Precise Spatial Information Research Articles

Developing a precision spatial information system of smallholder oil palm plantations for sustainable rural development

Oil-palm products are contributes significantly to economic growth, creating jobs and government revenue. Of the 14 million ha of Indonesian oil palm plantation, 40% are smallholder farming characterized by low productivity, irregular distribution, and fragmented. The lack of smallholder farming traceability information is one of the main issues for product certification, which make it difficult to access the international markets and is an obstacle to increasing productivity and developing rural-based processing. The OPTIMAL-IPB technology using Artificial Intelligence (AI) approach provides precise spatial information. The objectives of this research, which were carried out in Jambi Province are to detect precisely: (1) the positions, numbers, and areas of smallholder oil palm plantation in the village with precision; (2) the suitability of the spatial distribution of oil palm plantations compared to the spatial plan (RTRW); (3) potential income of farmers and land tax at village level based on production estimates. The study significantly increased the accuracy and detailed information of palm oil farming data, providing essential information for land use control and the traceability of oil palm products. Village and regency governments can utilize this spatial information system for income tax calculations, preparing various production, and income improvement programs.

Functional Alignment-Auxiliary Generative Adversarial Network-Based Visual Stimuli Reconstruction via Multi-Subject fMRI.

Functional Magnetic Resonance Imaging (fMRI) provides more precise spatial and temporal information to reconstruct stimulus images than other technologies that can be used to measure the human brain's neural responses. The fMRI scans, however, generally show heterogeneity among different subjects. The majority of the existing methods aim primarily at mining correlations between stimuli and evoked brain activity, disregarding the heterogeneity among subjects. Therefore, this heterogeneity will impair the reliability and applicability of multi-subject decoding results, leading to sub-optimal results. The present paper proposes the functional alignment-auxiliary generative adversarial network (FAA-GAN) as a novel multi-subject approach for visual image reconstruction that employs functional alignment to alleviate the heterogeneity between subjects. Our proposed FAA-GAN includes three key components: 1) a generative adversarial network (GAN) module for reconstructing visual stimuli, which consists of a visual image encoder as the generator that uses a nonlinear network to convert stimuli images into an implicit representation and a discriminator that generates the images comparable to the original images in detail; 2) a multi-subject functional alignment module, which is used to precisely align the individual fMRI response space of each subject in a common space to reduce the heterogeneity among different subjects; and 3) a cross-modal hashing retrieval module used for similarity retrieval of two modalities of data, i.e., the visual images and the evoked brain responses. Experiments on real-world datasets show that our FAA-GAN method does better than other state-of-the-art deep learning-based reconstruction methods with fMRI.

Development of an Adaptive Linear Mixture Model for Decomposition of Mixed Pixels to Improve Crop Area Estimation Using Artificial Neural Network

Precise spatial information of crop distribution is vital for government and research organizations to monitor agriculture activities like crop health monitoring, crop yield prediction, and food security. Mapping of crop area is challenging in smallholder farming like India, where crop parcels are smaller than two hectares. With an extension of artificial intelligence, an artificial neural network has ability to learn the spectral feature of multispectral satellite images and map them to a land cover class. However, mixed pixel is a challenging problem in pixel wise classification of coarse resolution satellite images. The linear mixture model is successfully utilized to unmix the signals of a mixed classes. The success of linear mixture model is depending on the selection of endmembers of a mixed class. Therefore, this paper presents an adaptive approach for automatic selection of endmembers of a mixed pixel in linear mixture model using spectral and spatial information. The proposed approach is capable of extracting the fraction area cover of each class by using a constrained least-squares error solution. The GPS field surveys, and drone images are employed to create reference data for the accuracy assessment of proposed algorithm. The experimentation results indicate that the solution of the proposed approach outperformed recent baseline methods in terms of efficiency and accuracy of pixelwise estimated area and overall estimated area of various land cover classes.

