scholarly journals Convolutional Neural Networks Predict Fish Abundance from Underlying Coral Reef Texture

2018 ◽  
Author(s):  
Grace Young ◽  
Vassileios Balntas ◽  
Victor Prisacariu
Keyword(s):  
Spatial Scales ◽  
A Priori ◽  
3D Structure ◽  
Three Dimensional ◽  
Structural Complexity ◽  
3D Models ◽  
Fish Abundance ◽  
Ecological Patterns ◽  
Abundance And Diversity ◽  
First Time

Coral reefs are among the most biodiverse ecosystems on Earth in large part owing to their unique three-dimensional (3D) structure, which provides niches for a variety of species. Metrics of structural complexity have been shown to correlate with the abundance and diversity of fish and other marine organisms, but they are imperfect representations of a surface that can oversimplify key structural elements and bias discoveries. Moreover, they require researchers to make relatively uninformed guesses about the features and spatial scales relevant to species of interest. This paper introduces a machine-learning method for automating inferences about fish abundance from reef 3D models. It demonstrates the capacity of a convolutional neural network (ConvNet) to learn ecological patterns that are extremely subtle, if not invisible, to the human eye. It is the first time in the literature that no a priori assumptions are made about the bathymetry–fish relationship.

scholarly journals COMPARISON OF COMMERCIAL STRUCTURE-FROM-MOTION PHOTOGRAMMETY SOFTWARE USED FOR UNDERWATER THREE-DIMENSIONAL MODELING OF CORAL REEF ENVIRONMENTS

2017 ◽  
Vol XLII-2/W3 ◽  
pp. 127-131 ◽  
Author(s):  
J. H. R. Burns ◽  
D. Delparte
Keyword(s):  
Coral Reef ◽  
Structural Characteristics ◽  
Spatial Scales ◽  
3D Structure ◽  
Light Availability ◽  
Three Dimensional ◽  
Structural Complexity ◽  
Physical Structure ◽  
Natural Environments ◽  
Ecological Processes

Structural complexity in ecosystems creates an assortment of microhabitat types and has been shown to support greater diversity and abundance of associated organisms. The 3D structure of an environment also directly affects important ecological parameters such as habitat provisioning and light availability and can therefore strongly influence ecosystem function. Coral reefs are architecturally complex 3D habitats, whose structure is intrinsically linked to the ecosystem biodiversity, productivity, and function. The field of coral ecology has, however, been primarily limited to using 2-dimensional (2D) planar survey techniques for studying the physical structure of reefs. This conventional approach fails to capture or quantify the intricate structural complexity of corals that influences habitat facilitation and biodiversity. A 3-dimensional (3D) approach can obtain accurate measurements of architectural complexity, topography, rugosity, volume, and other structural characteristics that affect biodiversity and abundance of reef organisms. Structurefrom- Motion (SfM) photogrammetry is an emerging computer vision technology that provides a simple and cost-effective method for 3D reconstruction of natural environments. SfM has been used in several studies to investigate the relationship between habitat complexity and ecological processes in coral reef ecosystems. This study compared two commercial SfM software packages, Agisoft Photoscan Pro and Pix4Dmapper Pro 3.1, in order to assess the cpaability and spatial accuracy of these programs for conducting 3D modeling of coral reef habitats at three spatial scales.

scholarly journals Associations between Benthic Cover and Habitat Complexity Metrics Obtained from 3D Reconstruction of Coral Reefs at Different Resolutions

2020 ◽  
Vol 12 (6) ◽  
pp. 1011 ◽  
Author(s):  
Atsuko Fukunaga ◽  
John H. R. Burns ◽  
Kailey H. Pascoe ◽  
Randall K. Kosaki
Keyword(s):  
Fractal Dimension ◽  
Coral Reefs ◽  
Habitat Structure ◽  
Coral Cover ◽  
Three Dimensional ◽  
Structural Complexity ◽  
Coralline Algae ◽  
Crustose Coralline Algae ◽  
Surface Areas ◽  
Abundance And Diversity

Quantifying the three-dimensional (3D) habitat structure of coral reefs is an important aspect of coral reef monitoring, as habitat architecture affects the abundance and diversity of reef organisms. Here, we used photogrammetric techniques to generate 3D reconstructions of coral reefs and examined relationships between benthic cover and various habitat metrics obtained at six different resolutions of raster cells, ranging from 1 to 32 cm. For metrics of 3D structural complexity, fractal dimension, which utilizes information on 3D surface areas obtained at different resolutions, and vector ruggedness measure (VRM) obtained at 1-, 2- or 4-cm resolution correlated well with benthic cover, with a relatively large amount of variability in these metrics being explained by the proportions of corals and crustose coralline algae. Curvature measures were, on the other hand, correlated with branching and mounding coral cover when obtained at 1-cm resolution, but the amount of variability explained by benthic cover was generally very low when obtained at all other resolutions. These results show that either fractal dimension or VRM obtained at 1-, 2- or 4-cm resolution, along with curvature obtained at 1-cm resolution, can effectively capture the 3D habitat structure provided by specific benthic organisms.

