Field-Deployable Computer Vision Wood Identification of Peruvian Timbers.
Ontology highlight
ABSTRACT:
SUBMITTER: Ravindran P
PROVIDER: S-EPMC8206804 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:The remarkable developments in computer vision and machine learning have changed the methodologies of many scientific disciplines. They have also created a new research field in wood science called computer vision-based wood identification, which is making steady progress towards the goal of building automated wood identification systems to meet the needs of the wood industry and market. Nevertheless, computer vision-based wood identification is still only a small area in wood science and is still unfamiliar to many wood anatomists. To familiarize wood scientists with the artificial intelligence-assisted wood anatomy and engineering methods, we have reviewed the published mainstream studies that used or developed machine learning procedures. This review could help researchers understand computer vision and machine learning techniques for wood identification and choose appropriate techniques or strategies for their study objectives in wood science.
| S-EPMC8082842 | biostudies-literature
Project description:The anatomical structure of wood is complex and contains considerable information about its specific species, physical properties, growth environment, and other factors. While conventional wood anatomy has been established by systematizing the xylem anatomical features, which enables wood identification generally up to genus level, it is difficult to describe all the information comprehensively. This study apply two computer vision approaches to optical micrographs: the scale-invariant feature transform algorithm and connected-component labelling. They extract the shape and pore size information, respectively, statistically from the whole micrographs. Both approaches enable the efficient detection of specific features of 18 species from the family Fagaceae. Although the methods ignore the positional information, which is important for the conventional wood anatomy, the simple information on the shape or size of the elements is enough to describe the species-specificity of wood. In addition, according to the dendrograms calculated from the numerical distances of the features, the closeness of some taxonomic groups is inconsistent with the types of porosity, which is one of the typical classification systems for wood anatomy, but consistent with the evolution based on molecular phylogeny; for example, ring-porous group Cerris and radial-porous group Ilex are nested in the same cluster. We analyse which part of the wood structure gave the taxon-specific information, indicating that the latewood zone of group Cerris is similar to the whole zone of group Ilex. Computer vision approaches provide statistical information that uncovers new aspects of wood anatomy that have been overlooked by conventional visual inspection.
| S-EPMC6690550 | biostudies-literature
Project description:The double-yolked (DY) egg is quite popular in some Asian countries because it is considered as a sign of good luck, however, the double yolk is one of the reasons why these eggs fail to hatch. The usage of automatic methods for identifying DY eggs can increase the efficiency in the poultry industry by decreasing egg loss during incubation or improving sale proceeds. In this study, two methods for DY duck egg identification were developed by using computer vision technology. Transmittance images of DY and single-yolked (SY) duck eggs were acquired by a CCD camera to identify them according to their shape features. The Fisher's linear discriminant (FLD) model equipped with a set of normalized Fourier descriptors (NFDs) extracted from the acquired images and the convolutional neural network (CNN) model using primary preprocessed images were built to recognize duck egg yolk types. The classification accuracies of the FLD model for SY and DY eggs were 100% and 93.2% respectively, while the classification accuracies of the CNN model for SY and DY eggs were 98% and 98.8% respectively. The CNN-based algorithm took about 0.12 s to recognize one sample image, which was slightly faster than the FLD-based (about 0.20 s). Finally, this work compared two classification methods and provided the better method for DY egg identification.