Lightweight Traffic Classification Model Based on Deep Learning

The development of mobile computing and the Internet of Things (IoT) has led to a surge in traffic volume, which creates a heavy burden for efficient network management. The network management requires high computational overheads to make traffic classification, which is even worse when in edge networks; existing approaches sacrifice the efficiency to obtain high-precision classification results, which are no longer suitable for limited resources edge network scenario. Given the problem, existing traffic classification generally has huge parameters and especially computational complexity. We propose a lightweight traffic classification model based on the Mobilenetv3 and improve it for an ingenious balance between performance and lightweight. Firstly, we adjust the model scale, width, and resolution to substantially reduce the number of model parameters and computations. Secondly, we embed precise spatial information on the attention mechanism to enhance the traffic flow-level feature extraction capability. Thirdly, we use the lightweight multiscale feature fusion to obtain the multiscale flow-level features of traffic. Experiments show that our model has excellent classification accuracy and operational efficiency. The accuracy of the traffic classification model designed in our work has reached more than 99.82%, and the parameter and computation amount are significantly reduced to 0.26 M and 5.26 M. In addition, the simulation experiments on Raspberry Pi prove the proposed model can realize real-time classification capability in the edge network.

Qualitative imaging of elemental spatial distribution in stalagmites through laser ablation inductively coupled plasma mass spectrometry analysis

The laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) technique is considered versatile for multi-elemental analysis and imaging because it is easy to handle, compatible with different types of solid samples, requires minimal sample preparation, and provides high spatial resolution and sensitivity. One of the challenges of imaging analysis is to obtain accurate and precise spatial information of elements distribution in the sample, so the optimization of the laser ablation (LA) parameters is essential. In this context, this study aimed to optimize the LA parameters for direct analysis of speleothem samples. Laser intensity, frequency, and spot diameter were evaluated through multivariate experimental design and multi-response data, the influence of ablation scan speed and the use of 44Ca as an internal standard (IS) for the qualitative image of 66Zn, 137Ba, 55Mn, 57Fe, 88Sr, 60Ni, and 26Mg distribution in the sample were also evaluated. The multivariate optimization revealed positive interactions between the parameters evaluated, i.e., the greater the laser intensity, LA frequency, and spot diameter, the greater the analyte signal and, thus the sensitivity. Therefore, 90% laser ablation intensity, 20 Hz repetition rate, and 100 µm spot diameter were selected. In the scan speed evaluation, the images obtained with 40 and 20 µm s-1 were very similar for all isotopes. The use of 44Ca as IS did not impact the resolution of the images. The use of 44Ca can provide important information about the speleothem formation and paleoclimate changes.

Deep Visual Proteomics defines single-cell identity and heterogeneity

Despite the availabilty of imaging-based and mass-spectrometry-based methods for spatial proteomics, a key challenge remains connecting images with single-cell-resolution protein abundance measurements. Here, we introduce Deep Visual Proteomics (DVP), which combines artificial-intelligence-driven image analysis of cellular phenotypes with automated single-cell or single-nucleus laser microdissection and ultra-high-sensitivity mass spectrometry. DVP links protein abundance to complex cellular or subcellular phenotypes while preserving spatial context. By individually excising nuclei from cell culture, we classified distinct cell states with proteomic profiles defined by known and uncharacterized proteins. In an archived primary melanoma tissue, DVP identified spatially resolved proteome changes as normal melanocytes transition to fully invasive melanoma, revealing pathways that change in a spatial manner as cancer progresses, such as mRNA splicing dysregulation in metastatic vertical growth that coincides with reduced interferon signaling and antigen presentation. The ability of DVP to retain precise spatial proteomic information in the tissue context has implications for the molecular profiling of clinical samples.

Optimal fluoroscopic viewing angles for stenting of the coronary aorto-ostial lesions.

BackgroundLong-term results after stenting aorto-coronary ostial lesions (AOL) are worse than those achieved in non-ostial locations. AOL interventions still pose a substantial challenge for interventional cardiologists. The aim of the study was to determine the optimal fluoroscopic viewing angles of the left and right coronary ostia, based on multislice computed tomography (MSCT) data.MethodsCardiac MSCT exams of 30 patients with clinical suspicion of coronary artery disease were analyzed. En face angles of both coronary ostia, as well as their optimal projection curves, were determined by 2 independent observers in a standard Dicom viewer, without any additional, specialized software add-ons, using a systematic, step-by-step approach. Spatial relations between the ostial plane and the aorta were also assessed.ResultsThe average en face angle of the left coronary ostium was RAO 23°, CAU 45°; for the right coronary ostium RAO 18°, CRA 5°. The mean inter-observer differences for the en face angles of the left and right coronary arteries were 5° and 7°, respectively.ConclusionsMultislice computed tomography data provide precise spatial information on the orientation of the coronary ostia and their relation to the aortic root. Their utilization for determining the patient-specific viewing angle may substantially facilitate percutaneous coronary interventions in AOL.

