Dynamic graph cuts for efficient inference in Markov Random Fields

Abstract-In this paper we present a fast new fully dynamic algorithm for the st-mincut/max-flow problem. We show how this algorithm can be used to efficiently compute MAP solutions for certain dynamically changing MRF models in computer vision such as image segmentation. Specifically, given the solution of the max-flow problem on a graph, the dynamic algorithm efficiently computes the maximum flow in a modified version of the graph. The time taken by it is roughly proportional to the total amount of change in the edge weights of the graph. Our experiments show that, when the number of changes in the graph is small, the dynamic algorithm is significantly faster than the best known static graph cut algorithm. We test the performance of our algorithm on one particular problem: the object-background segmentation problem for video. It should be noted that the application of our algorithm is not limited to the above problem, the algorithm is generic and can be used to yield similar improvements in many other cases that involve dynamic change.     

An ontology of slums for image-based classification

Information about rapidly changing slum areas may support the development of appropriate interventions by concerned authorities. Often, however, traditional data collection methods lack information on the spatial distribution of slum-dwellers. Remote sensing based methods could be used for a rapid inventory of the location and physical composition of slums. (Semi-)automatic detection of slums in image data is challenging, owing to the high variability in appearance and definitions across different contexts. This paper develops an ontological framework to conceptualize slums using input from 50 domain-experts covering 16 different countries. This generic slum ontology (GSO) comprises concepts identified at three levels that refer to the morphology of the built environment: the environs level, the settlement level and the object level. It serves as a comprehensive basis for image-based classification of slums, in particular, using object-oriented image analysis (OOA) techniques. This is demonstrated by with an example of local adaptation of GSO and OOA parameterization for a study area in Kisumu, Kenya. At the object level, building and road characteristics are major components of the ontology. At the settlement level, texture measures can be potentially used to represent the contrast between planned and unplanned settlements. At the environs level, factors which extend beyond the site itself are important indicators, e.g. hazards due to floods plains and marshy conditions. The GSO provides a comprehensive framework that includes all potentially relevant indicators that can be used for image-based slum identification. These characteristics may be different for other study areas, but show the applicability of the developed framework.     

Correlation of Mechanical Properties with Diameter and Cooling Rate of 1080 Wire-Rod

More than 540 heats of 1080 wire-rod were statistically analyzed by regression analyses to see whether tensile strength and percent reduction in area (%RA) relate to wire-rod diameter and composition. As diameter increases from 5.6 to 12.7 mm, the trend in %RA shows a decrease with negligible effect on the trend of the tensile strength. It was found that the estimated cooling rate at 700 °C during controlled cooling is responsible for the “diameter effect.” The effect of composition on %RA is minor when contrasted to the “diameter effect.” In particular, the effect of the concentrations of the residual elements on %RA within the compositional range studied is negligible.     

Robust Higher Order Potentials for Enforcing Label Consistency

This paper proposes a novel framework for labelling problems which is able to combine multiple segmentations in a principled manner. Our method is based on higher order conditional random fields and uses potentials defined on sets of pixels (image segments) generated using unsupervised segmentation algorithms. These potentials enforce label consistency in image regions and can be seen as a generalization of the commonly used pairwise contrast sensitive smoothness potentials. The higher order potential functions used in our framework take the form of the Robust P ⁿ model and are more general than the P ⁿ Potts model recently proposed by Kohli et al. We prove that the optimal swap and expansion moves for energy functions composed of these potentials can be computed by solving a st-mincut problem. This enables the use of powerful graph cut based move making algorithms for performing inference in the framework. We test our method on the problem of multi-class object segmentation by augmenting the conventional crf used for object segmentation with higher order potentials defined on image regions. Experiments on challenging data sets show that integration of higher order potentials quantitatively and qualitatively improves results leading to much better definition of object boundaries. We believe that this method can be used to yield similar improvements for many other labelling problems.     

Secondary membranoproliferative glomerulonephritis due to hemolytic uremic syndrome: an unusual presentation

In vitro selection can be used to generate nucleic acid ligands (aptamers) to target molecules ranging in size and structure from cations to cells. However, the selection process is repetitive and time-consuming. We have automated a protocol for in vitro selection using an augmented Beckman Biomek 2000 pipetting robot. The automated selection procedure requires the integration of four devices and the optimization of four molecular biology methods, and is one of the most complex automated protocols attempted to date. Initial attempts at selection yielded robust replication parasites, but optimization of the automated selection procedure suppressed the emergence of these parasites and led to the selection of true nucleic acid ligands. Automated selection can now be used to generate nucleic acid aptamers in days rather than weeks or months.     

Template synthesis of gold nanotubes in an anodic alumina membrane

Nanotube-containing membranes prepared by the template method show promise for use as highly selective filters for membrane-based chemical and biological separations. Most of the work to date has been done on gold nanotubes prepared by electroless deposition of Au within the pores of polymeric filtration membranes. These polymeric filters have very low porosities (< 1%), and, as a result, the flux through Au nanotube membranes based on these templates is very low. In contrast, the other popular template membranes-anodic aluminas-have high porosities-30% to 50%. In spite of this potential advantage of anodic alumina templates, there have been no reports of electrolessly plated Au nanotubes within the pores of these templates. This is because the electroless plating method used to deposit Au nanotubes in polymeric templates does not work in aluminas. We have developed a modified electroless plating strategy that can be used to deposit high-quality Au nanotubes within the pores of the alumina templates. We describe this new plating method here.     

Analysis and Performance of a Chopper Controlled Separately Excited DC Motor

A combination of Fourier analysis of load voltage waveform and Laplace transformation is applied for determining the performance of a separately excited dc motor fed from a chopper controlled supply, under all modes of operation. The analysis takes into account the commutation interval and the results show that contrary to the assumption made in works reported earlier, commutation interval significantly affects the performance. For the case of continuous conduction, independent expressions for current and speed are obtained in terms of applied voltage, load torque, and chopper and machine parameters. However, in the case of discontinuous conduction, Fourier coefficients include speed voltage as an additional quantity and hence the equations for current and speed have to be solved as simultaneous equations. The analysis is used to predict the steady-state and transient response. Computed torque-speed characteristics are compared with the results obtained experimentally.     

Mass Exchange Between Hamilton Harbour and Lake Ontario

The mass exchange between Hamilton Harbour and Lake Ontario waters through the Burlington Canal is important for estimating the dissolved oxygen budget of the harbour. Lake-harbour exchange is caused by either the oscillatory flow in the canal during isothermal conditions or the densimetric flow during thermal stratification. During the study period (September 1975), the canal water was found to be quasi-isothermal; consequently oscillatory flow existed. A computational method was developed to estimate the exchange based on excursion distance travelled for each limnological episode, and the final flow in each direction was checked with dissolved solids budget. An average of 2.04 × 10⁶ m³.d^?1 (24 m³.s^?l) of harbour water is estimated to flow into the lake, while 0.73 × 10⁶ m³.d^?1 (8 m³.s^?1) of lake water flowed toward the harbour during September 1975. This accounts for the total and net daily exchange of 0.98% and 0.48% of harbour volume, with net exchange being toward the lake. The total and net exchanges were respectively 8 and 4 times the natural drainage during the study period. On a monthly average, more water leaves than enters the harbour. The lake-harbour exchange is considered important for maintaining and even improving the existing harbour water quality. The harbour dilution factor was estimated as 0.0019 per day for the present study.