scholarly journals Variability of contemporary vegetation around Petuniabukta, central Spitsbergen

2012 ◽  
Vol 33 (4) ◽  
pp. 383-394 ◽  
Author(s):  
Karel Prach ◽  
Jitka Klimešová ◽  
Jiří Košnar ◽  
Olexii Redčenko ◽  
Martin Hais
Keyword(s):  
High Latitude ◽  
Vascular Plants ◽  
Native Species ◽  
Spatial Scales ◽  
A Priori ◽  
Vegetation Pattern ◽  
Svalbard Archipelago ◽  
Species Lists ◽  
Phytosociological Relevés ◽  
First Time

Abstract Vegetation was described in various spatial scales in the area of 37.8 km2 including distinguishing vegetation units, vegetation mapping, recording phytosociological relevés (53), and completing species lists of vascular plants (86), mosses (124) and lichens (40). Phytosociological relevés were elaborated using ordination methods DCA and CCA. The relevés formed clusters corresponding well to a priori assigned vegetation units. Slope and stoniness significantly influenced the vegetation pattern. Despite the high latitude (nearly 80° N), the vegetation is rather rich in species. Non-native species do not expand. The moss Bryum dichotomum is reported for the first time from Svalbard archipelago.

scholarly journals Reconstruction of Three-Dimensional Conformations of Bacterial ClpB from High-Speed Atomic-Force-Microscopy Images

2021 ◽  
Vol 8 ◽  
Author(s):  
Bhaskar Dasgupta ◽  
Osamu Miyashita ◽  
Takayuki Uchihashi ◽  
Florence Tama
Keyword(s):  
Atomic Force Microscopy ◽  
Computational Modeling ◽  
High Speed ◽  
3D Structure ◽  
Three Dimensional ◽  
3D Models ◽  
Computational Method ◽  
Molecular Surface ◽  
Force Microscopy ◽  
Atomic Force

ClpB belongs to the cellular disaggretase machinery involved in rescuing misfolded or aggregated proteins during heat or other cellular shocks. The function of this protein relies on the interconversion between different conformations in its native condition. A recent high-speed-atomic-force-microscopy (HS-AFM) experiment on ClpB from Thermus thermophilus shows four predominant conformational classes, namely, open, closed, spiral, and half-spiral. Analyses of AFM images provide only partial structural information regarding the molecular surface, and thus computational modeling of three-dimensional (3D) structures of these conformations should help interpret dynamical events related to ClpB functions. In this study, we reconstruct 3D models of ClpB from HS-AFM images in different conformational classes. We have applied our recently developed computational method based on a low-resolution representation of 3D structure using a Gaussian mixture model, combined with a Monte-Carlo sampling algorithm to optimize the agreement with target AFM images. After conformational sampling, we obtained models that reflect conformational variety embedded within the AFM images. From these reconstructed 3D models, we described, in terms of relative domain arrangement, the different types of ClpB oligomeric conformations observed by HS-AFM experiments. In particular, we highlighted the slippage of the monomeric components around the seam. This study demonstrates that such details of information, necessary for annotating the different conformational states involved in the ClpB function, can be obtained by combining HS-AFM images, even with limited resolution, and computational modeling.

scholarly journals Reconstruction of 3D Object Shape Using Hybrid Modular Neural Network Architecture Trained on 3D Models from ShapeNetCore Dataset

Sensors ◽  
2019 ◽  
Vol 19 (7) ◽  
pp. 1553 ◽  
Author(s):  
Audrius Kulikajevas ◽  
Rytis Maskeliūnas ◽  
Robertas Damaševičius ◽  
Sanjay Misra
Keyword(s):  
Neural Network ◽  
Network Architecture ◽  
A Priori ◽  
Three Dimensional ◽  
Qualitative Evaluation ◽  
3D Models ◽  
Depth Sensors ◽  
The Neural Network ◽  
Artificial Neural Network Ann ◽  
Commercial Applications

Depth-based reconstruction of three-dimensional (3D) shape of objects is one of core problems in computer vision with a lot of commercial applications. However, the 3D scanning for point cloud-based video streaming is expensive and is generally unattainable to an average user due to required setup of multiple depth sensors. We propose a novel hybrid modular artificial neural network (ANN) architecture, which can reconstruct smooth polygonal meshes from a single depth frame, using a priori knowledge. The architecture of neural network consists of separate nodes for recognition of object type and reconstruction thus allowing for easy retraining and extension for new object types. We performed recognition of nine real-world objects using the neural network trained on the ShapeNetCore model dataset. The results evaluated quantitatively using the Intersection-over-Union (IoU), Completeness, Correctness and Quality metrics, and qualitative evaluation by visual inspection demonstrate the robustness of the proposed architecture with respect to different viewing angles and illumination conditions.

