Search

Stereo superpixel segmentation aims to obtain the superpixel segmentation results of the left and right views more cooperatively and consistently, rather than simply performing independent segmentation directly. Thus, the correspondence between two views should be reasonably modeled and fully considered. In this paper, we propose a left-right interactive optimization framework for stereo superpixel segmentation. Considering the disparity in stereo image pairs, we first divide the images into paired region and non-paired region, and propose a collaborative optimization scheme to coordinately refine the matched superpixels of the left and right views in an interactive manner. This is, to the best of our knowledge, the first attempt to generate stereo superpixels considering the parallax consistency. Quantitative and qualitative experiments demonstrate that the proposed framework achieves superior performance in terms of consistency and accuracy compared with single-image superpixel segmentation.

There is an increasing recognition of the critical importance of charging for the safety and life of lithium-ion batteries. This paper proposes a computationally efficient optimal control approach for the problem of real-time charging control. By incorporating specific constraints that must be satisfied during charging, a health-aware operation is promoted. To determine the optimal charging current in the given setup, a recently proposed iterative framework for solving constrained optimal control problems is leveraged. It is found that the resulting optimal charging currents can be expressed in terms of a piecewise-affine time-invariant state feedback law, which results in a high computational efficiency for the optimal control solution.

Estimating the trustworthiness of a set of actors when all the available information is provided by the actors themselves is a hard problem. When two actors have conflicting reports about each other, how do we establish which of the two (if any) deserves our trust? In this paper, we model this scenario as a network problem: actors are nodes in a network and their reports about each other are the edges of the network. To estimate their trustworthiness levels, we develop an iterative framework which looks at all the reports about each connected actor pair to define its trustworthiness balance. We apply this framework to a customer/supplier business network. We show that our trustworthiness score is a significant predictor of the likelihood a business will pay a fine if audited. We show that the market network is characterized by homophily: businesses tend to connect to partners with similar trustworthiness degrees. This suggests that the topology of the network influences the behavior of the actors composing it, indicating that market regulatory efforts should take into account network theory to prevent further degeneration and failures.

Superpixel decomposition methods are widely used in computer vision and image processing applications. By grouping homogeneous pixels, the accuracy can be increased and the decrease of the number of elements to process can drastically reduce the computational burden. For most superpixel methods, a trade-off is computed between 1) color homogeneity, 2) adherence to the image contours and 3) shape regularity of the decomposition. In this paper, we propose a framework that jointly enforces all these aspects and provides accurate and regular Superpixels with Contour Adherence using Linear Path (SCALP). During the decomposition, we propose to consider color features along the linear path between the pixel and the corresponding superpixel barycenter. A contour prior is also used to prevent the crossing of image boundaries when associating a pixel to a superpixel. Finally, in order to improve the decomposition accuracy and the robustness to noise, we propose to integrate the pixel neighborhood information, while preserving the same computational complexity. SCALP is extensively evaluated on standard segmentation dataset, and the obtained results outperform the ones of the state-of-the-art methods. SCALP is also extended for supervoxel decomposition on MRI images., Computer Vision and Image Understanding (CVIU), 2018

Summary Early quantitative structure-activity relationship (QSAR) technologies have unsatisfactory versatility and accuracy in fields such as drug discovery because they are based on traditional machine learning and interpretive expert features. The development of Big Data and deep learning technologies significantly improve the processing of unstructured data and unleash the great potential of QSAR. Here we discuss the integration of wet experiments (which provide experimental data and reliable verification), molecular dynamics simulation (which provides mechanistic interpretation at the atomic/molecular levels), and machine learning (including deep learning) techniques to improve QSAR models. We first review the history of traditional QSAR and point out its problems. We then propose a better QSAR model characterized by a new iterative framework to integrate machine learning with disparate data input. Finally, we discuss the application of QSAR and machine learning to many practical research fields, including drug development and clinical trials., Graphical abstract, Data analysis in structural biology; Machine learning; Structural biology

A recently emerging seismic acquisition technology called simultaneous source shooting has attracted much attention from both academia and industry. The key topic in the newly developed technique is the removal of intense blending interferences caused by the simultaneous ignition of multiple airgun sources. In this paper, I propose a novel inversion strategy with multiple convex constraints to improve the deblending performance based on the projection onto convex sets (POCS) iterative framework. In the POCS iterative framework, as long as the multiple constraints are convex, the iterations are guaranteed to converge. In addition to the sparse constraint, I seek another important constraint from the untainted data. I create a blending mask in order to fully utilize the useful information hidden behind the noisy blended data. The blending mask is constructed by numerically blending a matrix with all its entries set to be one and then setting the non-one entries of the blended matrix zero. I use both synthetic and field data examples to demonstrate the successful performance of the proposed method.

