Search

Simple Summary: Breast cancer is a significant cause of female cancer-related deaths in the US. Checking how severe the cancer is helps in planning treatment. Modern AI methods are good at grading cancer, but they are not used much in hospitals yet. We developed and utilized ensemble deep learning algorithms for addressing the tasks of classifying (1) breast cancer subtype and (2) breast cancer invasiveness from whole slide image (WSI) histopathology slides. The ensemble models used were based on convolutional neural networks (CNNs) known for extracting distinctive features crucial for accurate classification. In this paper, we provide a comprehensive analysis of these models and the used methodology for breast cancer diagnosis tasks. Cancer diagnosis and classification are pivotal for effective patient management and treatment planning. In this study, a comprehensive approach is presented utilizing ensemble deep learning techniques to analyze breast cancer histopathology images. Our datasets were based on two widely employed datasets from different centers for two different tasks: BACH and BreakHis. Within the BACH dataset, a proposed ensemble strategy was employed, incorporating VGG16 and ResNet50 architectures to achieve precise classification of breast cancer histopathology images. Introducing a novel image patching technique to preprocess a high-resolution image facilitated a focused analysis of localized regions of interest. The annotated BACH dataset encompassed 400 WSIs across four distinct classes: Normal, Benign, In Situ Carcinoma, and Invasive Carcinoma. In addition, the proposed ensemble was used on the BreakHis dataset, utilizing VGG16, ResNet34, and ResNet50 models to classify microscopic images into eight distinct categories (four benign and four malignant). For both datasets, a five-fold cross-validation approach was employed for rigorous training and testing. Preliminary experimental results indicated a patch classification accuracy of 95.31% (for the BACH dataset) and WSI image classification accuracy of 98.43% (BreakHis). This research significantly contributes to ongoing endeavors in harnessing artificial intelligence to advance breast cancer diagnosis, potentially fostering improved patient outcomes and alleviating healthcare burdens. [ABSTRACT FROM AUTHOR]

The canopy height model (CHM) derived from LiDAR point cloud data is usually used to accurately identify the position and the canopy dimension of single tree. However, local invalid values (also called data pits) are often encountered during the generation of CHM, which results in low-quality CHM and failure in the detection of treetops. For this reason, this paper proposes an innovative method, called "pixels weighted differential gradient", to filter these data pits accurately and improve the quality of CHM. First, two characteristic parameters, gradient index (GI) and Z-score value (ZV) are extracted from the weighted differential gradient between the pit pixels and their eight neighbors, and then GIs and ZVs are commonly used as criterion for initial identification of data pits. Secondly, CHMs of different resolutions are merged, using the image processing algorithm developed in this paper to distinguish either canopy gaps or data pits. Finally, potential pits were filtered and filled with a reasonable value. The experimental validation and comparative analysis were carried out in a coniferous forest located in Triangle Lake, United States. The experimental results showed that our method could accurately identify potential data pits and retain the canopy structure information in CHM. The root-mean-squared error (RMSE) and mean bias error (MBE) from our method are reduced by between 73% and 26% and 76% and 28%, respectively, when compared with six other methods, including the mean filter, Gaussian filter, median filter, pit-free, spike-free and graph-based progressive morphological filtering (GPMF). The average F1 score from our method could be improved by approximately 4% to 25% when applied in single-tree extraction. [ABSTRACT FROM AUTHOR]

As an emerging cash crop, industrial hemp (Cannabis sativa L.) grown for cannabidiol (CBD) has spurred a surge of interest in the United States. Cultivar selection and harvest timing are important to produce CBD hemp profitably and avoid economic loss resulting from the tetrahydrocannabinol (THC) concentration in the crop exceeding regulatory limits. Hence there is a need for differentiating CBD hemp cultivars and growth stages to aid in cultivar and genotype selection and optimization of harvest timing. Current methods that rely on visual assessment of plant phenotypes and chemical procedures are limited because of its subjective and destructive nature. In this study, hyperspectral imaging was proposed as a novel, objective, and non-destructive method for differentiating hemp cultivars, growth stages as well as plant organs (leaves and flowers). Five cultivars of CBD hemp were grown greenhouse conditions and leaves and flowers were sampled at five growth stages 2–10 weeks in 2-week intervals after flower initiation and scanned by a benchtop hyperspectral imaging system in the spectral range of 400–1000 nm. The acquired images were subjected to image processing procedures to extract the spectra of hemp samples. The spectral profiles and scatter plots of principal component analysis of the spectral data revealed a certain degree of separation between hemp cultivars, growth stages, and plant organs. Machine learning based on regularized linear discriminant analysis achieved the accuracy of up to 99.6% in differentiating the five hemp cultivars. Plant organ and growth stage need to be factored into model development for hemp cultivar classification. The classification models achieved 100% accuracy in differentiating the five growth stages and two plant organs. This study demonstrates the effectiveness of hyperspectral imaging for differentiating cultivars, growth stages and plant organs of CBD hemp, which is a potentially useful tool for growers and breeders of CBD hemp. [ABSTRACT FROM AUTHOR]

The Pacific Northwest is an important pulse production region in the United States. Currently, pulse crop (chickpea, lentil, and dry pea) breeders rely on traditional phenotyping approaches to collect performance and agronomic data to support decision making. Traditional phenotyping poses constraints on data availability (e.g., number of locations and frequency of data acquisition) and throughput. In this study, phenomics technologies were applied to evaluate the performance and agronomic traits in two pulse (chickpea and dry pea) breeding programs using data acquired over multiple seasons and locations. An unmanned aerial vehicle-based multispectral imaging system was employed to acquire image data of chickpea and dry pea advanced yield trials from three locations during 2017–2019. The images were analyzed semi-automatically with custom image processing algorithm and features were extracted, such as canopy area and summary statistics associated with vegetation indices. The study demonstrated significant correlations (P < 0.05) between image-based features (e.g., canopy area and sum normalized difference vegetation index) with yield (r up to 0.93 and 0.85 for chickpea and dry pea, respectively), days to 50% flowering (r up to 0.76 and 0.85, respectively), and days to physiological maturity (r up to 0.58 and 0.84, respectively). Using image-based features as predictors, seed yield was estimated using least absolute shrinkage and selection operator regression models, during which, coefficients of determination as high as 0.91 and 0.80 during model testing for chickpea and dry pea, respectively, were achieved. The study demonstrated the feasibility to monitor agronomic traits and predict seed yield in chickpea and dry pea breeding trials across multiple locations and seasons using phenomics tools. Phenomics technologies can assist plant breeders to evaluate the performance of breeding materials more efficiently and accelerate breeding programs. [ABSTRACT FROM AUTHOR]

