Computer Assisted Image Analysis

Illumination invariance remains the most researched, yet the most challenging aspect of automatic face recognition. In this paper we investigate the discriminative power of colour-based invariants in the presence of large illumination changes between training and test data, when appearance changes due to cast shadows and non-Lambertian effects are significant. Specifically, there are three main contributions: (i) we employ a more sophisticated photometric model of the camera and show how its parameters can be estimated, (ii) we derive several novel colour-based face invariants, and (iii) on a large database of video sequences we examine and evaluate the largest number of colour-based representations in the literature. Our results suggest that colour invariants do have a substantial discriminative power which may increase the robustness and accuracy of recognition from low resolution images.

Dynamically changing background ("dynamic background") still presents a great challenge to many motion-based video surveillance systems. In the context of event detection, it is a major source of false alarms. There is a strong need from the security industry either to detect and suppress these false alarms, or dampen the effects of background changes, so as to increase the sensitivity to meaningful events of interest. In this paper, we restrict our focus to one of the most common causes of dynamic background changes: that of swaying tree branches and their shadows under windy conditions. Considering the ultimate goal in a video analytics pipeline, we formulate a new dynamic background detection problem as a signal processing alternative to the previously described but unreliable computer vision-based approaches. Within this new framework, we directly reduce the number of false alarms by testing if the detected events are due to characteristic background motions. In addition, we introduce a new dataset suitable for the evaluation of dynamic background detection. It consists of real-world events detected by a commercial surveillance system from two static surveillance cameras. The research question we address is whether dynamic background can be detected reliably and efficiently using simple motion features and in the presence of similar but meaningful events such as loitering. Inspired by the tree aerodynamics theory, we propose a novel method named local variation persistence (LVP), that captures the key characteristics of swaying motions. The method is posed as a convex optimization problem whose variable is the local variation. We derive a computationally efficient algorithm for solving the optimization problem, the solution of which is then used to form a powerful detection statistic. On our newly collected dataset, we demonstrate that the proposed LVP achieves excellent detection results and outperforms the best alternative adapted from existing art in the dynamic background literature.

The purpose of the present study was to evaluate two dental macroscopic age estimation criteria. Namely root dentine translucency (RDT) and dental root color measured as a mean value of the red, blue, and green components of color (RGB). The sample consisted of 416 freshly extracted single rooted permanent teeth collected from 311 individual referred to several dentists and dental clinics in Kom Ombou district, south Egypt in the Nubian resettlement area. RDT and RGB were assessed using image analysis quantification techniques.
Analysis reveals that there is high linear correlation between both RDT and RGB values with age, with r = 0.951 and –0.985 respectively. Regression analyses yields a  new set of equations  to estimate age among Egyptian Nubian population with high correlation values ranging from r = 0.90 to 0.97 and a standard error of estimated ranging from ± 5.067 to ±2.996.  The best estimates were obtained by combining both RGB and RDT.
From the results it is concluded that RDT and RGB can be used to assess an accurate estimate of age and this technique can form a basis of practical application in establishing age both in forensics and paleodemographic aspects.

Estimation of adult skeletal age-at-death is not only one of the most important identifying features for an unknown individual but also one of the most difficult to achieve especially among old adults. The main source of the problem is the nature of human aging. This feature is characterized by an accumulation of metabolic disorders that show great variation in the level and the degree of change with increasing age both within and between populations. Moreover, individual aging is determined by the interplay of complex set of genetic, culture and environmental factors.

Variation in the biological aging process has profound effects on age-t death assessment. The relationship between chronological age and skeletal age indicators is neither constant nor linear. The assumption that the underlying biological basis of the age-indicator relationship is constant across different populations is erroneous. Therefore, it is necessary to take into account that age changes are not uniform across populations.

It has been demonstrated that when applied an aging method to independent populations of known age at death, it proves less reliable than the results obtained when applying the method to samples from the population used to formulate the methods.

The purpose of the present study was to compare and evaluate two dental
macroscopic age estimation criteria. Namely root dentine translucency (RDT) and dental root color measured as a mean value of the red, blue, and green components of color (RGB). The sample consisted of 416 freshly extracted single rooted permanent teeth collected from 311 individuals referred to several dentists and dental clinics in Kom Ombou district, south Egypt in the Nubian resettlement area.

