Search NASA⌕ Search

SEARCH · Search NASA

Results for “Image processing.”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 631 records · Page 35

Statistical CT noise reduction with multiscale decomposition and penalized weighted least squares in the projection domain

Purposes: The suppression of noise in x-ray computed tomography (CT) imaging is of clinical relevance for diagnostic image quality and the potential for radiation dose saving. Toward this purpose, statistical noise reduction methods in either the image or projection domain have been proposed, which employ a multiscale decomposition to enhance the performance of noise suppression while maintaining image sharpness. Recognizing the advantages of noise suppression in the projection domain, the authors propose a projection domain multiscale penalized weighted least squares (PWLS) method, in which the angular sampling rate is explicitly taken into consideration to account for the possible variation of interview sampling rate in advanced clinical or preclinical applications. Methods: The projection domain multiscale PWLS method is derived by converting an isotropic diffusion partial differential equation in the image domain into the projection domain, wherein a multiscale decomposition is carried out. With adoption of the Markov random field or soft thresholding objective function, the projection domain multiscale PWLS method deals with noise at each scale. To compensate for the degradation in image sharpness caused by the projection domain multiscale PWLS method, an edge enhancement is carried out following the noise reduction. The performance of the proposed method is experimentally evaluated and verified using the projection data simulated by computer and acquired by a CT scanner. Results: The preliminary results show that the proposed projection domain multiscale PWLS method outperforms the projection domain single-scale PWLS method and the image domain multiscale anisotropic diffusion method in noise reduction. In addition, the proposed method can preserve image sharpness very well while the occurrence of 'salt-and-pepper' noise and mosaic artifacts can be avoided. Conclusions: Since the interview sampling rate is taken into account in the projection domain multiscale decomposition, the proposed method is anticipated to be useful in advanced clinical and preclinical applications where the interview sampling rate varies.

60 APPLIED LIFE SCIENCES↗

Impacts of advanced diesel combustion operation on soot nanostructure and reactivity

Advanced diesel combustion, accomplished via a single pulse fuel injection and high levels of exhaust gas recirculation, is shown to be a path to reduce oxides of nitrogen and particulate matter simultaneously. This is well established in the literature. Less established is how well such dilute combustion processes influence soot formation and affect soot that is emitted from diesel engines under such combustion modes. This work focuses on characterization of the nanostructure and oxidative reactivity of soot generated by a light-duty turbodiesel engine operating under a dilute, low-temperature combustion process referred to as high-efficiency clean combustion. The high-efficiency clean combustion soot samples are shown to have a fullerenic nanostructure, characterized by high levels of tortuosity of the fringe layers as seen in transmission electron micrograph images and as quantified using an image processing algorithm. Thermogravimetric analysis of the high-efficiency clean combustion soot samples shows that they have higher rates of oxidation than soot samples from a conventional diesel combustion mode. The linkage between the nanostructure of the high-efficiency clean combustion soot samples and their oxidative reactivity reinforces and supports the structure–property relationship for soot that greater curvature on the soot nanostructure leads to significant increases in oxidative reactivity.

Yehliu, Kuen↗

Automated Progress Monitoring in Modular Construction Factories Using Computer Vision and Building Information Modeling

Modular construction methods have recently gained interest due to the advantages offered in terms of safety, quality, and productivity for projects. In this method, a significant portion of the construction is performed off-site in factories where modular components are built in different workstations, assembled on the production line, and shipped to the site for installation. Due to the labor-intensive nature of tasks, cycle times in modular construction factories are highly variable, which commonly leads to major bottlenecks and delays in construction projects. To remedy this effect, recent methods rely on sensors such as RFID to monitor the production process, which is reportedly expensive, and intrusive to the work process. Recently, computer vision-based methods have been proposed to track the production process in modular construction factories. However, these methods overlook monitoring the assembly process on the production line. Therefore, this paper presents a method to monitor the assembly process by integrating computer vision-based methods with Building Information Modeling (BIM). The proposed method detects the modular units using object segmentation; superimposes the installation area with the corresponding 2D region using BIM, and identifies the installation of the components using image processing techniques. The proposed method has been validated using surveillance videos captured from a modular construction factory in the US. Successful implementation of the proposed method can lead to timely identification of delays during the assembly process and reduce delays in modular integrated construction projects.

building information modeling↗

Oak Ridge National Laboratory Progress Update on Segmentation of Neutron Tomographic Images, 2017