The magnetic domain structure is an important factor in discussing the coercivity development mechanism. We have researched on the new observation method of magnetic domain structures using multiple distances observation of magnetic force microscope (MFM) and image processing. We observed magnetic domain structures in thermally demagnetized state and alternating current (ac) demagnetized state of ferrite-sintered magnets using this method. In addition, the initial magnetization curve was measured in each demagnetized state. There were differences in the domain structures and the initial magnetization curves in each demagnetized state. These differences suggest that irreversible multi-domain gains (IRMDGs) confirmed in rare-earth magnets are present even in sintered ferrite magnets. [ABSTRACT FROM AUTHOR]

Teleseismic receiver functions (RFs) are frequently used to determine depths of seismic discontinuities in the crust and upper mantle. We developed an efficient reverse-time migration (RTM) method that is applied to teleseismic receiver functions directly. Both the primary P -to- S converted phases and their crustal multiples in RFs can be used for imaging seismic discontinuities. The method uses the phase-shift-plus-interpolation algorithm to extrapolate both the source and receiver wavefields in a 3-D velocity model, which greatly reduces the computation costs compared with those using a full wave-equation numerical solver. Tests using synthetic data in various crustal models demonstrate the effectiveness of the method and its superiority over the common-conversion-point stacking method. In particular, the method handles diffraction caused by strong lateral structural variations correctly and there is no limitation on the maximum dip of the interface. We applied the method to real data of a linear array in the Wabash Valley Seismic Zone in the central USA and obtained a crustal structural image across a failed continental rift. We suggest that future passive-source seismic recording experiments for crustal scale imaging use station spacing less than 5 km, and a 2-D array with even smaller station spacing is desired for regions with strong lateral structural variations. With increasing numbers of sensors used in passive-source recording experiments nowadays, our RF-RTM method can be a useful tool for structural imaging on scales ranging from sedimentary basins, crust to lithosphere. [ABSTRACT FROM AUTHOR]

The widespread use of social media has created a valuable but underused source of data for the environmental sciences. We demonstrate the potential for images posted to the website Twitter to capture variability in vegetation phenology across United States National Parks. We process a subset of images posted to Twitter within eight U.S. National Parks, with the aim of understanding the amount of green vegetation in each image. Analysis of the relative greenness of the images show statistically significant seasonal cycles across most National Parks at the 95% confidence level, consistent with springtime green-up and fall senescence. Additionally, these social media-derived greenness indices correlate with monthly mean satellite NDVI (r = 0.62), reinforcing the potential value these data could provide in constraining models and observing regions with limited high quality scientific monitoring. [ABSTRACT FROM AUTHOR]

Muscle myopathies or defects, such as white striping (WS), downgrade the quality of broiler meat, leading to the loss of millions of dollars annually for the United States poultry industry. Visual inspection is currently practiced for assessing myopathies in broiler meat, which is however prone to human evaluation error, labor-intensive, and costly. Imaging technology that employs uniform or diffuse illumination has been investigated for assessing broiler meat quality, but their performance is not always satisfactory, especially for detecting subtle defects with few visual symptoms. This study presents the first proof-of-concept evaluation of the utility of an innovative technique for improved imaging of structures of biological materials, i.e., structured-illumination reflectance imaging (SIRI), for detecting WS conditions in broiler breast meat. Broiler breast fillets with varying degrees of WS severity were imaged by an in-house assembled SIRI system under sinusoidal illumination at different spatial frequencies (0.05–0.40 cycles/mm). The acquired pattern images were demodulated into direct components (DC) and amplitude components (AC) at each spatial frequency. Texture features were extracted from DC and AC images for building classification models using regularized linear discriminant analysis, which were to classify the poultry meat samples into 2, 3, and 4 classes, respectively, according to WS conditions. AC images were found by visual inspection more effective than DC in resolving WS characteristics, depending on the spatial frequency of illumination patterns. Compared to DC, AC yielded the maximum accuracy improvements of 8.33%, 7.93%, and 12.3% in 2-, 3- and 4-class classifications, respectively. Classification models using ranked features yielded similar improvements of up to 9.53% by AC over DC. The SIRI technique is promising for enhanced detection of WS and potentially other defects of broiler breast meat. • Structured-illumination reflectance imaging (SIRI) was used for poultry quality detection. • Amplitude component (AC) images enabled resolving white striping (WS) of broiler breast meat. • Broiler breast fillets were classified by models into 2, 3 and 4 classes according to WS conditions. • SIRI offered substantial improvements of up to 12% in classifying WS-affected poultry samples. [ABSTRACT FROM AUTHOR]

The magnetotelluric component of the EarthScope USArray program has covered over 35% of the continental United States. Resistivity tomography models derived from these data image lithospheric structure and provide constraints on the distribution of fluids and melt within the lithosphere. We present a three-dimensional resistivity model of the northwestern United States which provides new insight into the tectonic assembly of western North America from the Archean to present. Comparison with seismic tomography models reveals regions of correlated and anti-correlated resistivity and velocity that help identify thermal and compositional variations within the lithosphere. Recent (Neogene) tectonic features reflected in the model include the subducting Juan de Fuca-Gorda plate which can be traced beneath the forearc to more than 100 km depth, high lithospheric conductivity along the Snake River Plain, and pronounced lower-crustal and upper-mantle conductivity beneath the Basin and Range. The latter is abruptly terminated to the northwest by the Klamath-Blue Mountains Lineament, which we interpret as an important structure during and since the Mesozoic assembly of the region. This boundary is interpreted to separate hot extended lithosphere from colder, less extended lithosphere. The western edge of Proterozoic North America, as indicated by the Cretaceous initial 87Sr/86Sr = 0.706 contour, is clearly reflected in the resistivity model. We further image an Archean crustal block ("Pend Oreille block") straddling the Washington/Idaho border, which we speculate separated from the Archean Medicine Hat block in the Proterozoic. Finally, in the modern Cascades forearc, the geometry and internal structure of the Eocene Siletz terrane is reflected in the resistivity model. The apparent eastern edge of the Siletz terrane under the Cascades arc suggests that pre-Tertiary rocks fill the Washington and Oregon back-arc. [ABSTRACT FROM AUTHOR]

