Dataset Information

Standardizing the protocol for hemispherical photographs: accuracy assessment of binarization algorithms.

ABSTRACT: Hemispherical photography is a well-established method to optically assess ecological parameters related to plant canopies; e.g. ground-level light regimes and the distribution of foliage within the crown space. Interpreting hemispherical photographs involves classifying pixels as either sky or vegetation. A wide range of automatic thresholding or binarization algorithms exists to classify the photographs. The variety in methodology hampers ability to compare results across studies. To identify an optimal threshold selection method, this study assessed the accuracy of seven binarization methods implemented in software currently available for the processing of hemispherical photographs. Therefore, binarizations obtained by the algorithms were compared to reference data generated through a manual binarization of a stratified random selection of pixels. This approach was adopted from the accuracy assessment of map classifications known from remote sensing studies. Percentage correct (Pc) and kappa-statistics (K) were calculated. The accuracy of the algorithms was assessed for photographs taken with automatic exposure settings (auto-exposure) and photographs taken with settings which avoid overexposure (histogram-exposure). In addition, gap fraction values derived from hemispherical photographs were compared with estimates derived from the manually classified reference pixels. All tested algorithms were shown to be sensitive to overexposure. Three of the algorithms showed an accuracy which was high enough to be recommended for the processing of histogram-exposed hemispherical photographs: "Minimum" (Pc 98.8%; K 0.952), "Edge Detection" (Pc 98.1%; K 0.950), and "Minimum Histogram" (Pc 98.1%; K 0.947). The Minimum algorithm overestimated gap fraction least of all (11%). The overestimation by the algorithms Edge Detection (63%) and Minimum Histogram (67%) were considerably larger. For the remaining four evaluated algorithms (IsoData, Maximum Entropy, MinError, and Otsu) an incompatibility with photographs containing overexposed pixels was detected. When applied to histogram-exposed photographs, these algorithms overestimated the gap fraction by at least 180%.

SUBMITTER: Glatthorn J

PROVIDER: S-EPMC4242535 | biostudies-literature | 2014

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Standardizing the protocol for hemispherical photographs: accuracy assessment of binarization algorithms.

Glatthorn Jonas J Beckschäfer Philip P

PloS one 20141124 11

Hemispherical photography is a well-established method to optically assess ecological parameters related to plant canopies; e.g. ground-level light regimes and the distribution of foliage within the crown space. Interpreting hemispherical photographs involves classifying pixels as either sky or vegetation. A wide range of automatic thresholding or binarization algorithms exists to classify the photographs. The variety in methodology hampers ability to compare results across studies. To identify ...[more]

PMID: 25420014

Similar Datasets

Project description:Hemispherical photography (HP), implemented with cameras equipped with "fisheye" lenses, is a widely used method for describing forest canopies and light regimes. A promising technological advance is the availability of low-cost fisheye lenses for smartphone cameras. However, smartphone camera sensors cannot record a full hemisphere. We investigate whether smartphone HP is a cheaper and faster but still adequate operational alternative to traditional cameras for describing forest canopies and light regimes. We collected hemispherical pictures with both smartphone and traditional cameras in 223 forest sample points, across different overstory species and canopy densities. The smartphone image acquisition followed a faster and simpler protocol than that for the traditional camera. We automatically thresholded all images. We processed the traditional camera images for Canopy Openness (CO) and Site Factor estimation. For smartphone images, we took two pictures with different orientations per point and used two processing protocols: (i) we estimated and averaged total canopy gap from the two single pictures, and (ii) merging the two pictures together, we formed images closer to full hemispheres and estimated from them CO and Site Factors. We compared the same parameters obtained from different cameras and estimated generalized linear mixed models (GLMMs) between them. Total canopy gap estimated from the first processing protocol for smartphone pictures was on average significantly higher than CO estimated from traditional camera images, although with a consistent bias. Canopy Openness and Site Factors estimated from merged smartphone pictures of the second processing protocol were on average significantly higher than those from traditional cameras images, although with relatively little absolute differences and scatter. Smartphone HP is an acceptable alternative to HP using traditional cameras, providing similar results with a faster and cheaper methodology. Smartphone outputs can be directly used as they are for ecological studies, or converted with specific models for a better comparison to traditional cameras.