Evolution of PET Detectors and Event Positioning Algorithms Using Monolithic Scintillation Crystals

The use of monolithic scintillator-based photon detectors in positron emission tomography (PET) has emerged as an attractive alternative to traditional pixelated array designs. Monolithic-based detector designs employ the scintillation light distribution (LD) shape to provide a single 3-D photon interaction position per event, enabling high spatial resolution throughout the crystal volume. Since there are no intercrystal gaps, monolithic designs provide higher intrinsic detection efficiency compared to pixelated designs. However, in order to make the monolithic detector design practical for clinical PET systems, some major drawbacks need to be addressed, such as the time-consuming and complex calibration procedures to obtain precise spatial and timing information. This article gives a historical review of monolithic-based PET detectors, a description of their main advantages and challenges, describes the state-of-the-art, including their use in current commercial system, and ends with a future prospective.

Spatial heterogeneity of vegetation extent and the response to water level fluctuations and micro-topography in Poyang Lake, China

Wetland vegetation dynamics are vital but difficult to understand due to complex impact factors and their spatial heterogeneities. Poyang Lake, the largest freshwater lake in China, is well-known for its ecological importance as a floodplain wetland. However, heterogeneity in vegetation extent and its response to water regimes across different geomorphic units of Poyang Lake remain unclear to date. To address this gap, Landsat images acquired during dry seasons from 1987 to 2018 were selected to explore vegetation extent and its response to water level fluctuations across multiple elevation bands and four wetland types, namely Riparian Wetland (RW), Delta Wetland (DW), Dish-shaped Wetland (DSW), and Fan-shaped wetland (FSW). Results show that vegetation extent in lower elevation bands of the four wetland types increased remarkably with an average increase rate of 0.53 km2/year between 2003 and 2018 compared to −0.17 km2/year−1 during 1987 to 2002. Hydrological variables indicating high-flow events negatively affected vegetation extent in DSW while variables indicating low-flow events appear to be positive for vegetation extent in RW, DW, and FSW. Over-long floods as happened in 2016 has caused fatal damage to vegetation extent in DSW. In addition, extreme droughts in recent years caused by the Three Gorges Dam promoted significant (p < 0.05) vegetation expansion in lower elevation bands of RW, DW, and FSW, but a deterioration of vegetation extent in higher elevation bands, making Poyang Lake at risk of shrinkage and degradation. Human impacts on the evolution Poyang Lake system have caused wide concern, which also contribute to the proposal of Poyang Lake water control structure. Heterogeneity in vegetation extent and their response to water regimes across Poyang Lake provide precise spatial information for restoration efforts of the wetland ecosystem.

Within-site drivers for soil nutrient variability in plantation forests: A case study from dry sub-humid New Zealand

Precise spatial information on soil properties in plantation forests is needed to improve soil nutrient management and to sustain productivity. Soil nitrogen, phosphorus, potassium, organic matter, carbon and boron are important determinants and indicators of soil fertility and quality. Particularly in forests, these soil properties are highly variable in space and time. In this study, soils were sampled from three plantation forest sites in a dry sub-humid region near Blenheim, New Zealand. Thirty sampling points were selected, and samples were collected from the three sites across a range of slope and aspect strata. Soil samples were analysed for total carbon (totC), total nitrogen (totN), total phosphorus (totP), extractable potassium (exK) and hot-water extractable boron (exB). All examined soil properties varied significantly (p < 0.05) within sites. A set of fine-scale (5 m resolution) topographic surfaces, that might explain this variability, were then interpolated or derived in geographic information system software. Topographic surfaces included elevation, aspect, slope, profile and plan curvature, topographic position index (TPI), topographic wetness index (TWI), wind exposition index (WEI), and morphometric protection index (MPI). A generalised linear mixed-effect model was applied to develop predictive models. The study found all soil properties were positively correlated with MPI and negatively correlated with the WEI. This indicated that soil properties were correlated with shelter from surrounding relief and wind. Interestingly, within-site boron levels were correlated with both profile curvature (PrCurv) and topographic wetness index, indicating boron movement through the surface with the movement of soil moisture. The modelling approach in this study has potential for application to sustainable management of plantation forests using spatially-precise estimates of soil fertility.

