An automated pipeline for supervised classification of petal color from citizen science photographs
Ontology highlight
ABSTRACT: Premise
Petal color is an ecologically important trait, and uncovering color variation over a geographic range, particularly in species with large distributions and/or short bloom times, requires extensive fieldwork. We have developed an alternative method that segments images from citizen science repositories using Python and k‐means clustering in the hue‐saturation‐value (HSV) color space. Methods
Our method uses k‐means clustering to aggregate like‐color pixels in sample images to generate the HSV color space encapsulating the color range of petals. Using the HSV values, our method isolates photographs containing clusters in that range and bins them into a classification scheme based on user‐defined categories. Results
We demonstrate the application of this method using two species: one with a continuous range of variation of pink‐purple petals in Geranium maculatum, and one with a binary classification of white versus blue in Linanthus parryae. We demonstrate results that are repeatable and accurate. Discussion
This method provides a flexible, robust, and easily adjustable approach for the classification of color images from citizen science repositories. By using color to classify images, this pipeline sidesteps many of the issues encountered using more traditional computer vision applications. This approach provides a tool for making use of large citizen scientist data sets.
SUBMITTER: Perez‐Udell R
PROVIDER: S-EPMC9934523 | biostudies-literature | 2023 Jan
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Color research continuously demands better methods and larger sample sizes. Citizen science (CS) projects are producing an ever-growing geo- and time-referenced set of photographs of organisms. These datasets have the potential to make a huge contribution to color research, but the reliability of these data need to be tested before widespread implementation.We compared the difference between color extracted from CS photographs with that of color extracted from controlled lighting conditions (i.e., the current gold standard in spectrometry) for both birds and plants. First, we tested the ability of CS photographs to quantify interspecific variability by assessing > 9,000 CS photographs of 537 Australian bird species with controlled museum spectrometry data. Second, we tested the ability of CS photographs to quantify intraspecific variability by measuring petal color data for two plant species using seven methods/sources with varying levels of control.For interspecific questions, we found that by averaging out variability through a large sample size, CS photographs capture a large proportion of across species variation in plumage color within the visual part of the spectrum (R2 = 0.68-0.71 for RGB space and 0.72-0.77 for CIE-LAB space). Between 12 and 14 photographs per species are necessary to achieve this averaging effect for interspecific studies. Unsurprisingly, the CS photographs taken with commercial cameras failed to capture information in the UV part of the spectrum. For intraspecific questions, decreasing levels of control increase the color variation but averaging larger sample sizes can partially mitigate this, aside from particular issues related to saturation and irregularities in light capture.CS photographs offer a very large sample size across space and time which offers statistical power for many color research questions. This study shows that CS photographs contain data that lines up closely with controlled measurements within the visual spectrum if the sample size is large enough, highlighting the potential of CS photographs for both interspecific and intraspecific ecological or biological questions. With regard to analyzing color in CS photographs, we suggest, as a starting point, to measure multiple random points within the ROI of each photograph for both patterned and unpatterned patches and approach the recommended sample size of 12-14 photographs per species for interspecific studies. Overall, this study provides groundwork in analyzing the reliability of a novel method, which can propel the field of studying color forward.
| S-EPMC8093748 | biostudies-literature
Project description:Plant functional traits ('traits') are essential for assessing biodiversity and ecosystem processes, but cumbersome to measure. To facilitate trait measurements, we test if traits can be predicted through visible morphological features by coupling heterogeneous photographs from citizen science (iNaturalist) with trait observations (TRY database) through Convolutional Neural Networks (CNN). Our results show that image features suffice to predict several traits representing the main axes of plant functioning. The accuracy is enhanced when using CNN ensembles and incorporating prior knowledge on trait plasticity and climate. Our results suggest that these models generalise across growth forms, taxa and biomes around the globe. We highlight the applicability of this approach by producing global trait maps that reflect known macroecological patterns. These findings demonstrate the potential of Big Data derived from professional and citizen science in concert with CNN as powerful tools for an efficient and automated assessment of Earth's plant functional diversity.
| S-EPMC8361087 | biostudies-literature
Project description:Citizen science and automated collection methods increasingly depend on image recognition to provide the amounts of observational data research and management needs. Recognition models, meanwhile, also require large amounts of data from these sources, creating a feedback loop between the methods and tools. Species that are harder to recognize, both for humans and machine learning algorithms, are likely to be under-reported, and thus be less prevalent in the training data. As a result, the feedback loop may hamper training mostly for species that already pose the greatest challenge. In this study, we trained recognition models for various taxa, and found evidence for a 'recognizability bias', where species that are more readily identified by humans and recognition models alike are more prevalent in the available image data. This pattern is present across multiple taxa, and does not appear to relate to differences in picture quality, biological traits or data collection metrics other than recognizability. This has implications for the expected performance of future models trained with more data, including such challenging species.