The present report documents progress regarding segmentation of fast-neutron tomographic images of objects consisting of assemblies of machined parts with distinct boundaries. This report satisfies the fiscal year 2017 technical deliverable for project OR16-3DTomography-PD3Jb, “3D Tomography and Image Processing Using Fast Neutrons,” to report on the extension of geometric shape-finding algorithms to three dimensions. The project has two overall goals. The first of these goals is to extend associated particle fast neutron transmission and, particularly, induced-reaction tomographic imaging algorithms to three dimensions. The second of these goals is to automatically segment the resultant tomographic images into constituent parts, and then extract information about the parts, such as the class of shape and potentially the shape parameters. This report addresses continued progress in the component of the project concerned with image segmentation.

3D tomography↗

Progress Update on Segmentation of Neutron Tomographic Images, 2018

The present report documents the techniques developed to segment fast-neutron tomographic images of objects consisting of assemblies of machined parts with distinct boundaries. This report satisfies the fiscal year 2018 technical deliverable for project OR16-3DTomography-PD3Jb, “3D Tomography and Image Processing Using Fast Neutrons,” to report on the extension of geometric shape-finding algorithms to three dimensions. The project has two overall goals. The first of these goals is to extend associated-particle fast-neutron transmission and, particularly, induced-reaction tomographic imaging algorithms to three dimensions. This aspect of the project is beyond the scope of this report. The second goal is to automatically segment the resultant tomographic images into constituent parts and then extract information about the parts, such as the class of shape and potentially the shape parameters. This report describes progress for the image segmentation part of the project. Imaging techniques have been developed as high-confidence methods for confirming the presence and configuration of special nuclear materials (SNM). This high confidence is achieved at the cost of revealing considerable information that may be undesirable to share with the operator of the equipment. One potential way to make use of the high-confidence of imaging methods without revealing imaging data is to employ automated analysis that can extract meaningful attributes of the SNM without showing imaging data to the operator. An essential step of this automated analysis is the segmentation of the image into its constituent parts. For example, a three-dimensional (3D) fast-neutron tomographic image of an assembly containing uranium could be reconstructed and then segmented into discrete parts. The properties of a constituent part identified as uranium would be of interest and could be extracted. For instance, the volume and density of the part would together determine its mass. Moreover, its shape might be expected to fall into a particular geometric class, such as a cube, cylinder, or sphere, and the parameters of the shape (such as side length, height, or diameter) would be expected to fall in a given range. In such an analysis, automated identification of the boundaries and properties of the uranium part is an essential step. A key goal of the present work is to infer the boundaries of objects in fast-neutron tomographic images via the general method of image segmentation. Development of this capability can be broken into more manageable steps, including 1. operator-guided segmentation of two-dimensional (2D) fast-neutron tomographic images, 2. operator-guided segmentation of 3D fast-neutron tomographic images, 3. development of automated segmentation algorithms, and 4. extraction of shape parameters from constituent volumes in 3D. In the first year, the project team reported progress on the first step. Last year, progress on the second step, the extension into 3D, was reported with application to simulated data. The present report describes additional progress on the second step, and progress toward the third step, automated segmentation algorithms. In particular, a multiphase level set approach for image segmentation in three dimensions has been applied to measured 3D data for the first time. Activities related to the fourth step will also be reported, but it will focus on estimation of 3D volumes rather than shape parameters. In this report, the steps before image segmentation will be described, and the multiphase level sets approach to image segmentation will be briefly reviewed. This will be followed by a discussion of the issues encountered when applying the level set technique to measured 3D transmission images, and then a description of the inclusion of induced-reaction data will be provided. Finally, segmentation results for various measured target assemblies will be presented and the image segmentation portion of this project will be summarized and future directions will be discussed.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Optimizing radiotherapy protocols using computer automata to model tumour cell death as a function of oxygen diffusion processes

The concept of hypofractionation is gaining momentum in radiation oncology centres, enabled by recent advances in radiotherapy apparatus. The gain of efficacy of this innovative treatment must be defined. We present a computer model based on translational murine data for in silico testing and optimization of various radiotherapy protocols with respect to tumour resistance and the microenvironment heterogeneity. This model combines automata approaches with image processing algorithms to simulate the cellular response of tumours exposed to ionizing radiation, modelling the alteration of oxygen permeabilization in blood vessels against repeated doses, and introducing mitotic catastrophe (as opposed to arbitrary delayed cell-death) as a means of modelling radiation-induced cell death. Published data describing cell death in vitro as well as tumour oxygenation in vivo are used to inform parameters. Our model is validated by comparing simulations to in vivo data obtained from the radiation treatment of mice transplanted with human prostate tumours. We then predict the efficacy of untested hypofractionation protocols, hypothesizing that tumour control can be optimized by adjusting daily radiation dosage as a function of the degree of hypoxia in the tumour environment. Further biological refinement of this tool will permit the rapid development of more sophisticated strategies for radiotherapy.

