Light field cameras are an emerging imaging device for acquiring 3-D information of a scene by capturing a field of light rays traveling in space. As light field cameras become portable, hand-held, and affordable, their potential as a 3-D measurement instrument is growing in many applications, including 3-D evidence imaging in crime scene investigations. We evaluated the lateral resolution of commercially available light field cameras, which is one of the fundamental specifications of imaging instruments. For the evaluation of the camera’s lateral resolution, we imaged Siemens stars under various imaging configurations and experimental conditions, including changes in distance between the camera and the resolution target plate, illumination, zoom level, location of the Siemens star in the camera’s field-of-view, and cameras of the same model. The analysis results from a full factorial experiment showed that (i) when a lower zoom level of the camera was used, the lateral resolution tended not to be affected by distance; however, when a higher zoom level was used, it tended to decrease significantly with respect to the distance, (ii) the center region of the camera’s field-of-view provided a better lateral resolution than the peripheral regions, (iii) a higher zoom level yielded a higher lateral resolution, (iv) the two cameras of the same model used in the study did not show a significant difference in the lateral resolution, and (v) changes in illumination did not affect the lateral resolution of the cameras.
Induced pluripotent stem cells (iPSCs) are reprogrammed cells that can have heterogeneous biological potential. Quality assurance metrics of reprogrammed iPSCs will be critical to ensure reliable use in cell therapies and personalized diagnostic tests. We present a quantitative phase imaging (QPI) workflow which includes acquisition, processing, and stitching multiple adjacent image tiles across a large field of view (LFOV) of a culture vessel.
Low magnification image tiles (10x) were acquired with a Phasics SID4BIO camera on a Zeiss microscope. iPSC cultures were maintained using a custom stage incubator on an automated stage. We implement an image acquisition strategy that compensates for non-flat illumination wavefronts to enable imaging of an entire well plate, including the meniscus region normally obscured in Zernike phase contrast imaging. Polynomial fitting and background mode correction was implemented to enable comparability and stitching between multiple tiles. LFOV imaging of reference materials indicated that image acquisition and processing strategies did not affect quantitative phase measurements across the LFOV. Analysis of iPSC colony images demonstrated mass doubling time was significantly different than area doubling time.
These measurements were benchmarked with prototype microsphere beads and etched-glass gratings with specified spatial dimensions designed to be QPI reference materials with optical pathlength shifts suitable for cell microscopy.
This QPI workflow and the use of reference materials can provide non-destructive traceable imaging method for novel iPSC heterogeneity characterization.
This work addresses the problem of automatic classification and labeling of 19th- and 20th-century quilts from
photographs. The photographs are classified according to the quilt patterns into crazy and non - crazy categories. Based
on the classification labels, humanists try to understand the distinct characteristics of an individual quilt-maker or
relevant quilt-making groups in terms of their choices of pattern selection, color choices, layout, and original deviations
from traditional patterns. While manual assignment of crazy and non-crazy labels can be achieved by visual inspection,
there does not currently exist a clear definition of the level of crazy-ness, nor an automated method for classifying
patterns as crazy and non-crazy.
We approach the problem by modeling the level of crazy-ness by the distribution of clusters of color-homogeneous
connected image segments of similar shapes. First, we extract signatures (a set of features) of quilt images that represent
our model of crazy-ness. Next, we use a supervised classification method, such as the Support Vector Machine (SVM)
with the radial basis function, to train and test the SVM model. Finally, the SVM model is optimized using N-fold cross
validation and the classification accuracy is reported over a set of 39 quilt images.
This paper addresses the problem of automating analyses of historical maps. The problem is motivated by the lack of
accuracy and consistency in the current comparison process of geographical objects found in historical maps by visual
inspections. The objective of our work is to compare shape characteristics of the Great Lakes region in a dataset of
approximately 40 French and British historical maps created in the 17th through the 19th centuries. Our approach
decomposes the visual inspection into steps such as object segmentation, spatial scale calibration, extraction of calibrated
object descriptors and comparison of descriptors over time and multiple cartographer houses. The automation of object
segmentation is achieved by template shape-based segmentation using the Hu moments as shape descriptors and ball-based
region growing. The automation of spatial calibration is accomplished by detection and classification of lines
along map borders and by mapping striped boundaries intersected by latitude and longitude lines into degrees of arc
length. Thus, shape characteristics of segmentation results in pixels can be converted to geographical units, for example,
an area of a lake in square miles. We report experimental evaluations of automation accuracy based on comparison with
manual segmentation results, as well as the knowledge obtained from the area comparisons.