Pavement markings provide guidance to road travelers and, when not properly removed during construction projects, can lead to fatal accidents. Current state guidelines, however, do not specify a level of acceptable damage and there is no reliable inspection tool available other than on-site individual visual inspections. To address these shortcomings, the main objective of this study was to identify effective temporary marking removal methods on pavements in roadway construction zones. Unlike previous studies, this research tested an extensive number of removal methods on both concrete and asphalt pavements, especially including dry-ice blasting and environmentally friendly non-MeCl chemical stripping. The methods were selected based on a nationwide survey and literature review. In addition, an innovative removal inspection method was developed using image-processing technology to accurately measure the visual damage to the pavement. Among the tested removal methods, the new non-MeCl chemical stripping method was most satisfactory in terms of performance quality. This study of pavement marking removal and inspection can help state transportation agencies significantly improve the current temporary marking removal practices and inspection protocols. [ABSTRACT FROM AUTHOR]

Background Basal cell carcinoma ( BCC) is the most commonly diagnosed cancer in the USA. In this research, we examine four different feature categories used for diagnostic decisions, including patient personal profile (patient age, gender, etc.), general exam (lesion size and location), common dermoscopic (blue-gray ovoids, leaf-structure dirt trails, etc.), and specific dermoscopic lesion (white/pink areas, semitranslucency, etc.). Specific dermoscopic features are more restricted versions of the common dermoscopic features. Methods Combinations of the four feature categories are analyzed over a data set of 700 lesions, with 350 BCCs and 350 benign lesions, for lesion discrimination using neural network-based techniques, including evolving artificial neural networks ( EANNs) and evolving artificial neural network ensembles. Results Experiment results based on 10-fold cross validation for training and testing the different neural network-based techniques yielded an area under the receiver operating characteristic curve as high as 0.981 when all features were combined. The common dermoscopic lesion features generally yielded higher discrimination results than other individual feature categories. Conclusions Experimental results show that combining clinical and image information provides enhanced lesion discrimination capability over either information source separately. This research highlights the potential of data fusion as a model for the diagnostic process. [ABSTRACT FROM AUTHOR]

Abstract: Automated alignment for the National Ignition Facility (NIF) is accomplished using a large-scale parallel control system that directs 192 laser beams along the 300m optical path. The beams are then focused down to a 50μ spot in the middle of the target chamber. The entire process is completed in less than 50min. The alignment system commands 9000 stepping motors for extremely precise adjustment of mirrors and other optics. 41 control loops per beamline perform parallel processing services running on a LINUX cluster to analyze high-resolution images of the beams and their references. This paper describes the status the NIF automatic alignment system and the challenges encountered as NIF development has transitioned from building the laser, to becoming a research project supporting a 24h, 7 days/week laser facility. NIF is now a continuously operated system where performance monitoring is increasingly more critical for operation, maintenance, and commissioning tasks. Equipment wear and the effects of high energy neutrons from fusion experiments are issues which affect alignment efficiency and accuracy. New sensors needing automatic alignment assistance are common. System modifications to improve efficiency and accuracy are prevalent. Handling these evolving alignment and maintenance needs while minimizing the impact on NIF experiment schedule is expected to be an on-going challenge for the planned 30 year operational life of NIF. [Copyright &y& Elsevier]

Characterization of urban land uses is essential for many applications. However, differentiating among thematically-detailed urban land uses (residential, commercial, industrial, institutional, recreational, etc.) over broad areas is challenging, in part because image-based solutions are not ideal for establishing the contextual basis for identifying economic function and use. At present no current United States national-scale mapping exists for urban land uses similar to the classical Anderson Level II classification. This paper describes a product that maps urban land uses, and is linked to and corresponds with the National Land Cover Database (NLCD) 2006. In this product, NLCD urban pixels, in addition to their current imperviousness intensity classification, are assigned one of nine urban use classes based on information drawn from multiple data sources. These sources include detailed infrastructure information, population characteristics, and historical land use. The result is a method for creating a 30 m national-scale grid providing thematically-detailed urban land use information which complements the NLCD. Initial results for 10 major metropolitan areas are provided as an on-line link. Accuracy assessment of initial products yielded an overall accuracy of 81.6 percent. [ABSTRACT FROM AUTHOR]

This paper presents a short overview of the available LEO weather satellites transmitting APT (Automatic Picture Transmission) images. Also, this paper shows the development and setup of an advanced VHF ground station capable of receiving APT images from LEO weather satellites. This ground station will be capable of automatic tracking of selected NOAA satellites by using a steerable high gain multi-purpose VHF X-Quad Yagi antenna. The 137 MHz FM signal from NOAA satellites is received using a wideband Icom receiver and is processed using an external USB soundcard with dedicated software for APT/WEFAX image decoding. The proposed solution will provide low-cost access to APT data from NOAA POES weather satellite fleet. Tests were completed with good results and based on the experimental data, it demonstrates the technical feasibility of this improved receiving system. [ABSTRACT FROM AUTHOR]

Abstract: Theoretical predictions with regard to dust devil (apparent) angular size–frequency distribution are made and critically compared with Mars Exploration Rover (MER) Spirit optical observations. For an idealized horizontal viewing geometry one should expect that the number of dust devils having the apparent angular diameter greater than a given angle α is inversely proportional to α squared. The actual dependency for Spirit dust devils is in between the inverse-squared and simple inverse laws, and close to the latter one for small and moderate angles α. It is emphasized that such a comparison can be considered as a benchmark for completeness and adequateness of dust-devil optical observations and correctness of competing analytical formulations for dust devil size–frequency distribution. [Copyright &y& Elsevier]