Since disruptions in railway networks are inevitable, railway operators and infrastructure managers need reliable measures and tools for disruption management. Current literature on railway disruption management focuses most of the time on rescheduling one resource (timetable, rolling stock or crew) at the time. In this research, we describe the application of an iterative framework in which all these three resources are considered. The framework applies existing models and algorithms for rescheduling the individual resources. We extensively test our framework on instances from Netherlands Railways and show that schedules which are feasible for all three resources can be obtained within short computation times. This case study shows that the framework and the existing rescheduling approaches can be of great value in practice. HighlightsWe present an iterative framework for real-time railway disruption management.Our framework reschedules the timetable, the rolling stock, and the crew.Our approach is tested on a huge set of real-world instances from the Netherlands.The framework converges within few iterations for all test instances.

Graphical abstractDisplay Omitted HighlightsAdd and delete operations are encoded as a list/string of integers (ADL).An effective hyper-heuristic approach operating with ADLs is proposed.Low level heuristics perform search over the space of feasible solutions.Proposed approach produces new best solutions to some instances.Proposed approach achieves generality across two variants of the timetabling problem. Hyper-heuristics are (meta-)heuristics that operate at a higher level to choose or generate a set of low-level (meta-)heuristics in an attempt of solve difficult optimization problems. Iterated local search (ILS) is a well-known approach for discrete optimization, combining perturbation and hill-climbing within an iterative framework. In this study, we introduce an ILS approach, strengthened by a hyper-heuristic which generates heuristics based on a fixed number of add and delete operations. The performance of the proposed hyper-heuristic is tested across two different problem domains using real world benchmark of course timetabling instances from the second International Timetabling Competition Tracks 2 and 3. The results show that mixing add and delete operations within an ILS framework yields an effective hyper-heuristic approach.

Living labs and co-production are increasingly popular strategies for universities to address sustainability challenges and yet the links between them remain largely implicit. This paper discusses the potential of living labs to provide a holistic and iterative framework for the co-production of knowledge. The University Living Lab initiative was launched in 2012 to transform the University of Manchester campus into a site for applied teaching and research around sustainability. Its goal was to provide a framework for students and academics to engage with the opportunities to work with Estates staff and their environmental consultants on applied sustainability challenges. This paper discusses the generation of living lab projects, the design of the campus as a living lab, and institutional visibility, identifying the key strengths of the living lab approach and the challenges of applying it more broadly.

A train slot selection model based on multicommodity network flow concepts is developed for determining freight train timetables for scheduling rail services along multiple interconnected routes. The model seeks to minimize operating costs incurred by carriers and delays incurred by shippers while ensuring that the schedules and demand levels are mutually consistent. A column generation-based methodology is proposed for train slot selection to meet frequency requirements. This methodology is embedded in a simulation-based iterative framework, where demand for rail services is re-computed in accordance with the train schedule obtained by solving the freight train scheduling problem.

In high level synthesis, module selection, scheduling, and resource binding are inter-dependent tasks. For a selected module set, the best schedule/binding should be generated in order to accurately assess the quality of a module selection. Exhaustively enumerating all module selections and constructing a schedule and binding for each one of them can be extremely expensive. In this paper, we present an iterative framework, called W i Z ard to solve module selection problem under resource, latency, and power constraints. The framework associates a utility measure with each module. This measurement reflects the usefulness of the module for a given a design goal. Using modules with high utility values should result in superior designs. We propose a heuristic which iteratively perturbs module utility values until they lead to good module selections. Our experiments show that by keeping modules with high utility values, W i Z ard can drastically reduce the module exploration space (approximately 99.2% reduction). Furthermore, the module selections formed by these modules belong to superior solutions in the enumerated set (top 15%).