RDT and RGB were assessed using traditional (manual) and image analysis quantification techniques. Analysis reveals that there is high linear correlation between both RDT and RGB values with age, with r = 0.951 and –0.985 respectively. Regression analyses yields a new set of equations to estimate age among Egyptian Nubian population with high correlation values ranging from r = 0.90 to 0.97 and a standard error of the estimate ranging from ± 5.067 to ±2.996. The best estimates were obtained by combining both RGB and RDT. Additionally, the analysis has shown that sex of the individual don’t affect age assessment.

It was concluded that RDT and RGB can be used to assess age at death
with high accuracy. Moreover, this technique can form a basis of practical
application in establishing age at death both in forensics and paleodemographic
aspects.

The author considers the Mona Lisa a a major site of artistic investigations of gender and identity. Specifically, she analyzes the symbiotic relationship between two landmark reinventions of the Mona Lisa: first, Marcel Duchamp’s notorious L.H.O.O.Q. (Mona-Lisa-with-a-mustache), and second, Lillian Schwartz’s pioneering use of computers to demonstrate that the main model for the Mona Lisa was Leonardo himself. Both L.H.O.O.Q. and Schwartz’s Mona/Leo project are a striking instance of the Bearded Lady archetype, which finds its match in Duchamp’s L.H.O.O.Q. rasée (L.H.O.O.Q. Shaved), a variant of the Shaven Man.

This is a summary of the (apparently unanimous) scholarly evaluation of the Hockney-Falco theory for 1420-1550, based on rigorous optical analyses, computer vision, material studies, art history, history of optics and science, and experimental tests.

Complex color documents with both graphics and text, where the text varies in color and size, call for specialized binarization techniques. We propose a novel method for binarization of color documents, whereby the foreground text is output as black and the background as white, regardless of the polarity of foreground and background shades. The method employs an edge-based connected component approach to determine text-like components and binarize them individually. The threshold for binarization and the logic for inverting the output are derived from the image data and do not require any manual tuning. Unlike existing binarization methods, our technique can handle documents with multi-colored texts with different background shades. The method is applicable to documents having text of widely varying sizes, usually not handled by local binarization methods. Experiments on a broad domain of target document types illustrate the effectiveness and adaptability of the method.

An on-line computer technique is described for the detection and quantitation of left-to-right shunts. The computer used Student's t test to evaluate the probability that variations in right heart oxygen saturations were due to left-to-right shunts. The computer method was found to compare favorably to the present step-up method of detecting left-to-right shunts. The computer also calculated pulmonary blood flow, systemic blood flow, their ratio, and shunt blood flow as accurately as standard methods, as evidenced by correlation coefficients of 0.98, 0.94, 0.95, and 0.71 between the computer and hand calculations.

Fish usually have a simple body shape and many morphological characters can be easily obtained by 2-dimensional image
analysis. Measuring errors are often analyzed using data from several researchers or repeated measurements, but the testing of accuracy and repeatability based on different methods has rarely been examined. Here we report the comparison of manual measurements using Proma 150D digital calipers and computer-assisted image analysis in LUCIA software to evaluate the accuracy of both methods and the suitability of digital images for exploration of fish morphology. Eighteen morphometric characters of 20 specimens of Prussian carp were examined by 2 researchers. The mean of the relative bias within and between researchers separately for the manual and computerassisted methods (expressed as a percentage of a character’s value) varied from –4.27% to 9.48%. Comparison of interresearcher results using digital caliper, interresearcher results using LUCIA, intraresearcher results using digital caliper, and the 2 methods using manual caliper and LUCIA revealed significant differences in several characters. Only testing of intraresearcher relative bias using LUCIA showed no significant difference.

The analysis of human crowds has widespread uses from law enforcement to urban engineering and traffic management. All of these require a crowd to first be detected, which is the problem addressed in this paper. Given an image, the algorithm we propose segments it into crowd and non-crowd regions. The main idea is to capture two key properties of crowds: (i) on a narrow scale, its basic element should look like a human (only weakly so, due to low resolution, occlusion, clothing variation etc.), while (ii) on a larger scale, a crowd inherently contains repetitive appearance elements. Our method exploits this by building a pyramid of sliding windows and quantifying how "crowd-like" each level of the pyramid is using an underlying statistical model based on quantized SIFT features. The two aforementioned crowd properties are captured by the resulting feature vector of window responses, describing the degree of crowd-like appearance around an image location as the surrounding spatial extent is increased.