Roe is an important product of the Alaska pollock (Theragra chalcogramma) industry. About 31% of the value for all pollock products comes from roe, yet roe is 5% of the weight of the fish. Currently, the size (weight), color, and maturity of the roe are subjectively evaluated. The objective of this study was to develop methods to predict the weight of Alaska pollock roe based on its view area from a camera and to differentiate between single and double roes. One hundred and forty-two pollock roes were picked from a processing line in a Kodiak, AK plant. Each roe was weighed, placed in a light box equipped with a digital video camera, images were taken at two different angles from one side, then turned over and presented at two different angles again (four images for each roe). A reference square of known surface area was placed by the roe. The following equations were used to fit the view area (X) versus weight (Y) data: linear, power, and second-order polynomial. Error rates for the classification of roes by weight decreased significantly when weight prediction equations for single and double roes were developed separately. A “turn angle” method, a “box” method, and a “modified box” method were tested to differentiate single and double roes by image analysis. Machine vision can accurately determine the weight of pollock roe. Practical Application Abstract: An image analysis method to accurately determine if pollock roe is a single or a double was developed. Then view area versus weight correlations were found for single and double roes that reduced incorrect weight classification rates to half that of human graders. [ABSTRACT FROM AUTHOR]

Abstract: This paper describes an implementation of a three-component, three-dimensional plane wave migration method to image P-to-S converted waves that arrive following the direct P waves of teleseismic earthquakes. The programs described assume that the input data have been deconvolved to produce a set of vector impulse response functions with the initial P pulse centered at time 0. This method was implemented as three different application programs called pwstack, pwmig, and gridstacker. pwstack produces local-scale plane wave decompositions using the pseudostation concept. The output of pwstack is an estimate of the plane wave response over a grid of slowness vectors interpolated onto a regular grid in space. The Gaussian smoother that is used for the pseudostation method has the additional benefit of serving as a type of spatial antialiasing filter. The program called pwmig takes the output of pwstack and inverts the plane wave data for radial and transverse scattering potential using an inverse generalized Radon transform. Data can be back-projected using a conventional longitudinal, radial, transverse coordinate system or a novel depth-variable method that computes a transformation matrix linking each vector sample to an equivalent dipping-layer coordinate system at the corresponding scattering point. Because pwmig is a prestack method, a final program, gridstacker, is needed to stack data from multiple events that characterize real data. gridstacker is a general solution for stacking data with different grid geometries and variable data quality. Weighting functions can be defined externally by a user-supplied recipe or one can use a set of robust estimation methods coded in gridstacker. The programs are validated with simulation data produced by recently developed elastic finite-difference code and a model that simulates the edge of a subduction zone. Finally, I show an example of an application of these programs to image the western United States with the USArray. [Copyright &y& Elsevier]

Relative house fly, Musca domestica L., activity at three large dairies in central California was monitored during the peak fly activity period from June to August 2005 by using spot cards, fly tapes, bait traps, and Alsynite traps. Counts for all monitoring methods were significantly related at two of three dairies; with spot card counts significantly related to fly tape counts recorded the same week, and both spot card counts and fly tape counts significantly related to bait trap counts 1-2 wk later. Mean fly counts differed significantly between dairies, but a significant interaction between dairies sampled and monitoring methods used demonstrates that between-dairy comparisons are unwise. Estimate precision was determined by the coefficient of variability (CV) (or SE/mean). Using a CV = 0.15 as a desired level of estimate precision and assuming an integrate pest management (IPM) action threshold near the peak house fly activity measured by each monitoring method, house fly monitoring at a large dairy would require 12 spot cards placed in midafternoon shaded fly resting sites near cattle or seven bait traps placed in open areas near cattle. Software (FlySpotter; http://ucanr.org/sites/FlySpotter/download/) using computer vision technology was developed to count fly spots on a scanned image of a spot card to dramatically reduce time invested in monitoring house flies. Counts provided by the FlySpotter software were highly correlated to visual counts. The use of spot cards for monitoring house flies is recommended for dairy IPM programs. [ABSTRACT FROM AUTHOR]

We clarify that the derivative-free fan-beam reconstruction formula [IEEE Trans. Image Process. 2, 543–547, 1993] only allows exact reconstruction of an object for a circular trajectory or at the origin of the coordinate system for a radially symmetric noncircular trajectory. [ABSTRACT FROM PUBLISHER]

Abstract: The National Cancer Institute has collected a large database of uterine cervix images termed “cervigrams”, for cervical cancer screening research. Tissues of interest within the cervigram, in particular the lesions, are of varying sizes and of complexnon-convex shapes. The tissues possess similar color features and their boundaries are not always clear. The main objective of the current work is to provide a segmentation framework for tissues of interest within the cervix, that can cope with these difficulties in an unsupervised manner and with a minimal number of parameters. The proposed framework transitions from pixels to a set of small coherent regions (superpixels), which are grouped bottom-up into larger, non-convex, perceptually similar regions. The merging process is performed utilizing a new graph-cut criterion termed the normalized-mean cut (NMCut) and an agglomerative clustering framework. Superpixels similarity is computed via a locally scaled similarity measure that combines region and edge information. Segmentation quality is evaluated by measuring the overlap accuracy of the generated segments and tissues that were manually marked by medical experts. Experiments are conducted on two sets of cervigrams and lead to the following set of observations and conclusions: 1) The generated superpixels provide an accurate decomposition of the different tissues; 2) The local scaling process improves the clustering results; 3) The influence of different graph-cut criterions on the segmentation accuracy is evaluated and the NMCut criterion is shown to provide the best results; 4) A comparison between several modifications to the agglomerative clustering process is conducted. The results are shown to be strongly influenced by the merging procedure; 5) The agglomerative clustering framework is shown to outperform a state-of-the-art spectral clustering algorithm. [Copyright &y& Elsevier]

In the late 1990s, image processing was first applied to steel bridge coating assessment in the United States. Yet, there has not been any robust method that could solve nonuniform illumination problems to date. In this regard, this paper aims to develop an approach that handles nonuniform illumination in rust image recognition. The proposed approach also contributes to rust intensity recognition and process automation. This paper adopts the a*b* color configuration (of the L*a*b* color space) which is shown to be the best coordinate system for rust recognition by the proposed method. Different rust colors are regarded as relating to different degrees of rust intensity. Due to the difficulty of defining light rust colors, the adaptive-network-based fuzzy inference system (ANFIS) is used as the framework for the box-and-ellipse-based ANFIS (BE-ANFIS). Illumination adjustment is used to overcome the nonuniform illumination problem. The proposed BE-ANFIS is trained using 120 rust images of the size of 256×256 pixels. The performance of BE-ANFIS is compared to the fuzzy C-means (FCM) method, one of the most popular image recognition methods, and the adaptive ellipse approach (AEA) which is the writers’ previous approach. Based on testing 35 rust images, the results show that BE-ANFIS can better recognize rust intensity (i.e., distinguishing light and dark rusts) than FCM and AEA. [ABSTRACT FROM AUTHOR]