Estimating Human Pose Efficiently by Parallel Pyramid Networks.

Good performance and high efficiency are both critical for estimating human pose in practice. Recent state-of-the-art methods have greatly boosted the pose detection accuracy through deep convolutional neural networks, however, the strong performance is typically achieved without high efficiency. In this paper, we design a novel network architecture for human pose estimation, which aims to strike a fine balance between speed and accuracy. Two essential tasks for successful pose estimation, preserving spatial location and extracting semantic information, are handled separately in the proposed architecture. Semantic knowledge of joint type is obtained through deep and wide sub-networks with low-resolution input, and high-resolution features indicating joint location are processed by shallow and narrow sub-networks. Because accurate semantic analysis mainly asks for adequate depth and width of the network and precise spatial information mostly requests preserving high-resolution features, good results can be produced by fusing the outputs of the sub-networks. Moreover, the computational cost can be considerably reduced comparing with existing networks, since the main part of the proposed network only deals with low-resolution features. We refer to the architecture as "parallel pyramid" network (PPNet), as features of different resolutions are processed at different levels of the hierarchical model. The superiority of our network is empirically demonstrated on two benchmark datasets: the MPII Human Pose dataset and the COCO keypoint detection dataset. PPNet outcompetes all recent methods by using less computation and memory to achieve better human pose estimation results.

Improved Static Hand Gesture Classification on Deep Convolutional Neural Networks Using Novel Sterile Training Technique

In this paper, we investigate novel data collection and training techniques towards improving classification accuracy of non-moving (static) hand gestures using a convolutional neural network (CNN) and frequency-modulated-continuous-wave (FMCW) millimeter-wave (mmWave) radars. Recently, non-contact hand pose and static gesture recognition have received considerable attention in many applications ranging from human-computer interaction (HCI), augmented/virtual reality (AR/VR), and even therapeutic range of motion for medical applications. While most current solutions rely on optical or depth cameras, these methods require ideal lighting and temperature conditions. mmWave radar devices have recently emerged as a promising alternative offering low-cost system-on-chip sensors whose output signals contain precise spatial information even in non-ideal imaging conditions. Additionally, deep convolutional neural networks have been employed extensively in image recognition by learning both feature extraction and classification simultaneously. However, little work has been done towards static gesture recognition using mmWave radars and CNNs due to the difficulty involved in extracting meaningful features from the radar return signal, and the results are inferior compared with dynamic gesture classification. This article presents an efficient data collection approach and a novel technique for deep CNN training by introducing “sterile” images which aid in distinguishing distinct features among the static gestures and subsequently improve the classification accuracy. Applying the proposed data collection and training methods yields an increase in classification rate of static hand gestures from 85% to 93% and 90% to 95% for range and range-angle profiles, respectively.

MHA-Net: Multipath Hybrid Attention Network for Building Footprint Extraction From High-Resolution Remote Sensing Imagery

Deep learning approaches have been widely applied to building footprint extraction using high-resolution imagery. However, the traditional fully convolution network still has problems in recovering spatial details and discriminating buildings with varying sizes and styles. We propose a novel multipath hybrid attention network (MHA-Net) to address these challenges. We design a separable convolution block attention module and an attention downsampling module as the basic modules with separable convolutions and channel attention. The MHA-Net architecture consists of three components: the encoding network, multipath hybrid dilated convolution (HDC), and dense upsampling convolution (DUC). The encoding network is used to encode the high-level semantic contexts of images. The multipath HDC aggregates multiscale features by combining rich semantic representations extracted by HDCs, which can achieve promising results in extracting tiny buildings. The DUC is capable of recovering precise spatial information of buildings. We evaluate our network on two public datasets: the WHU aerial building dataset and the Massachusetts building dataset. According to the experimental results, MHA-Net outperforms other classical semantic segmentation models and several recent building extraction models. In particular, MHA-Net can improve the extraction accuracy of small buildings and is robust to complicated building roofs.