60 APPLIED LIFE SCIENCES↗

The DECam Plane Survey: Optical Photometry of Two Billion Objects in the Southern Galactic Plane

The DECam Plane Survey is a five-band optical and near-infrared survey of the southern Galactic plane with the Dark Energy Camera at Cerro Tololo. The survey is designed to reach past the main-sequence turn-off of old populations at the distance of the Galactic center through a reddening $E(B-V)$ of 1.5 mag. Typical single-exposure depths are 23.7, 22.8, 22.3, 21.9, and 21.0 mag (AB) in the grizY bands, with seeing around $1^{\prime\prime} $. The footprint covers the Galactic plane with $| b| \lesssim 4^\circ $, $5^\circ \gt l\gt -120^\circ $. The survey pipeline simultaneously solves for the positions and fluxes of tens of thousands of sources in each image, delivering positions and fluxes of roughly two billion stars with better than 10 mmag precision. Most of these objects are highly reddened and deep in the Galactic disk, probing the structure and properties of the Milky Way and its interstellar medium. Furthermore, the fully-processed images and derived catalogs are publicly available.

79 ASTRONOMY AND ASTROPHYSICS↗

Deep data analytics for genetic engineering of diatoms linking genotype to phenotype via machine learning

Abstract Genome engineering for materials synthesis is a promising avenue for manufacturing materials with unique properties under ambient conditions. Biomineralization in diatoms, unicellular algae that use silica to construct micron-scale cell walls with nanoscale features, is an attractive candidate for functional synthesis of materials for applications including photonics, sensing, filtration, and drug delivery. Therefore, controllably modifying diatom structure through targeted genetic modifications for these applications is a very promising field. In this work, we used gene knockdown in Thalassiosira pseudonana diatoms to create modified strains with changes to structural morphology and linked genotype to phenotype using supervised machine learning. An artificial neural network (NN) was developed to distinguish wild and modified diatoms based on the SEM images of frustules exhibiting phenotypic changes caused by a specific protein (Thaps3_21880), resulting in 94% detection accuracy. Class activation maps visualized physical changes that allowed the NNs to separate diatom strains, subsequently establishing a specific gene that controls pores. A further NN was created to batch process image data, automatically recognize pores, and extract pore-related parameters. Class interrelationship of the extracted paraments was visualized using a multivariate data visualization tool, called CrossVis, and allowed to directly link changes in morphological diatom phenotype of pore size and distribution with changes in the genotype.

59 BASIC BIOLOGICAL SCIENCES↗

Extraction of Vibration Data with Imaging

To date, the primary sensing technology used to measure the vibration response has been accelerometers and strain gages mounted directly to the structure and using either wired or, more recently, wireless telemetry. Cost issues with these sensors and the associated data acquisition systems typically limit the numbers that are deployed on in situ structures. Although there are a few structures with larger sensing counts that in some cases exceed over 1000 sensors, more typical numbers range from ten to one hundred sensors resulting in low spatial resolution when they are applied to physically large systems. When one considers that nuclear power plant structures usually have complex geometries, material properties, connectivity and boundary conditions, it is clear these current approaches to vibration measurements can only provide limited information about a system’s dynamics response characteristics. As an alternative, many non-contact measurement technologies have emerged, including point wise measurement methods such as Global Positioning System (GPS), microwave interferometry, and laser Doppler vibrometry (LDV), as well as simultaneous full-field measurement methods such as electronic speckle pattern interferometry, holography interferometry, and muon tomography, some of which can provide high spatial resolution measurements. Among these methods, digital video imaging techniques have emerged as a feasible solution for full-field vibration measurements that provide significantly more detailed dynamic response information because every pixel becomes a measurement point. Furthermore, recent advances in image processing and computer vision algorithms have been successfully used to process video data for experimental and operational modal analysis. Such full-field measurements have the potential to significantly improve many current structural assessment procedures including system identification (modal parameter estimation), structural health monitoring, load reconstruction, model validation, and model updating. Furthermore, more recent full-field imaging techniques can be accomplished with relatively low-cost, commercially-available off-the-shelf cameras. However, these measurement procedures have other limitations that must be considered such as the ability to only measure visibly accessible points on a structure and a more limited dynamic range and bandwidth than can be achieved with accelerometers or strain gages.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Few-Shot Learning Enables Population-Scale Analysis of Leaf Traits in Populus trichocarpa