| S-EPMC5739493 | biostudies-literature
Project description:The West African Ebola virus outbreak underlined the importance of delivering mass diagnostic capability outside the clinical or primary care setting in effectively containing public health emergencies caused by infectious disease. Yet, to date, there is no solution for reliably deploying at the point of need the gold standard diagnostic method, real time quantitative reverse transcription polymerase chain reaction (RT-qPCR), in a laboratory infrastructure-free manner. In this proof of principle work, we demonstrate direct performance of RT-qPCR on fresh blood using far-red fluorophores to resolve fluorogenic signal inhibition and controlled, rapid freeze/thawing to achieve viral genome extraction in a single reaction chamber assay. The resulting process is entirely free of manual or automated sample pre-processing, requires no microfluidics or magnetic/mechanical sample handling and thus utilizes low cost consumables. This enables a fast, laboratory infrastructure-free, minimal risk and simple standard operating procedure suited to frontline, field use. Developing this novel approach on recombinant bacteriophage and recombinant human immunodeficiency virus (HIV; Lentivirus), we demonstrate clinical utility in symptomatic EBOV patient screening using live, infectious Filoviruses and surrogate patient samples. Moreover, we evidence assay co-linearity independent of viral particle structure that may enable viral load quantification through pre-calibration, with no loss of specificity across an 8?log-linear maximum dynamic range. The resulting quantitative rapid identification (QuRapID) molecular diagnostic platform, openly accessible for assay development, meets the requirements of resource-limited countries and provides a fast response solution for mass public health screening against emerging biosecurity threats.
| S-EPMC5694917 | biostudies-literature
Project description:BackgroundThe explosively radiating evolution of cichlid fishes of Lake Malawi has yielded an amazing number of haplochromine species estimated as many as 500 to 800 with a surprising degree of diversity not only in color and stripe pattern but also in the shape of jaw and body among them. As these morphological diversities have been a central subject of adaptive speciation and taxonomic classification, such high diversity could serve as a foundation for automation of species identification of cichlids.Methodology/principal findingHere we demonstrate a method for automatic classification of the Lake Malawi cichlids based on computer vision and geometric morphometrics. For this end we developed a pipeline that integrates multiple image processing tools to automatically extract informative features of color and stripe patterns from a large set of photographic images of wild cichlids. The extracted information was evaluated by statistical classifiers Support Vector Machine and Random Forests. Both classifiers performed better when body shape information was added to the feature of color and stripe. Besides the coloration and stripe pattern, body shape variables boosted the accuracy of classification by about 10%. The programs were able to classify 594 live cichlid individuals belonging to 12 different classes (species and sexes) with an average accuracy of 78%, contrasting to a mere 42% success rate by human eyes. The variables that contributed most to the accuracy were body height and the hue of the most frequent color.ConclusionsComputer vision showed a notable performance in extracting information from the color and stripe patterns of Lake Malawi cichlids although the information was not enough for errorless species identification. Our results indicate that there appears an unavoidable difficulty in automatic species identification of cichlid fishes, which may arise from short divergence times and gene flow between closely related species.
| S-EPMC3808401 | biostudies-literature
Project description:Mitigating global infectious disease requires diagnostic tools that are sensitive, specific, and rapidly field deployable. In this study, we demonstrate that the Cas13-based SHERLOCK (specific high-sensitivity enzymatic reporter unlocking) platform can detect Zika virus (ZIKV) and dengue virus (DENV) in patient samples at concentrations as low as 1 copy per microliter. We developed HUDSON (heating unextracted diagnostic samples to obliterate nucleases), a protocol that pairs with SHERLOCK for viral detection directly from bodily fluids, enabling instrument-free DENV detection directly from patient samples in <2 hours. We further demonstrate that SHERLOCK can distinguish the four DENV serotypes, as well as region-specific strains of ZIKV from the 2015-2016 pandemic. Finally, we report the rapid (<1 week) design and testing of instrument-free assays to detect clinically relevant viral single-nucleotide polymorphisms.
| S-EPMC6197056 | biostudies-literature
Project description:Background:Burkholderia pseudomallei causes melioidosis, a serious illness that can be fatal if untreated or misdiagnosed. Culture from clinical specimens remains the gold standard but has low diagnostic sensitivity. Method:In this study, we developed a rapid, sensitive and specific insulated isothermal Polymerase Chain Reaction (iiPCR) targeting bimA gene (Burkholderia Intracellular Motility A; BPSS1492) for the identification of B. pseudomallei. A pair of novel primers: BimA(F) and BimA(R) together with a probe were designed and 121 clinical B. pseudomallei strains obtained from numerous clinical sources and 10 ATCC non-targeted strains were tested with iiPCR and qPCR in parallel. Results:All 121 B. pseudomallei isolates were positive for qPCR while 118 isolates were positive for iiPCR, demonstrating satisfactory agreement (97.71%; 95% CI [93.45-99.53%]; k = 0.87). Sensitivity of the bimA iiPCR/POCKIT assay was 97.52% with the lower detection limit of 14 ng/µL of B. pseudomallei DNA. The developed iiPCR assay did not cross-react with 10 types of non-targeted strains, indicating good specificity. Conclusion:This bimA iiPCR/POCKIT assay will undoubtedly complement other methodologies used in the clinical laboratory for the rapid identification of this pathogen.