Specimens in a Broader Context: The National Ecological Observatory Network and the extended specimen

Community innovations in both specimen digitization (e.g., Morphbank; SlideAtlas; Inselect, Hudson et al. 2015) and data standards (e.g., the National Science Foundation initative "Advancing Digitization of Biodiversity Collections", Page et al. 2015, Nelson and Shari 2019; Darwin Core (Darwin Core Task Group 2009)), have resulted in digitized specimens with rich contextual metadata and the capacity to share such specimen information widely. These extended specimens have allowed for the exploration of cross-scale research questions that traverse multiple taxonomic, spatial and temporal scales. As a relatively new collection organization, the National Ecological Observatory Network (NEON; Keller et al. 2008) has curated and archived &amp;gt;200,000 specimens to date and is projected to archive between 80,000 and 120,000 specimens annually through its 30-year, continental-scale environmental monitoring program. NEON has embraced the Extended Specimen paradigm (introduced by Webster 2017; NEON's implementation described in Lendemer et al. 2020), and each sample is physically and digitally curated from the point of collection enabling sample discoverability that maximizes specimen Findability, Accessibility, Interoperability, and Reusability (the FAIR standard; Wilkinson et al. 2016). All archived specimens are associated with precise spatial and temporal information and (where available/applicable) NEON also integrates specimen images, morphometrics, genetic sequences and taxonomic data with the specimen records within a Symbiota platform. Any additional analyses or derived specimens created by the research community are also linked in the specimen record. NEON has benefited substantially from community development of tools and standards, but the process of data integration has not been without problems. Here, we will discuss challenges NEON has faced in the implementation of the extended specimen as well as solutions.

Culturally modified trees and forest structure at a Kaw\xe9sqar ancient settlement at R\xedo Batchelor, western Patagonia

Indigenous land use occurring on temporal scales over centuries or millennia shapes forests in specific ways and influences the dynamics of forest ecosystems. It is challenging to study such land use, but analysis of “culturally modified trees” (CMTs) can give precise spatial and temporal information on past land use by indigenous people. The aim of this study was to increase our knowledge of indigenous use of land and resources in Nothofagus forests by identifying CMTs and analyzing the forest structure dynamics in an ancient Kawésqar settlement site in western Patagonia. Our results show that there are CMTs at Río Batchelor and that the forest structure varies significantly within the site, indicating that Kawésqar people altered the forest by extracting various resources. We conclude that CMT studies have great potential in Nothofagus forests in southernmost America, but also face specific challenges due to environmental conditions and lack of corroborating historical information.

Quantification of Margalefidinium polykrikoides Blooms along the South Coast of Korea Using Airborne Hyperspectral Imagery

The red tide bloom-forming dinoflagellate Margalefidinium polykrikoides is well known for its harmful effects on marine organisms, and for killing fish in aquaculture cages via gill clogging at a high cell abundance. To minimize the damage caused by red tide blooms, it is essential to understand their detailed spatial distribution with high accuracy. Airborne hyperspectral imagery (HSI) is useful for quantifying red tide cell abundance because it provides substantial information on optical features related to red tide species. However, because published red tide indexes were developed for multichannel ocean color sensors, there are some limitations to applying them directly to HSI. In this study, we propose a new index for quantifying M. polykrikoides blooms along the south coast of Korea and generate a M. polykrikoides cell abundance map using HSI. A new index for estimating cell abundance was proposed using the pairs of M. polykrikoides cell abundances and in situ spectra. After optimization of the published red tide indexes and band correlation analyses, the green-to-fluorescence ratio (GFR) index was proposed based on red tide spectral characteristics. The GFR index was computed from the green (524 and 583 nm) and fluorescence wavelength bands (666 and 698 nm) and converted into red tide cell abundance using a second-order polynomial regression model. The newly proposed GFR index showed the best performance, with a coefficient of determination (R2) of 0.52, root mean squared error (RMSE) of 877.98 cells mL−1, and mean bias error (MBE) of −18.42 cells mL−1, when applied to atmospherically corrected HSI. The M. polykrikoides cell abundance map generated from the GFR index provides precise spatial distribution information and allowed us to estimate a wide range of cell abundance up to 5000 cells mL−1. This study indicates the potential of the GFR index for quantifying M. polykrikoides cell abundance from HSI with a reasonably high level of accuracy.