MR urography (MRU) is an emerging technique particularly useful in paediatric uroradiology. The most common indication is the investigation of hydronephrosis. Combined static and dynamic contrast-enhanced MRU (DCE-MRU) provides both morphological and functional information in a single examination. However, specific post-processing must be performed and to our knowledge, dedicated software is not available in conventional workstations. Investigators involved in MRU classically use homemade software that is not freely accessible. For these reasons, we have developed a software program that is freely downloadable on the National Institute of Health (NIH) website. We report and describe in this study the features of this software program. [ABSTRACT FROM AUTHOR]

A maximum a posteriori (MAP) technique is developed to identify solar features in cotemporal and cospatial images of line-of-sight magnetic flux, continuum intensity, and equivalent width observed with the NASA/National Solar Observatory (NSO) Spectromagnetograph (SPM). The technique facilitates human understanding of patterns in large data sets and enables systematic studies of feature characteristics for comparison with models and observations of long-term solar activity and variability. The method uses Bayes’ rule to compute the posterior probability of any feature segmentation of a trio of observed images from per-pixel, class-conditional probabilities derived from independently-segmented training images. Simulated annealing is used to find the most likely segmentation. New algorithms for computing class-conditional probabilities from three-dimensional Gaussian mixture models and interpolated histogram densities are described and compared. A new extension to the spatial smoothing in the Bayesian prior model is introduced, which can incorporate a spatial dependence such as center-to-limb variation. How the spatial scale of training segmentations affects the results is discussed, and a new method for statistical separation of quiet Sun and quiet network is presented. [ABSTRACT FROM AUTHOR]

Abstract: Green infrastructure is a popular framework for conservation planning. The main elements of green infrastructure are hubs and links. Hubs tend to be large areas of ‘natural’ vegetation and links tend to be linear features (e.g., streams) that connect hubs. Within the United States, green infrastructure projects can be characterized as: (1) reliant on classical geographic information system (GIS) techniques (e.g., overlay, buffering) for mapping; (2), mainly implemented by states and local jurisdictions; and (3) static assessments that do not routinely incorporate information on land-cover change. We introduce morphological spatial pattern analysis (MSPA) as a complementary way to map green infrastructure, extend the geographic scope to the conterminous United States, and incorporate land-cover change information. MSPA applies a series of image processing routines to a raster land-cover map to identify hubs, links, and related structural classes of land cover. We identified approximately 4000 large networks (>100 hubs) within the conterminous United States, of which approximately 10% crossed state boundaries. We also identified a net loss of up to 3.59 million ha of links and 1.72 million ha of hubs between 1992 and 2001. Our national assessment provides a backbone that states could use to coordinate their green infrastructure projects, and our incorporation of change illustrates the importance of land-cover dynamics for green infrastructure planning and assessment. [Copyright &y& Elsevier]

Immersive virtual environments with life-like interaction capabilities have very demanding requirements including high-precision motion capture and high-processing speed. These issues raise many challenges for computer vision-based motion estimation algorithms. In this study, we consider the problem of hand tracking using multiple cameras and estimating its 3D global pose (i.e., position and orientation of the palm). Our interest is in developing an accurate and robust algorithm to be employed in an immersive virtual training environment, called “Virtual GloveboX” (VGX) (Twombly et al. in J Syst Cybern Inf 2:30–34, 2005), which is currently under development at NASA Ames. In this context, we present a marker-based, hand tracking and 3D global pose estimation algorithm that operates in a controlled, multi-camera, environment built to track the user’s hand inside VGX. The key idea of the proposed algorithm is tracking the 3D position and orientation of an elliptical marker placed on the dorsal part of the hand using model-based tracking approaches and active camera selection. It should be noted that, the use of markers is well justified in the context of our application since VGX naturally allows for the use of gloves without disrupting the fidelity of the interaction. Our experimental results and comparisons illustrate that the proposed approach is more accurate and robust than related approaches. A byproduct of our multi-camera ellipse tracking algorithm is that, with only minor modifications, the same algorithm can be used to automatically re-calibrate (i.e., fine-tune) the extrinsic parameters of a multi-camera system leading to more accurate pose estimates. [ABSTRACT FROM AUTHOR]

In this paper we present a method for the recognition of hand- written digits and a practical implementation of this method for real-time recognition. A theoretical framework for the neural networks used to classify the handwritten digits is also presented. The classification task is performed using a Convolutional Neural Net- work (CNN). CNN is a special type of multi-layer neural network, being trained with an optimized version of the back-propagation learning algorithm. CNN is designed to recognize visual patterns directly from pixel images with minimal preprocessing, being capable to recognize patterns with extreme variability (such as handwritten characters), and with robustness to distortions and simple geometric transformations. The main contributions of this paper are related to the original methods for increasing the efficiency of the learning algorithm by preprocessing the images before the learning process and a method for increasing the precision and performance for real-time applications, by removing the non useful information from the background. By combining these strategies we have obtained an accuracy of 96.76%, using as training set the NIST (National Institute of Standards and Technology) database. [ABSTRACT FROM AUTHOR]

Abstract: NASA''s Cassini project has conducted a series of innovative “Scientist for a Day” outreach sessions with great success. Individual students and classrooms are presented, via the Cassini web site, with a selection of images that could be taken by the Cassini spacecraft in orbit around Saturn. Student essays are submitted indicating their preferred image, along with a detailed description of why that image was selected, often drawing heavily from material found elsewhere in the Cassini web resources. Winners from different age categories are chosen, and the selected images are sequenced. Very soon after the images reach the ground, winning classrooms are invited to participate via tele- and video-con with key Cassini scientists and engineers, where interactive discussions take place. This program has been very successful and is considered a key outreach highlight of Cassini and JPL in general. It reaches a significant number of students, involves them directly in image planning, and engages them in a highly interactive and inspirational environment requiring detailed study and creative thought. [Copyright &y& Elsevier]