| S-EPMC7261116 | biostudies-literature
Project description:Field-deployable detectors of disease biomarkers provide a simple and fast analysis of clinical specimens. However, most of the existing field-deployable diagnostics have poor sensitivity and are not suitable for the detection of biomarkers with low abundance. Herein, we report a highly sensitive and rapid colorimetric readout paper-based assay for pathogen detection by integrating the unique collateral activity of a Cas12a-activated universal field-deployable detector (CUFD). The collateral effect of Cas12a results in a nonspecific destruction of a fluorophore biotin-labeled ssDNA reporter for the CUFD. This technique can quantify seven different kinds of pathogens in blood samples without any purification procedure, with sensitivity as low as 10 aM for the Shigella dysenteriae DNA. This CUFD technique has significant potential for the detection of pathogenic DNA as well as other types of DNA or RNA targets at the point-of-care application.
| S-EPMC7315577 | biostudies-literature
Project description:Single domain antibodies (sdAb) are the recombinant variable heavy domains derived from camelid heavy-chain antibodies. While they have binding affinities equivalent to conventional antibodies, sdAb are only one-tenth the size and possess numerous advantages such as excellent thermal stability with the ability to refold following denaturation, and inexpensive production in Escherichia coli or yeast. However, their small size does have drawbacks, one being that they can lose activity upon attachment or adsorption to surfaces, or may fail to adsorb efficiently, as they are highly soluble. This can make the transition from using conventional antibodies to sdAb nontrivial for assay development. Specifically, it is often necessary to re-optimize the protocols and tailor the recombinant sdAb through protein engineering to function efficiently in handheld assays, which currently are utilized for point of care testing and field applications. This work focuses on optimizing the integration of sdAb into rapid vertical flow assays. To achieve this goal, we engineered sdAb-based constructs and developed general protocols for the attachment of the sdAb to both gold nanoparticles and a support membrane. We achieved a limit of detection of 0.11 µg/mL for toxins staphylococcal enterotoxin B and ricin, both potential biothreat agents. Additionally, we demonstrated the ability to detect the nucleocapsid protein of SARS-CoV-2, a common target of antigen tests for COVID-19.
| S-EPMC9624348 | biostudies-literature
Project description:Understanding the extremely variable, complex shape and venation characters of angiosperm leaves is one of the most challenging problems in botany. Machine learning offers opportunities to analyze large numbers of specimens, to discover novel leaf features of angiosperm clades that may have phylogenetic significance, and to use those characters to classify unknowns. Previous computer vision approaches have primarily focused on leaf identification at the species level. It remains an open question whether learning and classification are possible among major evolutionary groups such as families and orders, which usually contain hundreds to thousands of species each and exhibit many times the foliar variation of individual species. Here, we tested whether a computer vision algorithm could use a database of 7,597 leaf images from 2,001 genera to learn features of botanical families and orders, then classify novel images. The images are of cleared leaves, specimens that are chemically bleached, then stained to reveal venation. Machine learning was used to learn a codebook of visual elements representing leaf shape and venation patterns. The resulting automated system learned to classify images into families and orders with a success rate many times greater than chance. Of direct botanical interest, the responses of diagnostic features can be visualized on leaf images as heat maps, which are likely to prompt recognition and evolutionary interpretation of a wealth of novel morphological characters. With assistance from computer vision, leaves are poised to make numerous new contributions to systematic and paleobotanical studies.
| S-EPMC4812720 | biostudies-literature