We have investigated the computational scalability of image pyramid building needed for dissemination of very large
image data. The sources of large images include high resolution microscopes and telescopes, remote sensing and
airborne imaging, and high resolution scanners. The term 'large' is understood from a user perspective which means
either larger than a display size or larger than a memory/disk to hold the image data. The application drivers for our
work are digitization projects such as the Lincoln Papers project (each image scan is about 100-150MB or about
5000x8000 pixels with the total number to be around 200,000) and the UIUC library scanning project for historical maps
from 17th and 18th century (smaller number but larger images). The goal of our work is understand computational
scalability of the web-based dissemination using image pyramids for these large image scans, as well as the preservation
aspects of the data. We report our computational benchmarks for (a) building image pyramids to be disseminated using
the Microsoft Seadragon library, (b) a computation execution approach using hyper-threading to generate image
pyramids and to utilize the underlying hardware, and (c) an image pyramid preservation approach using various hard
drive configurations of Redundant Array of Independent Disks (RAID) drives for input/output operations. The benchmarks are obtained with a map (334.61 MB, JPEG format, 17591x15014 pixels). The discussion combines the speed and preservation objectives.
This paper addresses the problem of stitching Giga Pixel images from airborne images acquired over multiple flight
paths of Costa Rica in 2005. The set of input images contains about 10,158 images, each of size around 4072x4072
pixels, with very coarse georeferencing information (latitude and longitude of each image). Given the spatial coverage
and resolution of the input images, the final stitched color image is 294,847 by 269,195 pixels (79.3 Giga Pixels) and
corresponds to 238.2 GigaBytes. An assembly of such large images requires either hardware with large shared memory
or algorithms using disk access in tandem with available RAM providing data for local image operation. In addition to
I/O operations, the computations needed to stitch together image tiles involve at least one image transformation and
multiple comparisons to place the pixels into a pyramid representation for fast dissemination. The motivation of our
work is to explore the utilization of multiple hardware architectures (e.g., multicore servers, computer clusters) and
parallel computing to minimize the time needed to stitch Giga pixel images.
Our approach is to utilize the coarse georeferencing information for initial image grouping followed by an intensitybased
stitching of groups of images. This group-based stitching is highly parallelizable. The stitching process results in
image patches that can be cropped to fit a tile of an image pyramid frequently used as a data structure for fast image
access and retrieval. We report our experimental results obtained when stitching a four Giga Pixel image from the input
images at one fourth of their original spatial resolution using a single core on our eight core server and our preliminary
results for the entire 79.3 Gigapixel image obtained using a 120 core computer cluster.
We addressed the problem of finding salient characteristics of artists from two-dimensional (2D) images of historical
artifacts. Given a set of 2D images of historical artifacts by known authors, we discovered what salient characteristics
made an artist different from others, and then enabled statistical learning about individual and collective authorship. The
objective of this effort was to learn what would be unique about the style of each artist, and to provide the quantitative
results about salient characteristic. We accomplished this by exploring a large search space of low level image
descriptors. The motivation behind our framework was to assist humanists in discovering salient characteristics by
automated exploration of the key image descriptors. By employing our framework we had not only saved time of art
historians but also provided quantitative measures for incorporating their personal judgments and bridging the semantic
gap in image understanding. We applied the framework implementation to the face illustrations in Froissart's Chronicles
drawn by two anonymous authors. We reported the salient characteristics to be (HSV, histogram, k-nearest neighbor)
among the 55 triples considered with 5-fold validations. These low level characteristics were confirmed by the experts to
correspond semantically to the face skin colors.
This paper addresses the problem of robust and automated synchronization of multiple audio and video signals. The
input signals are from a set of independent multimedia recordings coming from several camcorders and microphones.
While the camcorders are static, the microphones are mobile as they are attached to people. The motivation for
synchronization of all signals is to support studies and understanding of human interaction in a decision support
environment that have been limited so far due to the difficulties in automated processing of any observations during the
decision making sessions. The application of our work is to environments supporting decisions. The data sets for this
work have been acquired during training exercises of response teams, rescue workers, and fire fighters at multiple
The developed synchronization methodology for a set of independent multimedia recordings is based on introducing
aural and visual landmarks with a bell and room light switches. Our approach to synchronization is based on detecting
the landmarks in audio and video signals per camcorder and per microphone, and then fusing the results to increase
robustness and accuracy of the synchronization. We report synchronization results that demonstrate accuracy of
synchronization based on video and audio.