Plant phenotyping is typically a time-consuming and expensive endeavor, requiring large groups of researchers to meticulously measure biologically relevant plant traits, and is the main bottleneck in understanding plant adaptation and the genetic architecture underlying complex traits at population scale. In this work, we address these challenges by leveraging few-shot learning with convolutional neural networks to segment the leaf body and visible venation of 2,906 Populus trichocarpa leaf images obtained in the field. In contrast to previous methods, our approach (a) does not require experimental or image preprocessing, (b) uses the raw RGB images at full resolution, and (c) requires very few samples for training (e.g., just 8 images for vein segmentation). Traits relating to leaf morphology and vein topology are extracted from the resulting segmentations using traditional open-source image-processing tools, validated using real-world physical measurements, and used to conduct a genome-wide association study to identify genes controlling the traits. In this way, the current work is designed to provide the plant phenotyping community with (a) methods for fast and accurate image-based feature extraction that require minimal training data and (b) a new population-scale dataset, including 68 different leaf phenotypes, for domain scientists and machine learning researchers. All of the few-shot learning code, data, and results are made publicly available.

59 BASIC BIOLOGICAL SCIENCES↗

TWO NOVEL ACM (ACTIVE CONTOUR MODEL) METHODS FOR INTRAVASCULAR ULTRASOUND IMAGE SEGMENTATION

One of the attractive image segmentation methods is the Active Contour Model (ACM) which has been widely used in medical imaging as it always produces sub-regions with continuous boundaries. Intravascular ultrasound (IVUS) is a catheter based medical imaging technique which is used for quantitative assessment of atherosclerotic disease. Two methods of ACM realizations are presented in this paper. The gradient descent flow based on minimizing energy functional can be used for segmentation of IVUS images. However this local operation alone may not be adequate to work with the complex IVUS images. The first method presented consists of basically combining the local geodesic active contours and global region-based active contours. The advantage of combining the local and global operations is to allow curves deforming under the energy to find only significant local minima and delineate object borders despite noise, poor edge information and heterogeneous intensity profiles. Results for this algorithm are compared to standard techniques to demonstrate the method's robustness and accuracy. In the second method, the energy function is appropriately modified and minimized using a Hopfield neural network. Proper modifications in the definition of the bias of the neurons have been introduced to incorporate image characteristics. The method overcomes distortions in the expected image pattern, such as due to the presence of calcium, and employs a specialized structure of the neural network and boundary correction schemes which are based on a priori knowledge about the vessel geometry. The presented method is very fast and has been evaluated using sequences of IVUS frames.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Fast Tomography: A Study Of Image Quality

We present tomographic slices of fast X-ray computed microtomography, focusing on one hand on the comparison of scans with identical scan time but varying experimental settings, on the other hand on comparing scans with total scan times that are two orders of magnitude apart. We show that streaks in the Fourier-space of the transformed images appear when the number of projections is reduced. These streaks exhibit spacing by the same angle as the one that is covered by the sample during the exposure time of one projection. Further a method for the estimation of information loss in fast scans is presented.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Development of an X-ray Micro-Laminography System at SPring-8

An x-ray micro-laminography system has been developed at SPring-8 and applied to measurements of planar objects that cannot be observed by x-ray computed tomography. To reconstruct a sectional image parallel to the planar surface, a filtered back-projection technique considering the inclination angle of a rotational axis was developed. Experimental conditions suitable for laminography measurements in this system were investigated. Spatial resolution in the reconstructed plane was almost determined by the detector resolution, while the spatial resolution perpendicular to the planar surface tended to be worse by the artifacts inherent of the laminographic scanning and reconstruction method. A laminated test sample was clearly observed in three dimensions.