Project description:BackgroundHemispherical photography (HP) is a popular method to estimate canopy structure and understorey light environment, which analyses photographs acquired with wide view-angle lens (i.e. fisheye lens). To increase HP accuracy, the approaches of most previous studies were to increase the preciseness of exposure and thresholding of photographs, while ours quantified effects of canopy properties (gap fraction and length of sky-canopy border (SCB)) and errors of exposure and thresholding on the accuracy of HP.ResultsThrough analysing photographs of real and model canopies, it was showed that HP inaccuracy resulted from the mismatch between exposure and thresholding rather than exposure or thresholding errors alone. HP inaccuracy was a function of the SCB length and the extent of exposure and thresholding errors, but independent of gap fraction.DiscussionIn photographs, SCBs are recorded as grey pixels which greyness is in between that of sky and canopy pixels. When there are exposure and thresholding errors, grey pixels are those prone to be misclassified in image analysis. Longer (vegetation with taller canopies) and wider (lower image sharpness) SCBs in photographs can both result in a higher amount of grey pixels and ultimately higher HP inaccuracy for a given extent of exposure and threshold errors.ConclusionsUsing lenses with view angle narrower rather than that of fisheye lens can shorten the SCB length in photographs and in turn reduce HP estimation inaccuracy for canopy structure and understorey light environment. Since short SCBs and low levels of exposure and thresholding errors can both result in low HP inaccuracy, to identify the true performance of new exposure and thresholding methods for HP, photographs recording canopies with long SCBs and acquired with fisheye lenses should be used. Because HP inaccuracy in a function of the amount of grey pixels resulting from SCBs, the amount of these pixels in photographs can be used as a universal parameter to quantify canopy properties influential to HP estimation and in turn make cross-study comparisons feasible.

Project description:PurposeRecent investigations failed to reproduce the positive rotor-guided ablation outcomes shown by initial studies for treating persistent atrial fibrillation (persAF). Phase singularity (PS) is an important feature for AF driver detection, but algorithms for automated PS identification differ. We aim to investigate the performance of four different techniques for automated PS detection.Methods2048-channel virtual electrogram (VEGM) and electrocardiogram signals were collected for 30 s from 10 patients undergoing persAF ablation. QRST-subtraction was performed and VEGMs were processed using sinusoidal wavelet reconstruction. The phase was obtained using Hilbert transform. PSs were detected using four algorithms: (1) 2D image processing based and neighbor-indexing algorithm; (2) 3D neighbor-indexing algorithm; (3) 2D kernel convolutional algorithm estimating topological charge; (4) topological charge estimation on 3D mesh. PS annotations were compared using the structural similarity index (SSIM) and Pearson's correlation coefficient (CORR). Optimized parameters to improve detection accuracy were found for all four algorithms using F β score and 10-fold cross-validation compared with manual annotation. Local clustering with density-based spatial clustering of applications with noise (DBSCAN) was proposed to improve algorithms 3 and 4.ResultsThe PS density maps created by each algorithm with default parameters were poorly correlated. Phase gradient threshold and search radius (or kernels) were shown to affect PS detections. The processing times for the algorithms were significantly different (p < 0.0001). The F β scores for algorithms 1, 2, 3, 3 + DBSCAN, 4 and 4 + DBSCAN were 0.547, 0.645, 0.742, 0.828, 0.656, and 0.831. Algorithm 4 + DBSCAN achieved the best classification performance with acceptable processing time (2.0 ± 0.3 s).ConclusionAF driver identification is dependent on the PS detection algorithms and their parameters, which could explain some of the inconsistencies in rotor-guided ablation outcomes in different studies. For 3D triangulated meshes, algorithm 4 + DBSCAN with optimal parameters was the best solution for real-time, automated PS detection due to accuracy and speed. Similarly, algorithm 3 + DBSCAN with optimal parameters is preferred for uniform 2D meshes. Such algorithms - and parameters - should be preferred in future clinical studies for identifying AF drivers and minimizing methodological heterogeneities. This would facilitate comparisons in rotor-guided ablation outcomes in future works.