We propose a methodology for making optimal registration decisions during 3D volume reconstruction in terms of (a) anticipated accuracy of aligned images, (b) uncertainty of obtained results during the registration process, (c) algorithmic repeatability of alignment procedure, and (d) computational requirements. We researched and developed a web-enabled, web services based, data-driven, registration decision support system. The registration decisions include (1) image spatial size (image sub-area or entire image), (2) transformation model (e.g., rigid, affine or elastic), (3) invariant registration feature (intensity, morphology or a sequential combination of the two), (4) automation level (manual, semi-automated, or fully-automated), (5) evaluations of registration results (multiple metrics and methods for establishing ground truth), and (6) assessment of resources (computational resources and human expertise, geographically local or distributed). Our goal is to provide mechanisms for evaluating the tradeoffs of each registration decision in terms of the aforementioned impacts. First, we present a medical registration methodology for making registration decisions that lead to registration results with well-understood accuracy, uncertainty, consistency and computational complexity characteristics. Second, we have built software tools that enable geographically distributed researchers to optimize their data-driven registration decisions by using web services and supercomputing resources. The support developed for registration decisions about 3D volume reconstruction is available to the general community with the access to the NCSA supercomputing resources. We illustrate performance by considering 3D volume reconstruction of blood vessels in histological sections of uveal melanoma from serial fluorescent labeled paraffin sections labeled with antibodies to CD34 and laminin. The specimens are studied by fluorescence confocal laser scanning microscopy (CLSM) images.
We address the problem of automated image alignment for 3D volume reconstruction from stacks of fluorescent confocal laser scanning microscope (CLSM) imagery acquired at multiple confocal depths, from a sequence of consecutive slides. We focus on automated image alignment based on centroid and area shape features by solving feature correspondence problem, also known as Procrustes problem, in highly deformable and ill-conditioned feature space. In result, we compare image alignment accuracy of a fully automated method with registration accuracy achieved by human subjects using a manual alignment method. Our work demonstrates significant benefits of automation for 3D volume reconstruction in terms of accuracy, consistency, and performance time. We also outline the limitations of fully automated and manual 3D volume reconstruction system.
In this paper we present the utilization of high-spectral resolution imagery for improving low-spectral resolution imagery. In our analysis, we assume that an acquisition of high spectral resolution images provides more accurate spectral predictions of low spectral resolution images than a direct acquisition of low spectral resolution images. We illustrate the advantages by focusing on a specific case of images acquired by a hyperspectral (HS) camera and a color (red, green, and blue or RGB) camera. First, we identify two directions for utilization of HS images, such as (a) evaluation and calibration of RGB colors acquired from commercial color cameras, and (b) color quality improvement by achieving sub-spectral resolution. Second, we elaborate on challenges of RGB color calibration using HS information due to non-ideal illumination sources and non-ideal hyperspectral camera characteristics. We describe several adjustment (calibration) approaches to compensate for wavelength and spatial dependencies of real acquisition systems. Finally, we evaluate two color cameras by establishing ground truth RGB values from hyperspectral imagery and by defining pixel-based, correlation-based and histogram-based error metrics. Our experiments are conducted with three illumination sources (fluorescent light, Oriel Xenon lamp and incandescent light); with one HS Opto-Knowledge Systems camera and two color (RGB) cameras, such as Sony and Canon. We show a data-driven color-calibration as a method for improving image color quality. The applications of the developed techniques for HS to RGB image calibrations and sub-spectral resolution predictions are related to real-time model-based scene classification and scene simulation.