Purpose - This paper aims to present an automated system that measures the volume of excavated soil in earthmoving operations. It focuses primarily on presenting the use of the radio frequency identification - real time location technology and image analysis techniques in isolation of stockpiles from their noisy backgrounds, extraction of their geometrical attributes and determination of their location. Design/methodology/approach - An extensive literature review of image analysis techniques is performed to identify the ones most applicable in developing the proposed automated system. A set of techniques are selected and experimented to evaluate their effectiveness in the proposed application. A review of the state-of-the-art distance measurement technologies is also conducted to identify the most suitable system to be used in the proposed system. Findings - A set of image analysis techniques that positively contribute to the accuracy of the proposed system is identified. Geometrical attributes of stockpiles are automatically extracted for volume measurement purposes. The use of an affordable and reliable automated distance measurement tool is also suggested. Research limitations/implications - Although the proposed methodology is believed to be applicable to most configurations of stockpiles, it was tested on conical ones only. Originality/value - The current practice of measuring the volume of excavated soil is time consuming and costly. Furthermore, it does not facilitate monitoring of excavation activities on close time intervals. The proposed system overcomes the problems associated with the current practice and provides an automated strategy that can be easily used by field personnel and/or home office management staff to closely monitor the progress of their earthmoving operations without physical human intervention. [ABSTRACT FROM AUTHOR]

The article presents an overview of the work and research being done by the organizations in the area of metadata creation related to cataloging and preservation of moving image content in the U.S. It highlights several organizations involved in creating metadata for moving images including the Motion Picture Experts Group (MPEG) and the Corporation for Public Broadcasting (CPB). Problems and suggestions encountered in the field of cataloging and preservation of moving images is also outlined.

The presence of clouds and their shadows in remotely sensed images limits their potential uses for extracting information. The commonly used methods for replacing clouded pixels by land cover reflection estimates usually yield poor results if the images being combined exhibit radical differences in target radiance due, for example, to large date separation and high temporal variability. This study focuses on introducing geostatistical techniques for interpolating the DN values of clouded pixels in multispectral remotely sensed images using traditional ordinary cokriging and standardized ordinary cokriging. Two case studies were conducted in this study. The first case study shows that the methods work well for the small clouds in a heterogeneous landscape even when the images being combined show high temporal variability. Although the basic spatial structure in large size clouds can be captured, image interpolation-related artefacts such as smoothing effects are visually apparent in a heterogeneous landscape. The second case study indicates that the cokriging methods work better in homogenous regions such as the dominantly agricultural areas in United States Midwest. Various statistics including both global statistics and local statistics are employed to confirm the reliability of the methods. [ABSTRACT FROM AUTHOR]

One aspect of grading table eggs is shell checks or cracks. Currently, USDA voluntary regulations require that humans grade a representative sample of all eggs processed. However, as processing plants and packing facilities continue to increase their volume and throughput, human graders are having difficulty matching the pace of the machines. Additionally, some plants also have a problem with micro-cracks that the graders often miss because they are very small and hard to see immediately post-processing but grow and become readily apparent before they reach market. An imaging system was developed to help the grader detect these small micro-cracks. The imaging system utilized one image captured at atmospheric pressure and a second at a slight negative pressure to enhance the crack and make detection much easier. A simple image processing algorithm was then applied to the ratio of these two images, and the resulting image, containing both cracked and/or intact eggs were color-coded to simplify identification. The imaging system was capable of imaging 15 eggs ill about 0.75 s, and the algorithm processing took about another 10 s. These times could easily be reduced with a compiled program specifically written for the application. In analyzing 1000 eggs, the system was 99.6% accurate overall with only 0.3% false positives, compared to 94.2% accurate overall for the human graders with 1.2% false positives. An international patent on the system has been filed, and further automation of the system is needed. [ABSTRACT FROM AUTHOR]

• Thirty different methods to estimate UAV-derived percent green cover were compared and discussed according to their ease of use and relationship with ground-level percent green cover estimations. • An HSV color space-based method was introduced to determine turfgrass percent green cover using UAV. • RGB-composites images are sufficient to determine turfgrass percent green cover. • The concept of percent green cover (green pixels only) is different from turfgrass coverage which may include green and non-green grass pixels. Turfgrass is an important urban crop in the United States. Determining the percent green cover (PGC) to assess turfgrass quality/health and the rate of establishment is a crucial parameter for evaluating different species and experimental lines within species. However, evaluating the PGC of individual plots within large breeding nurseries in a conventional way, either visually or through digital image analysis is a time-consuming and laborious process. In the present study, we used the unmanned aerial vehicle (UAV) with multispectral and RGB sensors to estimate PGC during turfgrass establishment. We evaluated thirty approaches with different levels of complexity based on vegetation indices, supervised and unsupervised machine learning classification methods, and image processing methods for high-throughput turfgrass PGC estimation. An HSV (Hue-Saturation-Value) color space-based green pixel identification (GPI) method was introduced for the first time for estimating UAV derived PGC (UAV PGC). The results indicate that the GPI achieved the highest coefficient of determination, 0.86–0.96, with lowest mean absolute error when compared to ground percent green cover (GroundPGC). Overall, UAV-derived RGB image-based support vector machine methods were in agreement with GroundPGC (R2 = 0.88–0.95). This suggests that UAV-derived RGB images are adequate in accurately determining percent green cover (green vegetation within an experimental plot); however, multispectral images might offer a solution to determine turfgrass coverage (green and non-green vegetation within an experimental plot) during turfgrass establishment to account for non-green vegetation which is not captured by RGB (visible light spectrum) based estimation of PGC. [ABSTRACT FROM AUTHOR]

The article reports on the Stennis Space Center's (SSC) Applied Science and Technology Project Office (ASTPO) and its use of the NASA's remote sensing, image processing, and oceanography to address issues related to Coastal Gulf of Mexico. ASTPO is a part of the NASA Applied Sciences Program which aims to demonstrate the utility of NASA Earth Science data, models, tools and techniques. It focuses on remote sensing and other tools of NASA for Gulf of Mexico applications. It also cooperates with the Gulf of Mexico Alliance (GOMA) and on the National Oceanic and Atmospheric Administration (NOAA) to ensure that the issues of the Gulf of Mexico community are accurately and appropriately identified.

