An Analysis of the Accuracy of Photo-Based Plant Identification Applications on Fifty-Five Tree Species

2022 ◽  
Vol 48 (1) ◽  
pp. 27-43
Author(s):  
Ryan Schmidt ◽  
Brianna Casario ◽  
Pamela Zipse ◽  
Jason Grabosky
Keyword(s):  
Tree Species ◽  
Species Level ◽  
Educational Tool ◽  
Plant Identification ◽  
High Confidence ◽  
Genus Level ◽  
Science Projects ◽  
Mobile Plant ◽  
Field Identification ◽  
Community Science

Background: With the creation of photo-based plant identification applications (apps), the ability to attain basic identifications of plants in the field is seemingly available to anyone who has access to a smartphone. The use of such apps as an educational tool for students and as a major identification resource for some community science projects calls into question the accuracy of the identifications they provide. We created a study based on the context of local tree species in order to offer an informed response to students asking for guidance when choosing a tool for their support in classes. Methods: This study tested 6 mobile plant identification apps on a set of 440 photographs representing the leaves and bark of 55 tree species common to the state of New Jersey (USA). Results: Of the 6 apps tested, PictureThis was the most accurate, followed by iNaturalist, with PlantSnap failing to offer consistently accurate identifications. Overall, these apps are much more accurate in identifying leaf photos as compared to bark photos, and while these apps offer accurate identifications to the genus-level, there seems to be little accuracy in successfully identifying photos to the species-level. Conclusions: While these apps cannot replace traditional field identification, they can be used with high confidence as a tool to assist inexperienced or unsure arborists, foresters, or ecologists by helping to refine the pool of possible species for further identification.

ERRORS ASSOCIATED WITH HERBAGE DISSECTION ANALYSES 2. Problems of species identification

1976 ◽  
pp. 213-225
Author(s):  
I.M. Ritchie ◽  
C.C. Boswell ◽  
A.M. Badland
Keyword(s):  
New Zealand ◽  
Dry Weight ◽  
Analytical Tool ◽  
Species Level ◽  
Weight Basis ◽  
Plant Identification ◽  
Leaf Fragment ◽  
Research Division ◽  
Or Groups ◽  
Botanical Analysis

HERBACE DISSECTION is the process in which samples of herbage cut from trials are separated by hand into component species. Heavy reliance is placed on herbage dissection as an analytical tool ,in New Zealand, and in the four botanical analysis laboratories in the Research Division of the Ministry of Agriculture and Fisheries about 20 000 samples are analysed each year. In the laboratory a representative subsample is taken by a rigorous quartering procedure until approximately 400 pieces of herbage remain. Each leaf fragment is then identified to species level or groups of these as appropriate. The fractions are then dried and the composition calculated on a percentage dry weight basis. The accuracy of the analyses of these laboratories has been monitored by a system of interchanging herbage dissection samples between them. From this, the need to separate subsampling errors from problems of plant identification was, appreciated and some of this work is described here.

scholarly journals Comparative Immature Morphology of Brazilian Fire Ants (Hymenoptera: Formicidae:Solenopsis)

2012 ◽  
Vol 2012 ◽  
pp. 1-10 ◽  
Author(s):  
Eduardo Gonçalves Paterson Fox ◽  
Daniel Russ Solis ◽  
Mônica Lanzoni Rossi ◽  
Jacques Hubert Charles Delabie ◽  
Rodrigo Fernando de Souza ◽  
...  
Keyword(s):  
Intraspecific Variation ◽  
Solenopsis Invicta ◽  
Morphological Traits ◽  
Species Level ◽  
Fire Ants ◽  
Fire Ant ◽  
Solenopsis Geminata ◽  
Body Dimensions ◽  
Larval Instars ◽  
Genus Level

Although common in Brazil, the biology of the fire antSolenopsis saevissima(Smith) is still poorly studied. Larval descriptions are useful to genus-level ant systematics and sometimes to species-level taxonomy. This study presents a detailed description of juveniles ofS. saevissimafrom Brazil, which were compared with Brazilian specimens ofSolenopsis invictaBuren,Solenopsis geminata(Fabricius), andSolenopsis altipunctataPitts. Different larval instars were separated by diagnostic morphological traits which were confirmed by observing moults. Reproductive larvae could be easily sorted by their distinctive body dimensions and shape. Contrary to previous reports on this species, the larvae ofS. saevissimaproved to be generally identical to those ofS. invicta, while a few specimens resembled those of other close species, such asSolenopsis megergatesTrager. Mature larvae thus presented considerable intraspecific variation in some characters recently proposed to aid fire ant species separation (morphology of head hairs).

Fractal dimension based features are useful descriptors of leaf complexity and shape

1999 ◽  
Vol 29 (9) ◽  
pp. 1301-1310 ◽  
Author(s):  
Wojciech Borkowski
Keyword(s):  
Fractal Dimension ◽  
Tree Species ◽  
Fractal Dimensions ◽  
Plant Identification ◽  
Simple Method ◽  
Mass Dimension ◽  
Scale Range ◽  
Computer Identification

An application of fractal dimensions as measures of leaf complexity to morphometric studies and automated plant identification is presented. Detailed algorithms for the calculation of compass dimension and averaged mass dimension together with a simple method of grasping the scale range related variability are given. An analysis of complexity of more than 300 leaves from 10 tree species is reported. Several classical biometric descriptors as well as 16 fractal dimension features were computed on digitized leaf silhouettes. It is demonstrated that properly defined fractal dimension based features may be used to discriminate between species with more than 90% accuracy, especially when used together with other measures. It seems, therefore, that they can be utilized in computer identification systems and for purely taxonomical purposes.

Reply to “Comment on ‘Which Earthquake Accounts Matter?’ by Susan E. Hough and Stacey S. Martin” by David J. Wald

2021 ◽  
Author(s):  
Susan E. Hough ◽  
Stacey S. Martin
Keyword(s):  
Scientific Community ◽  
Recent Article ◽  
Geological Survey ◽  
Data Sets ◽  
Macroseismic Data ◽  
Education Programs ◽  
Socioeconomic Groups ◽  
Science Projects ◽  
Community Science ◽  
The U.S

Abstract We thank David Wald (Wald, 2021; henceforth, W21) for his interest in our recent article (Hough and Martin, 2021; henceforth, HM21). Although different perspectives are vital in science, we are concerned that W21 misrepresents HM21 as an oblique criticism of the U.S. Geological Survey “Did You Feel It?” (DYFI) system, calling for HM21 to be retracted. Readers who are interested in the issues raised by HM21 and the statements made by us therein are referred to that article. In this brief reply, we respond to specific accusations made by W21 and return to the focus of HM21, calling attention to the extent to which macroseismic data sets and inferences drawn from them can be shaped by a lack of representation among individuals whose observations are available to science. HM21 never questioned the benefits of the community science DYFI project to science. HM21 noted, however, and we reiterate here, that community science also potentially benefits the community. Whether or not it matters for science, if participation in community science projects is unrepresentative across socioeconomic groups, it underscores the need for the scientific community to be proactive in its efforts to reach out to groups that have been underserved by current outreach and education programs. We appreciate this opportunity to continue the important conversation about representation.

scholarly journals Matrix-assisted Laser Desorption Ionization-Time-of-Flight Mass Spectrometry (MALDI-TOF MS) as a Reliable Tool to Identify Species of Catalase-negative Gram-positive Cocci not Belonging to the Streptococcus Genus

2016 ◽  
Vol 10 (1) ◽  
pp. 202-208 ◽  
Author(s):  
Marisa Almuzara ◽  
Claudia Barberis ◽  
Viviana Rojas Velázquez ◽  
Maria Soledad Ramirez ◽  
Angela Famiglietti ◽  
...  
Keyword(s):  
Extraction Methods ◽  
Species Level ◽  
Ionization Time ◽  
Maldi Tof Ms ◽  
Genus Level ◽  
Maldi Tof ◽  
Flight Mass Spectrometry ◽  
Level Identification ◽  
Gram Positive Cocci ◽  
Tof Ms

Objective:To evaluate the performance of matrix-assisted laser desorption ionization-time-of-flight mass spectrometry (MALDI-TOF MS) by using 190 Catalase-negative Gram-Positive Cocci (GPC) clinical isolates.Methods:All isolates were identified by conventional phenotypic tests following the proposed scheme by Ruoff and Christensen and MALDI-TOF MS (Bruker Daltonics, BD, Bremen, Germany). Two different extraction methods (direct transfer formic acid method on spot and ethanol formic acid extraction method) and different cut-offs for genus/specie level identification were used. The score cut-offs recommended by the manufacturer (≥ 2.000 for species-level, 1.700 to 1.999 for genus level and <1.700 no reliable identification) and lower cut-off scores (≥1.500 for genus level, ≥ 1.700 for species-level and score <1.500 no reliable identification) were considered for identification. A minimum difference of 10% between the top and next closest score was required for a different genus or species.MALDI-TOF MS identification was considered correct when the result obtained from MS database agreed with the phenotypic identification result.When both methods gave discordant results, the 16S rDNA orsodAgenes sequencing was considered as the gold standard identification method. The results obtained by MS concordant with genes sequencing, although discordant with conventional phenotyping, were considered correct. MS results discordant with 16S orsodA identification were considered incorrect.Results:Using the score cut-offs recommended by the manufacturer, 97.37% and 81.05% were correctly identified to genus and species level, respectively. On the other hand, using lower cut-off scores for identification, 97.89% and 94.21% isolates were correctly identified to genus and species level respectively by MALDI-TOF MS and no significant differences between the results obtained with two extraction methods were obtained.Conclusion:The results obtained suggest that MALDI-TOF MS has the potential of being an accurate tool for Catalase-negative GPC identification even for those species with difficult diagnosis asHelcococcus,Abiotrophia,Granulicatella, among others. Nevertheless, expansion of the library, especially including more strains with different spectra on the same species might overcome potential “intraspecies” variability problems. Moreover, a decrease of the identification scores for species and genus-level identification must be considered since it may improve the MALDI-TOF MS accuracy.

scholarly journals The Diversity of Epiphytic Wild Orchids in Mallawa Resort Area of Bantimurung Bulusaraung National Park, South Sulawesi, Indonesia

2015 ◽  
Vol 2 (2) ◽  
pp. 28-33
Author(s):  
St. Fatmah Hiola ◽  
Gufran D Dirawan ◽  
Muhammad Wiharto Caronge
Keyword(s):  
Data Collection ◽  
National Park ◽  
Digital Camera ◽  
Species Level ◽  
Orchid Species ◽  
Genus Level ◽  
South Sulawesi

This research aims to report the diversity of epiphytic wild orchids in Mallawa Resort area of Bantimurung Bulusaraung National Park (BBNP), South Sulawesi, Indonesia. Exploration methods were used in this study to search and record epiphyticwild orchids in this area. The technique of data collection comprised taking pictures with a digital camera for documentation and collecting specimens of wild orchids that were unidentified at the site. The identification of orchid species was conducted by matching the morphology and characterization of epiphytic wild orchids with appropriate photographs showing details to enable identification. The results of the study showed that there were 36 species of epiphytic wild orchids to be found in the study area. The identification to species level included 10 species, there were 17 specimens that were identified to genus level, and seven specimens remained unidentified. Sympodial type orchids dominated the suite of native orchids, with 23 species.Keyword: epiphytes, Mallawa Resort, Bantimurung Bulusaraung National Park, wild orchids

scholarly journals Response to spiders of Odisha: a preliminary checklist additions to the spider checklist of Odisha

2019 ◽  
Vol 11 (12) ◽  
pp. 14629-14630
Author(s):  
John T.D. Caleb
Keyword(s):  
The State ◽  
Species Level ◽  
Spider Species ◽  
Genus Level

Choudhury et al. (2018) presented a preliminary list of 248 spider species known to occur from Odisha State based on the compilation of all published literature and fresh collections carried out during 2016-17. This is the most recent paper providing an overall view of the spider diversity known in the state. However, the presented checklist seems to be incomplete, since not all species have been identified up to the species level. Several species, as many as 77 morphospecies, were identified up to the genus level only. Furthermore, it has also missed out on recording several species described from Odisha state itself.