Project description:Importance:Advances in smartphone photography (both quality and image transmission) may improve access to care via direct parent-to-clinician telemedicine. However, the accuracy of diagnoses that are reliant on parent-provided photographs has not been formally compared with diagnoses made in person. Objective:To assess whether smartphone photographs of pediatric skin conditions taken by parents are of sufficient quality to permit accurate diagnosis. Design, Setting, and Participants:A prospective study was conducted among 40 patient-parent dyads at a pediatric dermatology clinic at the Children's Hospital of Philadelphia from March 1 to September 30, 2016, to assess concordance between diagnoses made by an independent pediatric dermatologist based on in-person examination and those based on parental photographs. Half of the patient-parent dyads were randomized for a secondary analysis to receive instructions on how best to take photographs with smartphones. Clinicians were blinded to whether parents had received photography instructions. Exposures:Half of the patient-parent dyads received a simple, 3-step instruction sheet on how best to take photographs using a smartphone (intervention group); the other half did not (control group). Main Outcomes and Measures:Concordance between photograph-based vs in-person diagnosis in the intervention vs control groups, as quantified using Cohen ?, a measure of interrater agreement that takes into account the possibility of agreement occurring by chance. Results:Among the 40 patient-parent dyads (22 female children and 18 male children; mean [SD] age, 6.96 [5.23] years), overall concordance between photograph-based vs in-person diagnosis was 83% (95% CI, 71%-94%; ??=?0.81). Diagnostic concordance was 89% (95% CI, 75%-97%; ??=?0.88) in a subgroup of 37 participants with photographs considered of high enough quality to make a diagnosis. No statistically significant effect of photography instructions on concordance was detected (group that received instructions, 85%; group that did not receive instructions, 80%; P?=?.68). In cases of diagnostic disagreement, appropriate follow-up was suggested. Conclusions and Relevance:Parent-operated smartphone photography can accurately be used as a method to provide pediatric dermatologic care. Trial Registration:clinicaltrials.gov Identifier: NCT03246945.

Project description:ObjectiveTo develop a deep learning algorithm (DLA) to detect diabetic kideny disease (DKD) from retinal photographs of patients with diabetes, and evaluate performance in multiethnic populations.Materials and methodsWe trained 3 models: (1) image-only; (2) risk factor (RF)-only multivariable logistic regression (LR) model adjusted for age, sex, ethnicity, diabetes duration, HbA1c, systolic blood pressure; (3) hybrid multivariable LR model combining RF data and standardized z-scores from image-only model. Data from Singapore Integrated Diabetic Retinopathy Program (SiDRP) were used to develop (6066 participants with diabetes, primary-care-based) and internally validate (5-fold cross-validation) the models. External testing on 2 independent datasets: (1) Singapore Epidemiology of Eye Diseases (SEED) study (1885 participants with diabetes, population-based); (2) Singapore Macroangiopathy and Microvascular Reactivity in Type 2 Diabetes (SMART2D) (439 participants with diabetes, cross-sectional) in Singapore. Supplementary external testing on 2 Caucasian cohorts: (3) Australian Eye and Heart Study (AHES) (460 participants with diabetes, cross-sectional) and (4) Northern Ireland Cohort for the Longitudinal Study of Ageing (NICOLA) (265 participants with diabetes, cross-sectional).ResultsIn SiDRP validation, area under the curve (AUC) was 0.826(95% CI 0.818-0.833) for image-only, 0.847(0.840-0.854) for RF-only, and 0.866(0.859-0.872) for hybrid. Estimates with SEED were 0.764(0.743-0.785) for image-only, 0.802(0.783-0.822) for RF-only, and 0.828(0.810-0.846) for hybrid. In SMART2D, AUC was 0.726(0.686-0.765) for image-only, 0.701(0.660-0.741) in RF-only, 0.761(0.724-0.797) for hybrid.Discussion and conclusionThere is potential for DLA using retinal images as a screening adjunct for DKD among individuals with diabetes. This can value-add to existing DLA systems which diagnose diabetic retinopathy from retinal images, facilitating primary screening for DKD.