Reconnaissance, Surveillance and Target Acquisition (RSTA) systems are increasingly based on advanced technology and are required to handle a diversity of sensors to exploit the associated forms of collected imagery.  The future real time collection and dissemination of imagery will require co-operative use of multi-user workstations that have the capacity to display a variety of different forms of imagery (for example, Synthetic Aperture Radar (SAR), Moving Target Indicator (MTI) Radar, Electro Optic (EO) Sensors, and Infra-Red (IR)).
Currently, technology offers important assistance for the collection, display, storage, display and dissemination of imagery.  However, only minimal technical assistance is given to the IA who is required to analyse and exploit the presented imagery.  It is clear that in current 'technologically sophisticated systems' there is only limited support to the skilled performance of the cognitive and perceptual analyses that are required to interpret imagery.
  Furthermore, the introduction of this new technology has changed the nature and methods of imagery analysis and the associated work practices that the IA needs to undertake.  These new work practices are also being relocated into new work domains where the maintenance of situational awareness is becoming more important (Reference 1).  However, few studies have addressed the problem of taking a user-centred perspective to both qualitatively and quantitatively explore the cognitive and perceptual aspects of imagery analysis.    ISBN: 0 444 82183 X

Illumination invariance remains the most researched, yet the most challenging aspect of automatic face recognition. In this paper we propose a novel, general recognition framework for efficient matching of individual face images, sets or sequences. The framework is based on simple image processing filters that compete with unprocessed greyscale input to yield a single matching score between individuals. It is shown how the discrepancy between illumination conditions between novel input and the training data set can be estimated and used to weigh the contribution of two competing representations. We describe an extensive empirical evaluation of the proposed method on 171 individuals and over 1300 video sequences with extreme illumination, pose and head motion variation. On this challenging data set our algorithm consistently demonstrated a dramatic performance improvement over traditional filtering approaches. We demonstrate a reduction of 50-75% in recognition error rates, the best performing method-filter combination correctly recognizing 96% of the individuals.

This report describes the design, implementation, evaluation and original enhancements to the Live-Wire method for 2D and 3D image segmentation. Live-Wire 2D employs a semi-automatic paradigm; the user is asked to select a few boundary points of the object to segment, to steer the process in the right direction, while the result is displayed in real time. In our implementation segmentation is extended to three dimensions by performing this process on a slice-by-slice basis. User's time and involvement is further reduced by allowing him to specify object contours in planes orthogonal to the slices. If these planes are chosen strategically, Live-Wire 3D can perform 2D segmentation in the plane of each slice automatically. This report also proposes two improvements to the original method, path heating and a new graph edge feature function based on variance of path properties along the boundary. We show that these improvements lead up to a 33% reduction in interaction with the user, and improved delineation in presence of strong interfering edges.

Background in music psychology. Physical movement plays an important role in musical perception and production. It is generally agreed among professional singers and vocal teachers, for example, that there are relationships between the kinematics of a singer's body and the quality of their voice. Thus, we might expect to find relationships between quantifiable indicators of a singer's vocal performance and

Heavy metal resistance (HMR) in Eubacteria is regulated by a variety of systems including transcription factors from the MerR family (COG0789). The HMR systems are characterized by the complex signal structure (strong palindrome within a 19 or 20 bp promoter spacer), and usually consist of transporter and regulator genes. Some HMR regulons also include detoxification systems. The number of sequenced bacterial genomes is constantly increasing and even though HMR resistance regulons of the COG0789 type usually consist of few genes per genome, the computational analysis may contribute to the understanding of the cellular systems of metal detoxification. We studied the mercury (MerR), copper (CueR and HmrR), cadmium (CadR), lead (PbrR), and zinc (ZntR) resistance systems and demonstrated that combining protein sequence analysis and analysis of DNA regulatory signals it was possible to distinguish metal-dependent members of COG0789, assign specificity towards particular metals to unchara...