07 ISOTOPE AND RADIATION SOURCES↗

ORBITAL MOTION OF HR 8799 b, c, d USING HUBBLE SPACE TELESCOPE DATA FROM 1998: CONSTRAINTS ON INCLINATION, ECCENTRICITY, AND STABILITY

HR 8799 is currently the only multiple-planet system that has been detected with direct imaging, with four giant planets of masses 7-10 M{sub Jup} orbiting at large separations (15-68 AU) from this young late A star. Orbital motion provides insight into the stability and possible formation mechanisms of this planetary system. Dynamical studies can also provide constraints on the planets' masses, which help calibrate evolutionary models, yet measuring the orbital motion is a very difficult task because the long-period orbits (50-500 yr) require long time baselines and high-precision astrometry. This paper studies the three planets HR 8799b, c, and d in the archival data set of HR 8799 obtained with the Hubble Space Telescope (HST) NICMOS coronagraph in 1998. The detection of all three planets is made possible by a careful optimization of the Locally Optimized Combination of Images algorithm, and we used a statistical analysis of a large number of reduced images. This work confirms previous astrometry for planet b and presents new detections and astrometry for planets c and d. These HST images provide a ten-year baseline with the discovery images from 2008, and therefore offer a unique opportunity to constrain their orbital motion now. Recent dynamical studies of this system show the existence of a few possible stable solutions involving mean motion resonances (MMRs), where the interaction between c and d plays a major role. We study the compatibility of a few of these stable scenarios (1d:1c, 1d:2c, or 1d:2c:4d) with the new astrometric data from HST. In the hypothesis of a 1d:2c:4b MMR our best orbit fit is close to the stable solution previously identified for a three-planet system and involves low eccentricity for planet d (e{sub d} = 0.10) and moderate inclination of the system (i = 28.0 deg), assuming a coplanar system, circular orbits for b and c, and exact resonance with integer period ratios. Under these assumptions, we can place strong constraints on the inclination of the system (27.3-31.4 deg) and on the eccentricity for d e{sub d} < 0.46. Our results are robust to small departures from exact integer period ratios and consistent with previously published results based on dynamical studies for a three-planet system prior to the discovery of the fourth planet.

79 ASTRONOMY AND ASTROPHYSICS↗

Anatomical noise in contrast-enhanced digital mammography. Part I. Single-energy imaging

Purpose: The use of an intravenously injected iodinated contrast agent could help increase the sensitivity of digital mammography by adding information on tumor angiogenesis. Two approaches have been made for clinical implementation of contrast-enhanced digital mammography (CEDM), namely, single-energy (SE) and dual-energy (DE) imaging. In each technique, pairs of mammograms are acquired, which are then subtracted with the intent to cancel the appearance of healthy breast tissue to permit sensitive detection and specific characterization of lesions. Patterns of contrast agent uptake in the healthy parenchyma, and uncanceled signal from background tissue create a 'clutter' that can mask or mimic an enhancing lesion. This type of 'anatomical noise' is often the limiting factor in lesion detection tasks, and thus, noise quantification may be useful for cascaded systems analysis of CEDM and for phantom development. In this work, the authors characterize the anatomical noise in CEDM clinical images and the authors evaluate the influence of the x-ray energy used for acquisition, the presence of iodine in the breast, and the timing of imaging postcontrast administration on anatomical noise. The results are presented in a two-part report, with SE CEDM described here, and DE CEDM in Part II. Methods: A power law is used to model anatomical noise in CEDM images. The exponent, {beta}, which describes the anatomical structure, and the constant {alpha}, which represents the magnitude of the noise, are determined from Wiener spectra (WS) measurements on images. A total of 42 SE CEDM cases from two previous clinical pilot studies are assessed. The parameters {alpha} and {beta} are measured both from unprocessed images and from subtracted images. Results: Consistent results were found between the two SE CEDM pilot studies, where a significant decrease in {beta} from a value of approximately 3.1 in the unprocessed images to between about 1.1 and 1.8 in the subtracted images was observed. Increasing the x-ray energy from that used in conventional DM to those of typical SE CEDM spectra with mean energies above 33 keV significantly decreased {alpha} by about a factor of 19, in agreement with theory. Compared to precontrast images, in the unprocessed postcontrast images at 30 s postinjection, {alpha} was larger by about 7.4 Multiplication-Sign 10{sup -7} mm{sup 2} and {beta} was decreased by 0.2. While {alpha} did not vary significantly with the time after contrast administration, {beta} from the unprocessed image WS increased linearly, and {beta} from subtracted image WS increased with an initial quadratic relationship that plateaued by about 5 min postinjection. Conclusions: The presence of an iodinated contrast agent in the breast produced small, but significant changes in the power law parameters of unprocessed CEDM images compared to the precontrast images. Image subtraction in SE CEDM significantly reduced anatomical noise compared to conventional DM, with a reduction in both {alpha} and {beta} by about a factor of 2. The data presented here, and in Part II of this work, will be useful for modeling of CEDM backgrounds, for systems characterization and for lesion detectability experiments using models that account for anatomical noise.