Project description:Abstract Objectives Photo-based dietary assessment methods are becoming more feasible as artificial intelligence methods improve. However, advancement of these methods to the level usable in nutrition studies has been hindered by the lack of a dataset against which to benchmark algorithm performance. Here, we introduce the Surveying Nutrient Assessment with Photographs of Meals (SNAPMe) Study (ClinicalTrials ID: NCT05008653), describe the data, and discuss the utility of the data. Methods The purpose of the SNAPMe Study was to pair meal photographs with traditional food records. The goal was to collect approximately 1000 real-world meal photos from 100 participants consuming at least three meals per day, for a total of three study days. Participants were recruited nationally and completed enrollment meetings via web-based video conferencing. Participants uploaded and annotated their meal photos using a mobile phone app called Bitesnap and completed food records using the Automated Self-Administered 24-hour Dietary Assessment Tool (ASA24®) on the same day. A sizing marker with black and white boxes of known size were included in meal photos to assist with portion estimation. Participants included photos before and after eating non-packaged and multi-serving packaged meals, as well as photos of the front package label and ingredient label for single-serving packaged foods. Results By the end of the study, 90 participants had completed all three days of data collection. Examples showing the utility of SNAPMe data with respect to artificial intelligence will be presented. Conclusions The SNAPMe dataset will be made publicly available and will link meal photos, annotations, write-in notes, and ASA24 food records together. These data will be transformative for the improvement of artificial intelligence algorithms for the adoption of photo-based dietary assessment in nutrition research. Funding Sources This work was supported by the United States Department of Agriculture (USDA)/NSF AI Institute for Next Generation Food Systems (AIFS), USDA award number 2020-67,021-32,855.

Project description:ObjectiveTo evaluate the performance of a deep learning system (DLS) in classifying the severity of papilledema associated with increased intracranial pressure on standard retinal fundus photographs.MethodsA DLS was trained to automatically classify papilledema severity in 965 patients (2,103 mydriatic fundus photographs), representing a multiethnic cohort of patients with confirmed elevated intracranial pressure. Training was performed on 1,052 photographs with mild/moderate papilledema (MP) and 1,051 photographs with severe papilledema (SP) classified by a panel of experts. The performance of the DLS and that of 3 independent neuro-ophthalmologists were tested in 111 patients (214 photographs, 92 with MP and 122 with SP) by calculating the area under the receiver operating characteristics curve (AUC), accuracy, sensitivity, and specificity. Kappa agreement scores between the DLS and each of the 3 graders and among the 3 graders were calculated.ResultsThe DLS successfully discriminated between photographs of MP and SP, with an AUC of 0.93 (95% confidence interval [CI] 0.89-0.96) and an accuracy, sensitivity, and specificity of 87.9%, 91.8%, and 86.2%, respectively. This performance was comparable with that of the 3 neuro-ophthalmologists (84.1%, 91.8%, and 73.9%, p = 0.19, p = 1, p = 0.09, respectively). Misclassification by the DLS was mainly observed for moderate papilledema (Frisén grade 3). Agreement scores between the DLS and the neuro-ophthalmologists' evaluation was 0.62 (95% CI 0.57-0.68), whereas the intergrader agreement among the 3 neuro-ophthalmologists was 0.54 (95% CI 0.47-0.62).ConclusionsOur DLS accurately classified the severity of papilledema on an independent set of mydriatic fundus photographs, achieving a comparable performance with that of independent neuro-ophthalmologists.Classification of evidenceThis study provides Class II evidence that a DLS using mydriatic retinal fundus photographs accurately classified the severity of papilledema associated in patients with a diagnosis of increased intracranial pressure.

Dataset Information

Standardizing the protocol for hemispherical photographs: accuracy assessment of binarization algorithms.

Publications

Standardizing the protocol for hemispherical photographs: accuracy assessment of binarization algorithms.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets