Shape Model Fitting Research Articles

A spatiotemporal shape model fitting method for within-season crop phenology detection

Crop phenological information must be reliably acquired earlier in the growing season to benefit agricultural management. Although the popular shape model fitting (SMF) method and its various improved versions (e.g., SMF by the Separate phenological stage, SMF-S) have been successfully applied to after-season crop phenology detection, these existing methods cannot be applied to within-season crop phenology detection. This discrepancy arises due to the fact that, in the within-season scenario, phenological stages can beyond the defined cut-off time. Consequently, enhancing the alignment of the vegetation index (VI) curve segments prior to the cut-off time does not necessarily guarantee accurate within-season phenological detection. To resolve this issue, a new method named spatiotemporal shape model fitting (STSMF) was developed. STSMF does not seek to optimize the local curve matching between the target pixel and the shape model; instead, it determines similar local VI trajectories in the neighboring pixels of previous years. The within-season phenology of the target pixel was thus estimated from the corresponding phenological stage of the determined local VI trajectories. When compared with ground phenology observations, STSMF outperformed the existing SMF and SMF-S which were modified for the within-season scenario (SMFws and SMFSws) with the smallest mean absolute differences (MAE) between observed phenological stages and their corresponding model estimates. The MAE values averaged over all phenological stages for STSMF, SMFSws, and SMFws were 9.8, 12.4, and 27.1 days at winter wheat stations; 8.4, 14.9, and 55.3 days at corn stations; and 7.9, 12.4, and 64.6 days at soybean stations, respectively. Intercomparisons between after-season and within-season regional phenology maps also demonstrated the superior performance of STSMF (e.g., correlation coefficients for STSMF and SMFSws are 0.89 and 0.80 at the maturity stage of winter wheat). Furthermore, the performance of STSMF was less affected by the detection time and the determination of shape models. In conclusion, the straightforward, effective, and stable nature of STSMF makes it suitable for within-season detection of agronomic phenological stages.

Software Approach to BMI and Obesity Determination using Facial Extraction Techniques

The focus on COVID-19 pandemic has led doctors to reduce consultation and restrict hospital visitations by patients, a decision that has resulted in the increased cases of obesity coupled with restrictions on movement and exercises. Obesity places the health of people in danger, since it is associated with poor mental health, reduce quality of life, increases the risk of diabetes, heart disease, stroke and introduces certain types of cancer. The aim of this study therefore is to design and develop a face-to-BMI mobile application that enable people to learn about their Body Mass Index (BMI), get their obesity status, as well as receive doctor recommendations (or counsel) from the comfort of their homes using their mobile devices. The design science research methodology was used in the design and development of the mobile application for obesity assessment. A mobile phone is required to take the face photo of the users after which the algorithm will perform face detection, looking for all possible facial features such as the width-to-upper facial height ratio (WHR), cheek-to-jaw width (WJWR), perimeter of area ratio (PAR), lower face to face height ratio (FW/FH), mean of eyebrow height (MEH) using facial measurements such as the iris, mouth corners, eyebrows, and nostril. The algorithm then performs Active Shape Model (ASM) fitting. The mobile app was tested with five (5) participants and the results have shown significant improvement in obesity detection and ease of use

Detecting crop phenology from vegetation index time-series data by improved shape model fitting in each phenological stage

Crop phenology provides important information for crop growth management and yield estimations. The popular shape model fitting (SMF) method detects crop phenology from vegetation index (VI) time-series data, but it has two limitations. First, SMF assumes the same “relative position” of phenological stages for the pixels of the same crop type. This assumption is valid only if all target pixels, relative to the shape model, display a synchronized increase (or decrease) in length between any two phenological stages, which is uncommon in practice. Second, the variance in the resulting phenology estimates for a particular phenological stage is related to the stage itself; this makes it challenging to simulate spatial and temporal variations in crop phenology using SMF. Here, we address both limitations by developing the shape model fitting by the Separate phenological stage method (“SMF-S"). SMF-S uses a modified fitting function and an iterative procedure to match the shape model with the VI time series for each phenological stage in an adaptive local window. Comparisons between SMF-S and SMF in simulation experiments show the superior performance of SMF-S in different scenarios, regardless of noise. Comparisons involving winter wheat field observations from the North China Plain showed that the RMSE values averaged over nine phenological stages were smaller for SMF-S (RMSE = 9.5 d) than for SMF (RMSE = 13.4 d) and one variant of SMF (the shape model with accumulated growing degree days (SM-AGDD); RMSE=33.6 d). Moreover, SMF-S better described the spatial variations (i.e., variance) in the results and captured the temporal shifts in multiple phenological stages. In the derived regional phenology maps of winter wheat on the North China Plain, SMF-S generated more reasonable spatial patterns, whereas SMF underestimated (overestimated) the variance in the early (late) phenological stages. We expect that the improved crop phenology estimates obtained with SMF-S could benefit various agricultural activities.

Structural Analysis of an Octameric Resorcinarene Self-Assembly in Toluene and its Morphological Transition by Temperature.

Calix[4]resorcinarene derivatives such as C-undecylcalix[4]resorcinarene self-assemble into a hexameric capsule-like structure. This structure is expected in any apolar solvent, but we recently found a self-assembled octameric form of resorcinarene in toluene. To understand this unexpected form, we performed small-angle X-ray scattering (SAXS) measurements of the octameric self-assembly. In ab initio shape reconstruction and model fitting of the SAXS profile, a core-shell spherical structure resembling a reverse micelle was found in the octameric self-assembly. The shell and core were composed of alkyl chains and resorcinol moieties, respectively. We also evaluated the temperature dependence of the octameric self-assembly. Above 95 °C, the structure began directly decomposing into unimers, implying the partial cleavage of hydrogen bonds in the octamer core. Meanwhile, below 25 °C, the spherical structure of the octamer partially transformed to a cylindrical structure. This morphological transition could be understood by packing parameter theory, which is often applied to suspected micellar structures.

Assessing the Impact of Satellite Revisit Rate on Estimation of Corn Phenological Transition Timing through Shape Model Fitting

Agricultural monitoring is an important application of earth-observing satellite systems. In particular, image time-series data are often fit to functions called shape models that are used to derive phenological transition dates or predict yield. This paper aimed to investigate the impact of imaging frequency on model fitting and estimation of corn phenological transition timing. Images (PlanetScope 4-band surface reflectance) and in situ measurements (Soil Plant Analysis Development (SPAD) and leaf area index (LAI)) were collected over a corn field in the mid-Atlantic during the 2018 growing season. Correlation was performed between candidate vegetation indices and SPAD and LAI measurements. The Normalized Difference Vegetation Index (NDVI) was chosen for shape model fitting based on the ground truth correlation and initial fitting results. Plot-average NDVI time-series were cleaned and fit to an asymmetric double sigmoid function, from which the day of year (DOY) of six different function parameters were extracted. These points were related to ground-measured phenological stages. New time-series were then created by removing images from the original time-series, so that average temporal spacing between images ranged from 3 to 24 days. Fitting was performed on the resampled time-series, and phenological transition dates were recalculated. Average range of estimated dates increased by 1 day and average absolute deviation between dates estimated from original and resampled time-series data increased by 1/3 of a day for every day of increase in average revisit interval. In the context of this study, higher imaging frequency led to greater precision in estimates of shape model fitting parameters used to estimate corn phenological transition timing.

Cascaded statistical shape model based segmentation of the full lower limb in CT

Image segmentation has become an important tool in orthopedic and biomechanical research. However, it greatly remains a time-consuming and laborious task. In this manuscript, we propose a fully automatic model-based segmentation pipeline for the full lower limb in computed tomography (CT) images. The method relies on prior shape model fitting, followed by a gradient-defined free from deformation. The technique allows for the generation of anatomically corresponding surface meshes, which can subsequently be applied in anatomical and mechanical simulation studies. Starting from an initial, small (n ≤ 10) sample of manual segmentations, the model is continuously updated and refined with newly segmented training samples. Validation of the segmentation pipeline was performed by comparing the automatic segmentations against corresponding manual segmentations. Convergence of the segmentation pipeline was obtained in 250 cases and failed in three samples. The average distance error ranged from 0.53 to 0.76 mm and maximal error ranged from 2.0 to 7.8 mm for the 7 different osteological structures that were investigated. The accuracy of the shape model-based segmentation gradually increased as the number of training shapes in the updated population also increased. When optimized with the free form deformation, however, average segmentation accuracy rapidly plateaued from already as little as 20 training samples on. The maximum segmentation error plateaued from 100 training samples on.

Refined shape model fitting methods for detecting various types of phenological information on major U.S. crops

Crop phenological information is a critical variable in evaluating the influence of environmental stress on the final crop yield in spatio-temporal dimensions. Although the MODIS (Moderate Resolution Imaging Spectroradiometer) Land Cover Dynamics product (MCD12Q2) is widely used in place of crop phenological information, the definitions of MCD12Q2-derived phenological events (e.g. green-up date, dormancy date) were not completely consistent with those of crop development stages used in statistical surveys (e.g. emerged date, harvested date). It has been necessary to devise an alternative method focused on detecting continental-scale crop developmental stages using a different approach. Therefore, this study aimed to refine the Shape Model Fitting (SMF) method to improve its applicability to multiple major U.S. crops. The newly-refined SMF methods could estimate the timing of 36 crop-development stages of major U.S. crops, including corn, soybeans, winter wheat, spring wheat, barley, sorghum, rice, and cotton. The newly-developed calibration process did not require any long-term field observation data, and could calibrate crop-specific phenological parameters, which were used as coefficients in estimated equation, by using only freely accessible public data. The calibration of phenological parameters was conducted in two steps. In the first step, the national common phenological parameters, referred to as X0[base], were calibrated by using the statistical data of 2008. The SMF method coupled using X0[base] was named the rSMF[base] method. The second step was a further calibration to gain regionally-adjusted phenological parameters for each state, referred to as X0[local], by using additional statistical data of 2015 and 2016. The rSMF method using the X0[local] was named the rSMF[local] method. This second calibration process improved the estimation accuracy for all tested crops. When applying the rSMF[base] method to the validation data set (2009–2014), the root mean square error (RMSE) of the rSMF[base]-derived estimates ranged from 7.1 days (corn) to 15.7 days (winter wheat). When using the rSMF[local] method, the RMSE of the rSMF[local]-derived estimates improved and ranged from 5.6 days (corn) to 12.3 days (winter wheat). The results showed that the second calibration step for the rSMF[local] method could correct the region-dependent bias error between the rSMF[base]-derived estimates and the statistical data. A comparison between the performances of the refined SMF methods and the MCD12Q2 products, indicated that both of the rSMF methods were superior to the MCD12Q2 products in estimating all phenological stages, except for the case of the rSMF[base]-derived barley emerged stages. The phenological stages for which the rSMF[local] showed the best estimation accuracy were the corn silking stage (RMSE = 4.3 days); the soybeans dropping leaves stage (RMSE = 4.9 days); the headed stages of winter wheat (RMSE = 11.1 days), barley (RMSE = 6.1 days), and sorghum (RMSE = 9.5 days); the spring-wheat harvested stage (RMSE = 5.5 days); the rice emerged stage (RMSE = 5.5 days), and the cotton squaring stage (RMSE = 6.6 days). These were more accurate than the results achieved by the MCD12Q2 products. In addition, the rSMF[local]-derived estimates were superior in terms of the reproducibility of the annual variation range, particularly of the late reproductive stages, such as the mature and harvested stages. The crop phenology maps derived from the SMF [local] method were also in good agreement with the relevant maps derived from statistics, and could reveal the characteristic spatial pattern of the key phenological stages at the continental scale with fine spatial resolution. For example, the winter-wheat headed stage clearly became later from south to north. The cotton squaring stage became earlier from the central region towards both coastal regions.

Interactive patient-specific 3D approximation of scapula bone shape from 2D X-ray images using landmark-constrained statistical shape model fitting.

We report on an interactive tool for patientspecific 3D approximation of scapula bone shape from 2D X-ray images using landmark-constrained statistical shape model (SSM) fitting. The 3D localization of points on the 2D X-ray images was done through X-ray stereophotogrammetry. The inferior angle, acromion and the coracoid process were identified as reliable landmarks from the anteroposterior (AP) and oblique lateral views in a landmark selection study. The 3D scapula surface was approximated through fitting the scapula SSM to the 3D reconstructed coordinates of the selected landmarks. 3D point localization yielded average (X, Y, Z) coordinate reconstruction errors of (X=0.14, Y=0.07, Z=0.04) mm. The landmark-constrained fitting algorithm yielded an average error between the mean posterior model landmarks and the corresponding target landmarks of 0.49 mm using the three landmarks, and later 0.19 mm with sixteen landmarks. Average surface to surface error between the CT ground truth model and approximated model from within the dataset improved from 3.20 mm to 2.46 mm using three landmarks and using sixteen landmarks, respectively. Average surface to surface error between the CT ground truth model and the approximated model from outside the dataset improved from 4.28 mm to 3.20 mm using three landmarks and using sixteen landmarks, respectively.

A hybrid approach for detecting corn and soybean phenology with time-series MODIS data

Abstract Monitoring crop phenology provides essential information for crop management, as well as for understanding regional to global scale vegetation dynamics. In this study, a hybrid phenology detection method is presented that incorporates the “shape-model fitting” concept of the two-step filtering method and a simulation concept of the crop models to detect the critical vegetative stages and reproductive stages of corn (Zea mays L.) and soybeans (Glycine max L.) from MODIS 250-m Wide Dynamic Range Vegetation Index (WDRVI) time-series data and 1000-m Land Surface Temperature (LST) data. The method was first developed and tested at the field scale over a ten-year period (2003–2012) for three experimental study sites in eastern Nebraska of USA, where the estimated phenology dates were compared to the ground-based phenology observations for both corn and soybeans. The average root mean square error (RMSE) of phenology stage estimation of the individual development stages across all sites ranged from 1.9 to 4.3 days for corn and from 1.9 to 4.9 days for soybeans. The approach was then tested at a regional scale over eastern Nebraska and the state of Iowa to evaluate its ability to characterize the spatio-temporal variation of targeted corn and soybean phenology stage dates over a larger area. Quantitative regional assessments were conducted by comparing the estimated crop stage dates with crop developmental stage statistics reported by the USDA NASS Crop Progress Reports (NASS-CPR) for both eastern Nebraska and Iowa. The accuracy of the regional-scale phenology estimation in Iowa (RMSE ranged from 2.6 to 3.9 days for corn and from 3.2 to 3.9 days for soybeans) was slightly lower than in eastern Nebraska (RMSE ranged from 1.8 to 2.9 days for corn and from 1.7 to 2.9 days for soybeans), However, the estimation accuracy in Iowa was still reasonable with the estimated phenology dates being within 4 days or less of the observed dates for both corn and soybeans.

Implementation of Robust Image Artifact Removal in SWarp through Clipped Mean Stacking

We implement an algorithm for detecting and removing artifacts from astronomical images by means of outlier rejection during stacking. Our method is capable of addressing both small, highly significant artifacts such as cosmic rays and, by applying a filtering technique to generate single frame masks, larger area but lower surface brightness features such as secondary (ghost) images of bright stars. In contrast to the common method of building a median stack, the clipped, or outlier-filtered mean stacked point-spread function (PSF), is a linear combination of the single frame PSFs as long as the latter are moderately homogeneous, a property of great importance for weak lensing shape measurement or model fitting photometry. In addition, our method has superior noise properties, allowing a significant reduction in exposure time compared to median stacking. We make publicly available a modified version of SWarp that implements clipped mean stacking and software to generate single frame masks from the list of outlier pixels.

Pulmonary lobe segmentation based on ridge surface sampling and shape model fitting

Performing lobe-based quantitative analysis of the lung in computed tomography (CT) scans can assist in efforts to better characterize complex diseases such as chronic obstructive pulmonary disease (COPD). While airways and vessels can help to indicate the location of lobe boundaries, segmentations of these structures are not always available, so methods to define the lobes in the absence of these structures are desirable. The authors present a fully automatic lung lobe segmentation algorithm that is effective in volumetric inspiratory and expiratory computed tomography (CT) datasets. The authors rely on ridge surface image features indicating fissure locations and a novel approach to modeling shape variation in the surfaces defining the lobe boundaries. The authors employ a particle system that efficiently samples ridge surfaces in the image domain and provides a set of candidate fissure locations based on the Hessian matrix. Following this, lobe boundary shape models generated from principal component analysis (PCA) are fit to the particles data to discriminate between fissure and nonfissure candidates. The resulting set of particle points are used to fit thin plate spline (TPS) interpolating surfaces to form the final boundaries between the lung lobes. The authors tested algorithm performance on 50 inspiratory and 50 expiratory CT scans taken from the COPDGene study. Results indicate that the authors' algorithm performs comparably to pulmonologist-generated lung lobe segmentations and can produce good results in cases with accessory fissures, incomplete fissures, advanced emphysema, and low dose acquisition protocols. Dice scores indicate that only 29 out of 500 (5.85%) lobes showed Dice scores lower than 0.9. Two different approaches for evaluating lobe boundary surface discrepancies were applied and indicate that algorithm boundary identification is most accurate in the vicinity of fissures detectable on CT. The proposed algorithm is effective for lung lobe segmentation in absence of auxiliary structures such as vessels and airways. The most challenging cases are those with mostly incomplete, absent, or near-absent fissures and in cases with poorly revealed fissures due to high image noise. However, the authors observe good performance even in the majority of these cases.

MODIS-based corn grain yield estimation model incorporating crop phenology information

A crop yield estimation model using time-series MODIS WDRVI was developed. The main feature of the proposed model is the incorporation of crop phenology detection using MODIS data, called the “Shape-Model Fitting Method”. MODIS WDRVI taken 7–10days before the corn silking stage had strong linear correlation with corn final grain yield at both field and regional scales. The model revealed spatial patterns of corn final grain yield all over the U.S. from 2000 to 2011. State-level corn yield was estimated accurately with coefficient of variation below 10% especially for the 18 major corn producing states including Iowa, Illinois, Delaware, Minnesota, Ohio, West Virginia, Wisconsin, Michigan, Indiana, Nebraska, Kentucky, New York, South Dakota, Missouri, Pennsylvania, Tennessee, New Jersey and Maryland. The results corresponded well with the spatial pattern of high-yield regions derived from the USDA/NASS data. However, the model tended to underestimate corn grain yield in three irrigated regions: the Midwestern region depending on the Ogallala Aquifer, the downstream basin of the Mississippi, and the southwestern region of Georgia. In contrast, it tended to overestimate corn grain yield around the outlying regions of the U.S. Corn Belt, specifically, the East Coast, North Dakota, Minnesota, Wisconsin, and Missouri. The estimation accuracy of the proposed model differed depending on the region. However, the annual variation of state level corn grain yield could be detected with high accuracy, especially in the major corn producing states.

A Two-Step Filtering approach for detecting maize and soybean phenology with time-series MODIS data

The crop developmental stage represents essential information for irrigation scheduling/fertilizer management, understanding seasonal ecosystem carbon dioxide (CO2) exchange, and evaluating crop productivity. In this study, we devised an approach called the Two-Step Filtering (TSF) for detecting the phenological stages of maize and soybean from time-series Wide Dynamic Range Vegetation Index (WDRVI) data derived from Moderate Resolution Imaging Spectroradiometer (MODIS) 250-m observations. The TSF method consists of a Two-Step Filtering scheme that includes: (i) smoothing the temporal WDRVI data with a wavelet-based filter and (ii) deriving the optimum scaling parameters from shape-model fitting procedure. The date of key crop development stages are then estimated by using the optimum scaling parameters and an initial value of the specific phenological date on the shape model, which are preliminary defined in reference to ground-based crop growth stage observations. The shape model is a crop-specific WDRVI curve with typical seasonal features, which were defined by averaging smoothed, multi-year WDRVI profiles from MODIS 250-m data collected over irrigated maize and soybean study sites.In this study, the TSF method was applied to MODIS-derived WDRVI data over a 6-year period (2003 to 2008) for two irrigated sites and one rainfed site planted to either maize or soybean as part of the Carbon Sequestration Program (CSP) at the University of Nebraska-Lincoln. A comparison of satellite-based retrievals with ground-based crop growth stage observations collected by the CSP over the six growing seasons for these three sites showed that the TSF method can accurately estimate the date of four key phenological stages of maize (V2.5: early vegetative stage, R1: silking stage, R5: dent stage and R6: maturity) and soybean (V1: early vegetative stage, R5: beginning seed, R6: full seed and R7: beginning maturity). The root mean square error (RMSE) of phenological-stage estimation for maize ranged from 2.9 [R1] to 7.0 [R5] days and from 3.2 [R6] to 6.9 [R7] days for soybean, respectively. In addition, the TSF method was also applied for two years (2001 and 2002) over eastern Nebraska to test its ability to characterize the spatio-temporal patterns of these key phenological stages over a larger geographic area. The MODIS-derived crop phenological stage dates agreed well with the statistical crop progress data reported by the United State Department of Agriculture (USDA) National Agricultural Statistics Service (NASS) for eastern Nebraska's three crop agricultural statistic districts (ASDs). At the ASD-level, the RMSE of phenological-stage estimation ranged from 1.6 [R1] to 5.6 [R5] days for maize and from 2.5 [R7] to 5.3 [R5] days for soybean.