| S-EPMC9890120 | biostudies-literature
Project description:PurposeIn assessing the severity of age-related macular degeneration (AMD), the Age-Related Eye Disease Study (AREDS) Simplified Severity Scale predicts the risk of progression to late AMD. However, its manual use requires the time-consuming participation of expert practitioners. Although several automated deep learning systems have been developed for classifying color fundus photographs (CFP) of individual eyes by AREDS severity score, none to date has used a patient-based scoring system that uses images from both eyes to assign a severity score.DesignDeepSeeNet, a deep learning model, was developed to classify patients automatically by the AREDS Simplified Severity Scale (score 0-5) using bilateral CFP.ParticipantsDeepSeeNet was trained on 58 402 and tested on 900 images from the longitudinal follow-up of 4549 participants from AREDS. Gold standard labels were obtained using reading center grades.MethodsDeepSeeNet simulates the human grading process by first detecting individual AMD risk factors (drusen size, pigmentary abnormalities) for each eye and then calculating a patient-based AMD severity score using the AREDS Simplified Severity Scale.Main outcome measuresOverall accuracy, specificity, sensitivity, Cohen's kappa, and area under the curve (AUC). The performance of DeepSeeNet was compared with that of retinal specialists.ResultsDeepSeeNet performed better on patient-based classification (accuracy = 0.671; kappa = 0.558) than retinal specialists (accuracy = 0.599; kappa = 0.467) with high AUC in the detection of large drusen (0.94), pigmentary abnormalities (0.93), and late AMD (0.97). DeepSeeNet also outperformed retinal specialists in the detection of large drusen (accuracy 0.742 vs. 0.696; kappa 0.601 vs. 0.517) and pigmentary abnormalities (accuracy 0.890 vs. 0.813; kappa 0.723 vs. 0.535) but showed lower performance in the detection of late AMD (accuracy 0.967 vs. 0.973; kappa 0.663 vs. 0.754).ConclusionsBy simulating the human grading process, DeepSeeNet demonstrated high accuracy with increased transparency in the automated assignment of individual patients to AMD risk categories based on the AREDS Simplified Severity Scale. These results highlight the potential of deep learning to assist and enhance clinical decision-making in patients with AMD, such as early AMD detection and risk prediction for developing late AMD. DeepSeeNet is publicly available on https://github.com/ncbi-nlp/DeepSeeNet.
| S-EPMC6435402 | biostudies-literature
Project description:Citizen science programs using organism photographs have become popular, but there are two problems related to photographs. One problem is the low quality of photographs. It is laborious to identify species in photographs taken outdoors because they are out of focus, partially invisible, or under different lighting conditions. The other is difficulty for non-experts to identify species. Organisms usually have interspecific similarity and intraspecific variation, which hinder species identification by non-experts. Deep learning solves these problems and increases the availability of organism photographs. We trained a deep convolutional neural network, Xception, to identify bee species using various quality of bee photographs that were taken by citizens. These bees belonged to two honey bee species and 10 bumble bee species with interspecific similarity and intraspecific variation. We investigated the accuracy of species identification by biologists and deep learning. The accuracy of species identification by Xception (83.4%) was much higher than that of biologists (53.7%). When we grouped bee photographs by different colors resulting from intraspecific variation in addition to species, the accuracy of species identification by Xception increased to 84.7%. The collaboration with deep learning and experts will increase the reliability of species identification and their use for scientific researches.
| S-EPMC8786926 | biostudies-literature
Project description:Non-targeted analysis is nowadays applied in many different domains of analytical chemistry such as metabolomics, environmental and food analysis. Conventional processing strategies for GC-MS data include baseline correction, feature detection, and retention time alignment before multivariate modeling. These techniques can be prone to errors and therefore time-consuming manual corrections are generally necessary. We introduce here a novel fully automated approach to non-targeted GC-MS data processing. This new approach avoids feature extraction and retention time alignment. Supervised machine learning on decomposed tensors of segmented chromatographic raw data signal is used to rank regions in the chromatograms contributing to differentiation between sample classes. The performance of this novel data analysis approach is demonstrated on three published datasets.
| S-EPMC7587030 | biostudies-literature
Project description:Sharing cooking recipes is a great way to exchange culinary ideas and provide instructions for food preparation. However, categorizing raw recipes found online into appropriate food genres can be challenging due to a lack of adequate labeled data. In this study, we present a dataset named the "Assorted, Archetypal, and Annotated Two Million Extended (3A2M+) Cooking Recipe Dataset" that contains two million culinary recipes labeled in respective categories with extended named entities extracted from recipe descriptions. This collection of data includes various features such as title, NER, directions, and extended NER, as well as nine different labels representing genres including bakery, drinks, non-veg, vegetables, fast food, cereals, meals, sides, and fusions. The proposed pipeline named 3A2M+ extends the size of the Named Entity Recognition (NER) list to address missing named entities like heat, time or process from the recipe directions using two NER extraction tools. 3A2M+ dataset provides a comprehensive solution to the various challenging recipe-related tasks, including classification, named entity recognition, and recipe generation. Furthermore, we have demonstrated traditional machine learning, deep learning and pre-trained language models to classify the recipes into their corresponding genre and achieved an overall accuracy of 98.6%. Our investigation indicates that the title feature played a more significant role in classifying the genre.
| S-EPMC11774350 | biostudies-literature
Project description:PurposeTo assess the utility of deep learning in the detection of geographic atrophy (GA) from color fundus photographs and to explore potential utility in detecting central GA (CGA).DesignA deep learning model was developed to detect the presence of GA in color fundus photographs, and 2 additional models were developed to detect CGA in different scenarios.ParticipantsA total of 59 812 color fundus photographs from longitudinal follow-up of 4582 participants in the Age-Related Eye Disease Study (AREDS) dataset. Gold standard labels were from human expert reading center graders using a standardized protocol.MethodsA deep learning model was trained to use color fundus photographs to predict GA presence from a population of eyes with no AMD to advanced AMD. A second model was trained to predict CGA presence from the same population. A third model was trained to predict CGA presence from the subset of eyes with GA. For training and testing, 5-fold cross-validation was used. For comparison with human clinician performance, model performance was compared with that of 88 retinal specialists.Main outcome measuresArea under the curve (AUC), accuracy, sensitivity, specificity, and precision.ResultsThe deep learning models (GA detection, CGA detection from all eyes, and centrality detection from GA eyes) had AUCs of 0.933-0.976, 0.939-0.976, and 0.827-0.888, respectively. The GA detection model had accuracy, sensitivity, specificity, and precision of 0.965 (95% confidence interval [CI], 0.959-0.971), 0.692 (0.560-0.825), 0.978 (0.970-0.985), and 0.584 (0.491-0.676), respectively, compared with 0.975 (0.971-0.980), 0.588 (0.468-0.707), 0.982 (0.978-0.985), and 0.368 (0.230-0.505) for the retinal specialists. The CGA detection model had values of 0.966 (0.957-0.975), 0.763 (0.641-0.885), 0.971 (0.960-0.982), and 0.394 (0.341-0.448). The centrality detection model had values of 0.762 (0.725-0.799), 0.782 (0.618-0.945), 0.729 (0.543-0.916), and 0.799 (0.710-0.888).ConclusionsA deep learning model demonstrated high accuracy for the automated detection of GA. The AUC was noninferior to that of human retinal specialists. Deep learning approaches may also be applied to the identification of CGA. The code and pretrained models are publicly available at https://github.com/ncbi-nlp/DeepSeeNet.
| S-EPMC6810830 | biostudies-literature
Project description:Videos of animal behavior are used to quantify researcher-defined behaviors of interest to study neural function, gene mutations, and pharmacological therapies. Behaviors of interest are often scored manually, which is time-consuming, limited to few behaviors, and variable across researchers. We created DeepEthogram: software that uses supervised machine learning to convert raw video pixels into an ethogram, the behaviors of interest present in each video frame. DeepEthogram is designed to be general-purpose and applicable across species, behaviors, and video-recording hardware. It uses convolutional neural networks to compute motion, extract features from motion and images, and classify features into behaviors. Behaviors are classified with above 90% accuracy on single frames in videos of mice and flies, matching expert-level human performance. DeepEthogram accurately predicts rare behaviors, requires little training data, and generalizes across subjects. A graphical interface allows beginning-to-end analysis without end-user programming. DeepEthogram's rapid, automatic, and reproducible labeling of researcher-defined behaviors of interest may accelerate and enhance supervised behavior analysis. Code is available at: https://github.com/jbohnslav/deepethogram.
| S-EPMC8455138 | biostudies-literature
Project description:Sonmat, a citizen-science enabled Kimjang kimchi case study on associations between hand and kimchi microbiota
| PRJEB80362 | ENA