scholarly journals A novel 3D template for mandible and maxilla reconstruction: Rapid prototyping using stereolithography

2015 ◽  
Vol 48 (03) ◽  
pp. 263-273 ◽  
Author(s):  
Samir Kumta ◽  
Monica Kumta ◽  
Leena Jain ◽  
Shrirang Purohit ◽  
Rani Ummul
Keyword(s):  
Rapid Prototyping ◽  
Computer Aided Design ◽  
Treatment Plan ◽  
3D Structure ◽  
Three Dimensional ◽  
Bone Grafts ◽  
3D Models ◽  
Mirror Image ◽  
Anatomical Model ◽  
Patient Counselling

ABSTRACT Introduction: Replication of the exact three-dimensional (3D) structure of the maxilla and mandible is now a priority whilst attempting reconstruction of these bones to attain a complete functional and aesthetic rehabilitation. We hereby present the process of rapid prototyping using stereolithography to produce templates for modelling bone grafts and implants for maxilla/mandible reconstructions, its applications in tumour/trauma, and outcomes for primary and secondary reconstruction. Materials and Methods: Stereolithographic template-assisted reconstruction was used on 11 patients for the reconstruction of the mandible/maxilla primarily following tumour excision and secondarily for the realignment of post-traumatic malunited fractures or deformity corrections. Data obtained from the computed tomography (CT) scans with 1-mm resolution were converted into a computer-aided design (CAD) using the CT Digital Imaging and Communications in Medicine (DICOM) data. Once a CAD model was constructed, it was converted into a stereolithographic format and then processed by the rapid prototyping technology to produce the physical anatomical model using a resin. This resin model replicates the native mandible, which can be thus used off table as a guide for modelling the bone grafts. Discussion: This conversion of two-dimensional (2D) data from CT scan into 3D models is a very precise guide to shaping the bone grafts. Further, this CAD can reconstruct the defective half of the mandible using the mirror image principle, and the normal anatomical model can be created to aid secondary reconstructions. Conclusion: This novel approach allows a precise translation of the treatment plan directly to the surgical field. It is also an important teaching tool for implant moulding and fixation, and helps in patient counselling.

scholarly journals Three-Dimensional Reconstruction of Coronary Arteries and Its Application in Localization of Coronary Artery Segments Corresponding to Myocardial Segments Identified by Transthoracic Echocardiography

2013 ◽  
Vol 2013 ◽  
pp. 1-8 ◽  
Author(s):  
Chunyan Zhong ◽  
Yanli Guo ◽  
Haiyun Huang ◽  
Liwen Tan ◽  
Yi Wu ◽  
...  
Keyword(s):  
Coronary Arteries ◽  
Transthoracic Echocardiography ◽  
3D Structure ◽  
Three Dimensional ◽  
3D Models ◽  
Localization Accuracy ◽  
Significant Difference ◽  
2D Ultrasound ◽  
Ct Data ◽  
Contrast Ct

Objectives.To establish 3D models of coronary arteries (CA) and study their application in localization of CA segments identified by Transthoracic Echocardiography (TTE).Methods.Sectional images of the heart collected from the first CVH dataset and contrast CT data were used to establish 3D models of the CA. Virtual dissection was performed on the 3D models to simulate the conventional sections of TTE. Then, we used 2D ultrasound, speckle tracking imaging (STI), and 2D ultrasound plus 3D CA models to diagnose 170 patients and compare the results to coronary angiography (CAG).Results.3D models of CA distinctly displayed both 3D structure and 2D sections of CA. This simulated TTE imaging in any plane and showed the CA segments that corresponded to 17 myocardial segments identified by TTE. The localization accuracy showed a significant difference between 2D ultrasound and 2D ultrasound plus 3D CA model in the severe stenosis group (P<0.05) and in the mild-to-moderate stenosis group (P<0.05).Conclusions.These innovative modeling techniques help clinicians identify the CA segments that correspond to myocardial segments typically shown in TTE sectional images, thereby increasing the accuracy of the TTE-based diagnosis of CHD.