End-to-end jet classification of quarks and gluons with the CMS Open Data

We describe the construction of novel end-to-end jet image classifiers to discriminate quark- versus gluon-initiated jets using the simulated CMS Open Data. These multi-detector images correspond to true maps of the low-level energy deposits in the detector, giving the classifiers direct access to the maximum recorded event information about the jet, differing fundamentally from conventional jet images constructed from reconstructed particle-level information. Using this approach, we achieve classification performance competitive with current state-of-the-art jet classifiers that are dominated by particle-based algorithms. We find the performance to be driven by the availability of precise spatial information, highlighting the importance of high-fidelity detector images. We then illustrate how end-to-end jet classification techniques can be incorporated into event classification workflows using Quantum Chromodynamics di-quark versus di-gluon events. We conclude with the end-to-end event classification of full detector images, which we find to be robust against the effects of underlying event and pileup outside the jet regions-of-interest.

Dangerous or Desirable: Utilizing Augmented Content for Field Policing

ABSTRACT Location-based systems (LBS) can be of benefit to obtain information when and where it is needed. What is not captured with LBS are points of interest (dynamic hotspots) that are not fixed in one location, such as people. Furthermore, using 2D interface elements makes it difficult to communicate precise spatial information to users. The current study aims to assess the viability of dynamic hotspots and augmented content in a mobile hotspot policing application. Mock cases were constructed to mimic real field policing work. The system, comprised of a smartphone and smartwatch, gave static and dynamic hotspot information, which was enriched with the use of augmented elements viewed through the smartphone-camera. Interviews, observations and usability measures were used to gain insight for augmented reality (AR) design considerations in the domain of field-policing. The detection of dynamic hotspots was seen as desirable. While static and dynamic hotspot information was seen as valuable, AR information, as it was utilized in the current study, has too many drawbacks to be used in the field. Alternative modalities, insights into possible future iterations and technical limitations of AR for the emergency response domain are discussed.

Global Spatio-Temporal Attention for Action Recognition Based on 3D Human Skeleton Data

The human skeleton joints captured by RGB-D camera are widely used in action recognition for its robust and comprehensive 3D information. Presently, most action recognition methods based on skeleton joints treat all skeletal joints with the same importance spatially and temporally. However, the contributions of skeletal joints vary significantly. Hence, a GL-LSTM+Diff model is proposed to improve the recognition of human actions. A global spatial attention (GSA) model is proposed to express the different weights for different skeletal joints to provide precise spatial information for human action recognition. The accumulative learning curve (ALC) model is introduced to highlight which frames contribute most to the final decision making by giving varying temporal weights to each intermediate accumulated learning results. By integrating the proposed GSA (for spatial information) and ALC (for temporal processing) models into the LSTM framework and taking the human skeletal joints as inputs, a global spatio-temporal action recognition framework (GL-LSTM) is constructed to recognize human actions. Diff is introduced as the preprocessing method to enhance the dynamic of the features, thus to get distinguishable features in deep learning. Rigorous experiments on the largest dataset NTU RGB+D and the common small dataset SBU show that the algorithm proposed in this paper outperforms other state-of-the-art methods.

Drawings of real-world scenes during free recall reveal detailed object and spatial information in memory

Understanding the content of memory is essential to teasing apart its underlying mechanisms. While recognition tests have commonly been used to probe memory, it is difficult to establish what specific content is driving performance. Here, we instead focus on free recall of real-world scenes, and quantify the content of memory using a drawing task. Participants studied 30 scenes and, after a distractor task, drew as many images in as much detail as possible from memory. The resulting memory-based drawings were scored by thousands of online observers, revealing numerous objects, few memory intrusions, and precise spatial information. Further, we find that visual saliency and meaning maps can explain aspects of memory performance and observe no relationship between recall and recognition for individual images. Our findings show that not only is it possible to quantify the content of memory during free recall, but those memories contain detailed representations of our visual experiences.