The U.S. Landsat satellite series provide the longest dedicated land remote sensing data record with a balance between requirements for localized high spatial resolution studies and global monitoring. As with any other optical wavelength satellite sensor, cloud contamination greatly compromises image usability for land surface studies. Additionally, selective scene acquisition due to payload, ground station and mission cost constraints further reduces Landsat image availability. Since the 1999 launch of the Landsat Enhanced Thematic Mapper Plus (ETM+) a Long-term Acquisition Plan (LTAP) has been used to anticipate user requests with the goal of annually refreshing a global daytime archive of cloud-free ETM+ data. This research evaluates the availability of cloud-free Landsat ETM+ data over the conterminous U.S. and globally using 3 years of ETM+ cloud fraction metadata archived by the U.S. Landsat project. Landsat application requirements including obtaining at least one cloud-free observation in a year, a season, and two different seasons, or at least a pair of cloud-free observations occurring no more than 16, 32, 48, 64, and 80 days apart within a year and season are considered. Probabilistic analyses indicate that over the conterminous U.S., land applications requiring at least one cloud-free observation in a year, a season, two different seasons, or at least two cloud-free observations occurring within any period of the year, are on average largely unaffected by cloud cover, except for certain Winter applications and cloudy scenes near the U.S.–Canada border and the Great Lakes. Cloud becomes a constraint when at least two cloud-free observations are required from the same season over the conterminous U.S., especially when the separation between observations is restricted to short time intervals. Global applications requiring at least one cloud-free observation in a season, in two different seasons, and applications requiring at least two cloud-free observations in a year, are all severely affected by cloud and data availability constraints; and globally it is generally not practical to consider land applications that require at least two cloud-free observations in any season. Globally, only land applications requiring at least one cloud-free observation per year are largely unaffected by cloud cover and the reduced global ETM+ data availability. These results are specific only to the U.S. Landsat ETM+ archive; they suggest the need for an increased global Landsat acquisition rate for the current and future Landsat missions and/or the development of new approaches to mitigating cloud contamination in the U.S. global Landsat ETM+ archive. [Copyright &y& Elsevier]

Over 1 billion biodiversity collection specimens ranging from fungi to fish to fossils are housed in more than 1,600 natural history collections across the United States. The digitization of these specimens has risen significantly within the last few decades and this is only likely to increase, as the use of digitized data gains more importance every day. Numerous experiments with automated image analysis have proven the practicality and usefulness of digitized biodiversity images by computational techniques such as neural networks and image processing. However, most of the computational techniques to analyze images of biodiversity collection specimens require a good curation of this data. One of the challenges in curating multimedia data of biodiversity collection specimens is the quality of the multimedia objects--in our case, two dimensional images. To tackle the image quality problem, multimedia needs to be captured in a specific format and presented with appropriate descriptors. In this study we present an analysis of two image repositories each consisting of 2D images of fish specimens from several institutions--the Integrated Digitized Biocollections (iDigBio) and the Great Lakes Invasives Network (GLIN). Approximately 70 thousand images have been processed from the GLIN repository and 450 thousand images have been processed from the iDigBio repository and their suitability assessed for use in neural network-based species identification and trait extraction applications. Our findings showed that images that came from the GLIN dataset were more successful for image processing and machine learning purposes. Almost 40% of the species have been represented with less than 10 images while only 20% have more than 100 images per species. We identified and captured 20 metadata descriptors that define quality and usability of the image. According to the captured metadata information, 70% of the GLIN dataset images were found to be useful for further analysis according to the overall image quality score. Quality issues with the remaining images included: curved specimens, non-fish objects in the images such as tags, labels and rocks that obstructed the view of the specimen; color, focus and brightness issues; folded or overlapping parts as well as missing parts. We used both the web interface and the API (Application Programming Interface) for downloading images from iDigBio. We searched for all fish genera, families and classes in three different searches with the images-only option selected. Then we combined all of the search results and removed duplicates. Our search on the iDigBio database for fish taxa returned approximately 450 thousand records with images. We narrowed this down to 90 thousand fish images aided by the multimedia metadata with the downloaded search results, excluding some non-fish images, fossil samples, X-ray and CT (computed tomography) scans and several others. Only 44% of these 90 thousand images were found to be suitable for further analysis. In this study, we discovered some of the limitations of biodiversity image datasets and built an infrastructure for assessing the quality of biodiversity images for neural network analysis. Our experience with the fish images gathered from two different image repositories has enabled describing image quality metadata features. With the help of these metadata descriptors, one can simply create a dataset for a desired image quality for the purpose of analysis. Likewise, the availability of the metadata descriptors will help advance our understanding of quality issues, while helping data technicians, curators and the other digitization staff be more aware of multimedia. [ABSTRACT FROM AUTHOR]

This research presents a segmentation-based, image processing method to automate the extraction of tidal datum referenced shorelines from airborne light detection and ranging (LIDAR) data. The method first segments a LIDAR digital elevation model (DEM) into a binary image, consisting of land and water pixels, by intersecting the LIDAR DEM and the tidal datum surface. A chain of image processing algorithms, including region grouping and labeling, two passes of image region scanning, a mathematical morphology operation, line tracing, and vectorization, is sequentially applied to the segmented binary image. Our applications to the upper Texas gulf coast show that the method is efficient, accurate, objective, and replicable. Spatially detailed shorelines can be derived from the LIDAR data with minimal human intervention. The method is a substantial technical improvement over standard cross-shore profile and contouring methods. We also examine and quantify the effects of vertical measurement error of the LIDAR system and the uncertainty in tidal datum determination on the shoreline extraction process using the Monte Carlo simulation technique. The Monte Carlo technique allows confidence intervals and summary error statistics to be calculated for each section of the extracted shoreline. Our analysis suggests that the horizontal position of the MHW shoreline derived from the LIDAR data is accurate within 4.5 m at the 95% confidence level. [ABSTRACT FROM AUTHOR]

The article discusses the techniques used in locating a specific type of image or file in the U.S. Because of the existence of different types of images which varies in color, type and description, it is difficult for user to locate a specific image that he wants. Previously, the Library of Congress uses Thesaurus for Graphic Materials I and Thesaurus for Graphic Materials II in cataloguing and retrieving images. With the onset of new technology, collaborative tagging is used in assigning keywords to textual documents and image resources. Such a tool is used by sharing sites such as Flick because it allows massive image storage and web diffusion wherein users upload their own images and index them using their own tags.