Nomenclatural notes on some taxa of Bidessus Sharp and designation of a neotype (Coleoptera: Dytiscidae: Hydroporinae: Bidessini)

Zootaxa ◽  
2019 ◽  
Vol 4565 (4) ◽  
pp. 579 ◽  
Author(s):  
HANS FERY
Keyword(s):  
Taxonomic Status ◽  
Species Level ◽  
Type Locality ◽  
Modern Material ◽  
Male Specimen ◽  
Genus Level ◽  
Nomina Dubia ◽  
The Stability

The nomenclature of some taxa of the tribe Bidessini is dealt with. Currently, the names of these taxa are mostly interpreted as synonyms of Bidessus unistriatus (Goeze, 1777). The four oldest names for this species are shown to be nomina dubia due to insufficient descriptions and lack of name-bearing types. These names are in their original combinations Dyticus parvulus O.F. Müller, 1776, Dytiscus unistriatus Goeze, 1777, Dytiscus unistriatus Schrank, 1781, and Dyticus monostriatus Geoffroy in Fourcroy, 1785. The uncertainties of these species-level names also affect the identity of the genus-level names Bidessus Sharp, 1882, and Hydroglyphus Motschulsky, 1853a, meaning that the stability of nomenclature in the tribe Bidessini is considerably threatened. To eliminate this threat by clarifying the taxonomic status and the type locality of each of these nominal taxa, one and the same male specimen is designated as neotype for unistriatus Goeze, unistriatus Schrank and monostriatus Geoffroy in Fourcroy, and thus these three names become objective synonyms. The neotype is selected from modern material collected near Paris (France) because the type localities of all three taxa include "Paris environs". This is the locality from which Geoffroy (1762) described the non-binominal "ditique à une seule strie" and to which is referred in the descriptions of the latter three taxa. To prevent the threat of the stability by the name parvulus O.F. Müller it is intended to make an application to the ICZN in order to suppress this name for the purpose of priority. 

scholarly journals An End-to-End Deep Fusion Model for Mapping Forests at Tree Species Levels with High Spatial Resolution Satellite Imagery

2020 ◽  
Vol 12 (20) ◽  
pp. 3324
Author(s):  
Ying Guo ◽  
Zengyuan Li ◽  
Erxue Chen ◽  
Xu Zhang ◽  
Lei Zhao ◽  
...  
Keyword(s):  
Deep Learning ◽  
Tree Species ◽  
High Spatial Resolution ◽  
Kappa Coefficient ◽  
Species Level ◽  
Forest Types ◽  
Remote Sensing Images ◽  
Proposed Model ◽  
Processing Step ◽  
Forest Farm

Mapping the distribution of forest resources at tree species levels is important due to their strong association with many quantitative and qualitative indicators. With the ongoing development of artificial intelligence technologies, the effectiveness of deep-learning classification models for high spatial resolution (HSR) remote sensing images has been proved. However, due to the poor statistical separability and complex scenarios, it is still challenging to realize fully automated and highly accurate forest types at tree species level mapping. To solve the problem, a novel end-to-end deep learning fusion method for HSR remote sensing images was developed by combining the advantageous properties of multi-modality representations and the powerful features of post-processing step to optimize the forest classification performance refined to the dominant tree species level in an automated way. The structure of the proposed model consisted of a two-branch fully convolutional network (dual-FCN8s) and a conditional random field as recurrent neural network (CRFasRNN), which named dual-FCN8s-CRFasRNN in the paper. By constructing a dual-FCN8s network, the dual-FCN8s-CRFasRNN extracted and fused multi-modality features to recover a high-resolution and strong semantic feature representation. By imbedding the CRFasRNN module into the network as post-processing step, the dual-FCN8s-CRFasRNN optimized the classification result in an automatic manner and generated the result with explicit category information. Quantitative evaluations on China’s Gaofen-2 (GF-2) HSR satellite data showed that the dual-FCN8s-CRFasRNN provided a competitive performance with an overall classification accuracy (OA) of 90.10%, a Kappa coefficient of 0.8872 in the Wangyedian forest farm, and an OA of 74.39%, a Kappa coefficient of 0.6973 in the GaoFeng forest farm, respectively. Experiment results also showed that the proposed model got higher OA and Kappa coefficient metrics than other four recently developed deep learning methods and achieved a better trade-off between automaticity and accuracy, which further confirmed the applicability and superiority of the dual-FCN8s-CRFasRNN in forest types at tree species level mapping tasks.

Sign in / Sign up

Export Citation Format

Share Document