Genome-wide computational analysis of alternative splicing (AS) in several flowering plants has revealed that pre-mRNAs from about 30% of genes undergo AS. Chlamydomonas, a simple unicellular green alga, is part of the lineage that includes land plants. However, it diverged from land plants about one billion years ago. Hence, it serves as a good model system to study alternative splicing in early photosynthetic eukaryotes, to obtain insights into the evolution of this process in plants, and to compare splicing in simple unicellular photosynthetic and non-photosynthetic eukaryotes. We performed a global analysis of alternative splicing in Chlamydomonas reinhardtii using its recently completed genome sequence and all available ESTs and cDNAs. Our analysis of AS using BLAT and a modified version of the Sircah tool revealed AS of 498 transcriptional units with 611 events, representing about 3% of the total number of genes. As in land plants, intron retention is the most prevalent form of AS. Retained introns and skipped exons tend to be shorter than their counterparts in constitutively spliced genes. The splice site signals in all types of AS events are weaker than those in constitutively spliced genes. Furthermore, in alternatively spliced genes, the prevalent splice form has a stronger splice site signal than the non-prevalent form. Analysis of constitutively spliced introns revealed an over-abundance of motifs with simple repetitive elements in comparison to introns involved in intron retention. In almost all cases, AS results in a truncated ORF, leading to a coding sequence that is around 50% shorter than the prevalent splice form. Using RT-PCR we verified AS of two genes and show that they produce more isoforms than indicated by EST data. All cDNA/EST alignments and splice graphs are provided in a website at http://combi.cs.colostate.edu/as/chlamy. The extent of AS in Chlamydomonas that we observed is much smaller than observed in land plants, but is much higher than in simple unicellular heterotrophic eukaryotes. The percentage of different alternative splicing events is similar to flowering plants. Prevalence of constitutive and alternative splicing in Chlamydomonas, together with its simplicity, many available public resources, and well developed genetic and molecular tools for this organism make it an excellent model system to elucidate the mechanisms involved in regulated splicing in photosynthetic eukaryotes.

Successful application of posttranscriptional gene silencing (PTGS) for gene function study in both plants and animals depends on high target specificity and silencing efficiency. By computational analysis with genome and/or transcriptome sequences of 25 plant species, we predicted that about 50% to 70% of gene transcripts in plants have potential off-targets when used for PTGS that could obscure experimental results.

We analysed multi-channel electroencephalographic (EEG) recordings during a spatial Working Memory (WM) task in order to test the hypothesis that segmentation of perception and action is present when the visual stimulus has been stored in spatial WM. To detect the interactions between different regions of the brain depending on the task we employed both Short Time Fourier Transformation (STFT) and the concept of Granger Causality (GC). Our computational analysis supports evidence that the Parietal Cortex (PC) is involved in WM processing.

Background: Morphometry and quantitative image analysis have shown considerable potential for diagnostic applications because they eliminate subjectivity and increase the measurement reproduc-ibility. In our study, we investigated the use of morphometric quantitative image analysis as a tool to aid diagnosis of lesion tissues stained with seven staining techniques. Material and methods: Thirteen already established lesions were examined in this study. Tissues were sectioned at 5 mm thickness using the rotary microtome and prepared for staining according to standard methods before staining them with seven different histochemical dyes: Haematoxylin and Eosin (H&E), Masson's Trichrome (MT), Periodic Acid Schiff (PAS), Phosphotungstic Acid Haematoxylin (PTAH), Southgate Mucincamine (SGM), Alcian Blue (AB) and Verhoeff Van Gieson (VVG). Photomicrographs were analyzed using imageJ for percentage area and intensity measurement. Results: The average % area and intensity measurement based on the seven staining techniques used in this study were 47.88% and 122.23 points (pts). The lesions, adenomyosis in MT (80%), squamous cell carcinoma in PAS (80%) and fibroma in PTAH (77%) stained above the average % area. The squamous cell carcinoma stained with MT was the most intense tissue (55 pts), followed by Brenner tumour in PTAH (61 pts) and squamous cell carcinoma in PAS (69 pts). There were significant correlations between the % area of tissues covered by PAS and H&E (r ¼ 0.599, p ¼ 0.030), and PAS and MT (r ¼ 0.572, p ¼ 0.041) as well as in the intensity of tissues in PAS and H&E (r¼0.615, p¼0.025), VVG and H&E (r ¼ 0.707, p ¼ 0.007), and VVG and PTAH (r ¼ 0.577, p ¼ 0.039). Conclusion: MT staining technique produced the best results in the % area of tissues covered and the intensity measurements, and therefore was recommended for routine use alongside with H&E in diagnostic histopathology.

... the dependence of the ventilation rate on the greenhouse design and improve its ventilation ...Natural ventilation is driven by pressure differences created at ventilating openings either by the wind or ... Ventilation due to the buoyancy effect is important in situations of relatively high ...