A method for simulating the effects of visual adaptation using image processing techniques was developed, and then tested for accuracy in predicting the reductions in visual performance that occur with adaptation. The simulation process entailed transforming digital photographs of realistic exterior scenes into the spatial frequency domain, filtering the transformed data using contrast sensitivity functions for different adaptation levels, transforming the filtered data back into the luminance domain, and producing a new digital image of the scene. Two experiments were conducted that compared performance on a realistic visual detection task when the subjects were adapted to different luminances, to performance on the same task when subjects' adaptation level was held constant while they viewed the simulated adaptation effects. Results showed that the simulations produced decreases in performance in the same direction as actual adaptation effects, but the effects were of smaller magnitude. A third experiment was conducted to better quantify the differences between the simulated and actual adaptation conditions; this experiment utilized a standardized visual test target, the United States Air Force (USAF) 1951 test pattern. [ABSTRACT FROM AUTHOR]

This article is my personal account on the work at Stanford on Stanley, the winning robot in the DARPA Grand Challenge. Between July 2004 and October 2005, my then-postdoc Michael Montemerlo and I led a team of students, engineers, and professionals with the single vision of claiming one of the most prestigious trophies in the field of robotics: the DARPA Grand Challenge (DARPA 2004). The Grand Challenge, organized by the U.S. government, was unprecedented in the nation's history. It was the first time that the U.S. Congress had appropriated a cash price for advancing technological innovation. My team won this prize, competing with some 194 other teams. Stanley was the fastest of five robotic vehicles that, on October 8, 2005, successfully navigated a 131.6-mile-long course through California's Mojave Desert. This essay is not about the technology behind our success; for that I refer the interested reader to recent articles on the technical aspects of Stanley (Dahlkamp et al. 2006; Montemerlo et al. 2006; Stavens and Thrun 2006; Thrun, Montemerlo, and Aron 2006; Thrun et al. 2006). Instead, this is my personal story of leading the Stanford Racing Team. It is the story of a team of people who built an autonomous robot in record time. It is also a success story for the field of artificial intelligence, as Stanley used some state of the art AI methods in areas such as probabilistic inference, machine learning, and computer vision. Of course, it is also the story of a step towards a technology that, one day, might fundamentally change our lives. [ABSTRACT FROM AUTHOR]

The article presents information on the X-ray scanners which help in detecting illegitimate materials because of their ability to identify any substance such as explosives at the molecular and atomic levels. It is also because of its less expensive cost that the technology is used extensively. However, the R-L technology created at the Virginia Tech University is considered the first multisensing technology used to detect explosives.

Counterfeit is one of the major crimes in financial market. The objective of this project was to develop a bill-counterfeit prevention (BCP) algorithm used in multi-function peripherals (MFPs) that automatically detects reproduction of U.S. banknotes. Our BCP algorithm included: (1) complex background removal; (2) banknote location identification; (3) orientation correction; (4) area measurement; (5) front/back identification; (6) treasury seal analysis; (7) upside/down correction; (8) signature analysis; and (9) currency value analysis. To test our BCP algorithm, a database of 132 digital images with various amounts, i.e., $1, $5, $10, $20, $50, and $100, of U.S. banknotes were collected. In this preliminary study, our technique has achieved a detection rate of 78.3% in U.S. banknotes. In summary, our technique can be potentially used in detecting banknote reproduction such that extra enforcement measures (e.g., no scanning or printing, watermark & tracing system, etc.) can be achieved. Future investigation will also include hardware implementation adopting the BCP algorithm used in MFP copiers, scanners and/or printers. [ABSTRACT FROM AUTHOR]

Prostatic adenocarcinoma is the most commonly occurring cancer among men in the United States, second only to skin cancer. Currently, the only definitive method to ascertain the presence of prostatic cancer is by trans-rectal ultrasound (TRUS) directed biopsy. Owing to the poor image quality of ultrasound, the accuracy of TRUS is only 20%-25%. High-resolution magnetic resonance imaging (MRI) has been shown to have a higher accuracy of prostate cancer detection compared to ultrasound. Consequently, several researchers have been exploring the use of high resolution MRI in performing prostate biopsies. Visual detection of prostate cancer, however, continues to be difficult owing to its apparent lack of shape, and the fact that several malignant and benign structures have overlapping intensity and texture characteristics. In this paper, we present a fully automated computer-aided detection (CAD) system for detecting prostatic adenocarcinoma from 4 Tesla ex vivo magnetic resonance (MR) imagery of the prostate. After the acquired MR images have been corrected for background inhomogeneity and nonstandardness, novel three-dimensional (3-D) texture features are extracted from the 3-D MRI scene. A Bayesian classifier then assigns each image voxel a "likelihood" of malignancy for each feature independently. The "likelihood" images generated in this fashion are then combined using an optimally weighted feature combination scheme. Quantitative evaluation was performed by comparing the CAD results with the manually ascertained ground truth for the tumor on the MRI. The tumor labels on the MR slices were determined manually by an expert by visually registering the MR slices with the corresponding regions on the histology slices. We evaluated our CAD system on a total of 33 two-dimensional (2-D) MR slices from five different 3-D MR prostate studies. Five slices from two different glands were used for training. Our feature combination scheme was found to outperform the individual texture features, and also other popularly used feature combination methods, including AdaBoost, ensemble averaging, and majority voting. Further, in several instances our CAD system performed better than the experts in terms of accuracy, the expert segmentations being determined solely from visual inspection of the MRI data. In addition, the intrasystem variability (changes in CAD accuracy with changes in values of system parameters) was significantly lower than the corresponding intraobserver and interobserver variability. CAD performance was found to be very similar for different training sets. Future work will focus on extending the methodology to guide high-resolution MRI-assisted in vivo prostate biopsies. [ABSTRACT FROM AUTHOR]

This article addresses a very particular domain in the area of peritextual studies- that of the illustrated covers which clothe some photographic books. It seeks first to analyse how this type of peritext is distinguished from peritext in general. Next, it attempts to understand the peritextual differences between the major sub- categories: `photographs in books', on the one hand, and `photographic books', on the other. It concludes with an analysis of one particularly well-known example- Robert Frank's The Americans. [ABSTRACT FROM AUTHOR]

Explores the use of raster-oriented technology to support public Internet applications that manage large raster imagery collections in the United States. Use of partial pyramiding to allow databases to update only the changing areas of an imagery collection in response to the insertion of new imagery; Use of software to enhance the performance of analytical operations; Raster data management.