Lure-assisted visual census: a new method for quantifying fish abundance, behaviour, and predation risk in shallow coastal habitats

2010 ◽  
Vol 61 (12) ◽  
pp. 1349 ◽  
Author(s):  
Claudia Kruschel ◽  
Stewart Schultz
Keyword(s):  
Predation Risk ◽  
Structural Complexity ◽  
Marine Ecology ◽  
Fish Abundance ◽  
Coastal Habitats ◽  
Field Methods ◽  
Bare Sediment ◽  
Faunal Abundance ◽  
Abundance And Diversity ◽  
Ambush Predators

Understanding the causes of variation in faunal abundance and diversity across shallow coastal habitats is a fundamental goal of marine ecology. Field methods for inferring a habitat effect on population density and predation risk are informative only if method biases are equal across habitats and species. We hypothesised that observation of fixed lures has a species by bias interaction if sampled species have different modes of predation, and that these biases are overcome by use of moving lures. We tested this hypothesis by observation of fixed and moving lures within seagrass and bare sediment in the Novigrad Sea, Croatian Adriatic. Both methods showed that ambush predators peaked in seagrass, wait–chasers peaked over bare sediment, and move–chasers were abundant in both. Stationary lures underestimated wait–chase and wait–ambush predators relative to moving lures, whereas moving lures did not underestimate the density of predators. These results indicate that stationary lures can underestimate both fish abundance and predation risk in the presence of waiting predators, and that if waiting predators are more abundant in structured habitat, then stationary lures will underestimate the predation risk within such habitats. Use of moving lures may be preferable for comparing habitats differing in structural complexity and frequency of predation modes.

scholarly journals REALDIST: Real-valued protein distance prediction

2020 ◽  
Author(s):  
Badri Adhikari
Keyword(s):  
Structure Prediction ◽  
3D Structure ◽  
Three Dimensional ◽  
Prediction Method ◽  
3D Models ◽  
Classification Problem ◽  
Intermediate Step ◽  
Protein Distance ◽  
Multi Class Classification ◽  
Distance Prediction

AbstractProtein structure prediction continues to stand as an unsolved problem in bioinformatics and biomedicine. Deep learning algorithms and the availability of metagenomic sequences have led to the development of new approaches to predict inter-residue distances—the key intermediate step. Different from the recently successful methods which frame the problem as a multi-class classification problem, this article introduces a real-valued distance prediction method REALDIST. Using a representative set of 43 thousand protein chains, a variant of deep ResNet is trained to predict real-valued distance maps. The contacts derived from the real-valued distance maps predicted by this method, on the most difficult CASP13 free-modeling protein datasets, demonstrate a long-range top-L precision of 52%, which is 17% higher than the top CASP13 predictor Raptor-X and slightly higher than the more recent trRosetta method. Similar improvements are observed on the CAMEO ‘hard’ and ‘very hard’ datasets. Three-dimensional (3D) structure prediction guided by real-valued distances reveals that for short proteins the mean accuracy of the 3D models is slightly higher than the top human predictor AlphaFold and server predictor Quark in the CASP13 competition.

scholarly journals Identification and “in silico” Structural Analysis of the Glutamine-rich Protein Qrp (YheA) in Staphylococcus Aureus

2019 ◽  
Vol 12 (1) ◽  
pp. 18-29
Author(s):  
Javier Escobar-Perez ◽  
Katterine Ospina-Garcia ◽  
Zayda Lorena Corredor Rozo ◽  
Ricaurte Alejandro Marquez-Ortiz ◽  
Jaime E Castellanos ◽  
...  
Keyword(s):  
Staphylococcus Aureus ◽  
Structural Analysis ◽  
Protein Function ◽  
In Silico ◽  
3D Structure ◽  
Three Dimensional ◽  
3D Models ◽  
Protein Domain ◽  
Dimensional Structure ◽  
Biofilm Growth

Background: YlbF and YmcA are two essential proteins for the formation of biofilm, sporulation, and competence in Bacillus subtilis. In these two proteins, a new protein domain called com_ylbF was recently discovered, but its role and protein function has not yet been established. Objective: In this study, we identified and performed an “in silico” structural analysis of the YheA protein, another com_ylbF-containing protein, in the opportunistic pathogen Staphylococcus aureus. Methods: The search of the yheA gene was performed using BLAST-P and tBLASn algorithms. The three-dimensional (3D) models of YheA, as well as YlbF and YmcA proteins, were built using the I-TASSER and Quark programs. The identification of the native YheA in Staphylococcus aureus was carried out through chromatography using the FPLC system. Results: We found that YheA protein is more widely distributed in Gram-positive bacteria than YlbF and YmcA. Two new and important characteristics for YheA and other com_ylbF-containing proteins were found: a highly conserved 3D structure and the presence of a putative conserved motif located in the central region of the domain, which could be involved in its function. Additionally, we established that Staphylococcus aureus expresses YheA protein in both planktonic growth and biofilm. Finally, we suggest renaming YheA as glutamine-rich protein (Qrp) in S. aureus. Conclusion: The Grp (YheA), YlbF, and YmcA proteins adopt a highly conserved three-dimensional structure, harboring a protein-specific putative motif within the com_ylbF domain, which possibly favors the interaction with their substrates. Finally, Staphylococcus aureus expresses the Grp (YheA) protein in both planktonic and biofilm growth.

Sign in / Sign up

Export Citation Format

Share Document