We present a novel methodology for evaluating statistically predicted versus measured multi-modal imagery, such as Synthetic Aperture Radar (SAR), Electro-Optical (EO), Multi-Spectral (MS) and Hyper-Spectral (HS) modalities. While several scene modeling approaches have been proposed in the past for multi-modal image predictions, the problem of evaluating synthetic and measured images has remained an open issue. Although analytical prediction models would be appropriate for accuracy evaluations of man-made objects, for example, SAR target modeling based on Xpatch, the analytical models cannot be applied to prediction evaluation of natural scenes because of their randomness and high geometrical complexity imaged by any of the aforementioned sensor modality. Thus, statistical prediction models are frequently chosen as more appropriate scene modeling approaches and there is a need to evaluate the accuracy of statistically predicted versus measured imagery. This problem poses challenges in terms of selecting quantitative and qualitative evaluation techniques, and establishing a methodology for systematic comparisons of synthetic and measured images. In this work, we demonstrate clutter accuracy evaluations for modified measured and predicted synthetic images with statistically modeled clutter. We show experimental results for color (red, green and blue) and HS imaging modalities, and for statistical clutter models using Johnson's family of probability distribution functions (PDFs). The methodology includes several evaluation techniques for comparing image samples and their similarity, image histograms, statistical central moments, and estimated probability distribution functions (PDFs). Particularly, we assess correlation, histogram, chi-squared, pixel and PDF parameter based error metrics quantitatively, and relate them to a human visual perception of predicted image quality. The work is directly applicable to multi-sensor phenomenology modeling for exploitation, recognition and identification.
We present a novel semi-automated registration technique for 3D volume reconstruction from fluorescent laser scanning confocal microscope (LSCM) imagery. The developed registration procedure consists of (1) highlighting segmented regions as salient feature candidates, (2) defining two region correspondences by a user, (3) computing a pair of region centroids, as control points for registration, and (4) transforming images according to estimated transformation parameters determined by solving a set of linear equations with input control points. The presented semi-automated method is designed based on our observations that (a) an accurate point selection is much harder for a human than an accurate region (segment) selection, (b) a centroid selection of any region is less accurate by a human than by a computer, and (c) registration based on structural shape of a region rather than on intensity-defined point is more robust to noise and to morphological deformation of features across stacks. We applied the method to image mosaicking and image alignment registration steps and evaluated its performance with 20 human subjects on LSCM images with stained blood vessels. Our experimental evaluation showed significant benefits of automation for 3D volume reconstruction in terms of achieved accuracy, consistency of results and performance time. In addition, the results indicate that the differences between registration accuracy obtained by experts and by novices disappear with an advanced automation while the absolute registration accuracy increases.
We present an information gathering system for medical image inspection that consists of software tools for capturing computer-centric and human-centric information. Computer-centric information includes (1) static annotations, such as (a) image drawings enclosing any selected area, a set of areas with similar colors, a set of salient points, and (b) textual descriptions associated with either image drawings or links between pairs of image drawings, and (2) dynamic (or temporal) information, such as mouse movements, zoom level changes, image panning and frame selections from an image stack. Human-centric information is represented by video and audio signals that are acquired by computer-mounted cameras and microphones. The short-term goal of the presented system is to facilitate learning of medical novices from medical experts, while the long-term goal is to data mine all information about image inspection for assisting in making diagnoses.
In this work, we built basic software functionality for gathering computer-centric and human-centric information of the aforementioned variables. Next, we developed the information playback capabilities of all gathered information for educational purposes. Finally, we prototyped text-based and image template-based search engines to retrieve information from recorded annotations, for example, (a) find all annotations containing the word "blood vessels", or (b) search for similar areas to a selected image area. The information gathering system for medical image inspection reported here has been tested with images from the Histology Atlas database.
This paper presents a novel approach to multi-sensor statistical modeling of bi-directional texture functions (BTF). Our proposed BTF modeling approach is based on (1) conducting an analytical study that relates a sensor resolution to the size and shape of elements forming material surface, (2) developing a robotic system for laboratory BTF data acquisition, (3) researching an application of the Johnson family of statistical probability distribution functions (PDF) to BTF modeling, (4) selecting optimal feature space for statistical BTF modeling, (5) building a database of parameters for the Johnson family of PDFs that after interpolations forms a high-dimensional statistical BTF model and (6) researching several statistical quality metrics that can be used for verification and validation of the obtained BTF models. The motivation for developing the proposed statistical BTF modeling approach comes from the facts that (a) analytical models have to incorporate randomness of outdoor scene clutter surfaces and (b) models have to be computationally feasible with respect to the complexity of modeled interactions between light and materials. The major advantages of our approach over other approaches are (a) the low computational requirements on BTF modeling (BTF model storage, fast BTF model-based generation), (b) flexibility of the Johnson family of PDFs to cover a wide range of PDF shapes and (c) applicability of the BTF model to a wide range of spectral sensors, e.g., color, multi-spectral or hyperspectral cameras. The prime applications for the proposed BTF model are multi-sensor automatic target recognition (ATR), and scene understanding and simulation.