performed to obtain the motion parameters of each sub-node with inertial measurement units (IMU). Finally, the motion parameters of other sub-nodes without IMUs are obtained by the least squares fitting. The mathematical simulation and semi-physical simulation based on flight experiment show that the proposed method is not only effective but also provides us some new insights into the multi-node measurement of DPOS.
Non-syndromic cleft lip with or without cleft palate (NSCL/P) is a common congenital facial malformation with a complex, incompletely understood origin. Long noncoding RNAs (lncRNAs) have emerged as ...pivotal regulators of gene expression, potentially shedding light on NSCL/P's etiology. This study aimed to identify critical lncRNAs and construct regulatory networks to unveil NSCL/P's underlying molecular mechanisms. Integrating gene expression profiles from the Gene Expression Omnibus (GEO) database, we pinpointed 30 dysregulated NSCL/P-associated lncRNAs. Subsequent analyses enabled the creation of competing endogenous RNA (ceRNA) networks, lncRNA-RNA binding protein (RBP) interaction networks, and lncRNA cis and trans regulation networks. RT-qPCR was used to examine the regulatory networks of lncRNA in vivo and in vitro. Furthermore, protein levels of lncRNA target genes were validated in human NSCL/P tissue samples and murine palatal shelves. Consequently, two lncRNAs and three mRNAs: FENDRR (log2FC = - 0.671, P = 0.040), TPT1-AS1 (log2FC = 0.854, P = 0.003), EIF3H (log2FC = - 1.081, P = 0.041), RBBP6 (log2FC = 0.914, P = 0.037), and SRSF1 (log2FC = 0.763, P = 0.026) emerged as potential contributors to NSCL/P pathogenesis. Functional enrichment analyses illuminated the biological functions and pathways associated with these lncRNA-related networks in NSCL/P. In summary, this study comprehensively delineates the dysregulated transcriptional landscape, identifies associated lncRNAs, and reveals pivotal sub-networks relevant to NSCL/P development, aiding our understanding of its molecular progression and setting the stage for further exploration of lncRNA and mRNA regulation in NSCL/P.
...greater saturated hydraulic conductivity values (Ks) are observed in pristine peat than in degraded peat (Figure 2B). Macropores in low to moderately degraded peat soils (e.g., bulk density <0.2 g ...cm−3) are formed by the undecomposed parent plant material, which functions as a channel/pipe system (Figure 1). Pore structure (total porosity and macroporosity; A) and saturated hydraulic conductivity (Ks; B) of peat soils along a bulk density gradient. The macro-porosity of peat soil differs from those of mineral soils because macro-pores in mineral soils belong to the secondary pore space originating from biological activity (worm burrows, plant roots) and formation of soil peds (aggregation).
The research of robotic autonomous radioactivity detection or radioactive source search plays an important role in the monitoring and disposal of nuclear safety and biological safety. In this paper, ...a method for autonomously searching for radioactive sources through mobile robots was proposed. In the method, by using a partially observable Markov decision process (POMDP), the search of autonomous unknown radioactive sources was realized according to a series of radiation information measured by mobile robot. First, the factors affecting the accuracy of radiation measurement during the robot’s movement were analyzed. Based on these factors, the behavior set of POMDP was designed. Secondly, the parameters of the radioactive source were estimated in the Bayesian framework. In addition, through the reward strategy, autonomous navigation of the robot to the position of the radiation source was achieved. The search algorithm was simulated and tested, and the TurtleBot robot platform was used to conduct a real search experiment on the radio source Cs-137 with an activity of 37 MBq indoors. The experimental results showed the effectiveness of the method. Additionally, from the experiments, it could been seen that the robot was affected by the linear velocity, angular velocity, positioning accuracy and the number of measurements in the process of autonomous search for the radioactive source. The proposed mobile robot autonomous search method can be applied to the search for lost radioactive sources, as well as for the leakage of substances (nuclear or chemical) in nuclear power plants and chemical plants.