60 APPLIED LIFE SCIENCES↗

Postprocessing Algorithm for Driving Conventional Scanning Tunneling Microscope at Fast Scan Rates

We present an image postprocessing framework for Scanning Tunneling Microscope (STM) to reduce the strong spurious oscillations and scan line noise at fast scan rates and preserve the features, allowing an order of magnitude increase in the scan rate without upgrading the hardware. The proposed method consists of two steps for large scale images and four steps for atomic scale images. For large scale images, we first apply for each line an image registration method to align the forward and backward scans of the same line. In the second step we apply a “rubber band” model which is solved by a novel Constrained Adaptive and Iterative Filtering Algorithm (CIAFA). The numerical results on measurement from copper(111) surface indicate the processed images are comparable in accuracy to data obtained with a slow scan rate, but are free of the scan drift error commonly seen in slow scan data. For atomic scale images, an additional first step to remove line-by-line strong background fluctuations and a fourth step of replacing the postprocessed image by its ranking map as the final atomic resolution image are required. The resulting image restores the lattice image that is nearly undetectable in the original fast scan data.

47 OTHER INSTRUMENTATION↗

Physics guided machine learning for multi-material decomposition of tissues from dual-energy CT scans of simulated breast models with calcifications

We introduce a physics guided data-driven method for image-based multi-material decomposition for dual-energy computed tomography (CT) scans. The method is demonstrated for CT scans of virtual human phantoms containing more than two types of tissues. The method is a physics-driven supervised learning technique. We take advantage of the mass attenuation coefficient of dense materials compared to that of muscle tissues to perform a preliminary extraction of the dense material from the images using unsupervised methods. We then perform supervised deep learning on the images processed by the extracted dense material to obtain the final multi-material tissue map. The method is demonstrated on simulated breast models with calcifications as the dense material placed amongst the muscle tissues. The physics-guided machine learning method accurately decomposes the various tissues from input images, achieving a normalized root-mean-squared error of 2.75%.

Gopalakrishnan Meena, Murali↗

A Combined Computer Vision and Deep Learning Approach for Rapid Drone-Based Optical Characterization of Parabolic Troughs

Optical accuracy is a primary driver of parabolic trough concentrating solar power (CSP) plant performance, but can be damaged by wind loads, gravity, error during installation, and regular plant operation. Collecting and analyzing optical measurements over an entire operating parabolic trough plant is difficult, given the large scale of typical installations. Distant Observer, a software tool developed at the National Renewable Energy Laboratory, uses images of the absorber tube reflected in the collector mirror to measure both surface slope in the parabolic mirror and offset of the absorber tube from the ideal focal point. This technology has been adapted for fast data collection using low-cost commercial drones, but until recently still required substantial human labor to process large amounts of data. A new method leveraging advanced deep learning and computer vision tools can drastically reduce the time required to process images. This new method addresses the primary analysis bottleneck, identifying featureless, reflective mirror corner points to a high degree of accuracy. Recent work has shown promising results using computer vision methods. The combined deep learning and computer vision approach presented here proved highly effective and has the potential to further automate data collection and analysis, making the tool more robust. The method presented in this paper automatically identified 74.3% of mirror corners within 2 pixels of their manually marked counterparts and 91.9% within 3 pixels. This level of accuracy is sufficient for practical Distant Observer analysis within a target uncertainty. A commercial drone collected video of over 100 parabolic trough modules at an operating CSP plant to demonstrate the deep learning and computer vision method's usefulness in processing large amounts of data. These troughs were successfully analyzed using Distant Observer, paired with the new deep learning and computer vision algorithm, and can provide plant operators and trough designers with valuable insight about plant performance, operating strategies, and plant-wide optical error trends.

computer vision↗