Multiple objects tracking in the presence of long-term occlusions

We present a robust object tracking algorithm that handles spatially extended and temporally long object occlusions. The proposed approach is based on the concept of “object permanence” which suggests that a totally occluded object will re-emerge near its occluder. The proposed method does not require prior training to account for differences in the shape, size, color or motion of the objects to be tracked. Instead, the method automatically and dynamically builds appropriate object representations that enable robust and effective tracking and occlusion reasoning. The proposed approach has been evaluated on several image sequences showing either complex object manipulation tasks or human activity in the context of surveillance applications. Experimental results demonstrate that the developed tracker is capable of handling several challenging situations, where the labels of objects are correctly identified and maintained over time, despite the complex interactions among the tracked objects that lead to several layers of occlusions.     

An ontology based multi-agent system to support HABIO outsourcing framework

In recent years the widespread application of outsourcing to help improve finance and maintain competitiveness is growing rapidly, and has been accompanied by a growth of academic research in this field. Most of the outsourcing literature has focused primarily on the social and cultural impacts, the organisational and management perspectives of outsourcing and very few tools have been developed to assist managers in making outsourcing decisions. Incorrect decision-making in outsourcing can lead to devastating consequences for their business. To guide decision makers in managing the complexity of outsourcing a holistic approach to outsourcing has been developed by taking business, information and organisational perspectives (HABIO). The purpose of this paper is to demonstrate how HABIO can be applied to help managers evaluate the impact of their decision-making by providing them with a set of ‘what-if’ scenarios, which are enacted by intelligent agents. The case example, discussed in this paper, demonstrates that HABIO can be applied effectively using a multi-agent system to analyse the outsourcing decision for a call centre sector in the retail industry. A prototype of the multi-agent system is presented where intelligent agents are endowed with specific domain ontologies, appropriate to solve a specific problem, and with a reasoning mechanism to achieve specific goals and to collaborate together to produce a set of “what-if scenarios” to the outsourcing managers.     

ERP data sharing framework using the Generic Product Model (GPM)

Nowadays, all product life cycle processes are investigated deeply in order to get an advantage over competitors. To support these processes, several software applications are available. However, this wide range of heterogeneous applications leads to a large variety of hardware and operating systems, data management software, data models, schemas, and data semantics that hinder the information sharing process. To tackle this problem, Hitachi Company has developed a modeling language called Generic Product Model (GPM) in addition to several translators from native formats to GPM for storing, sharing and visualizing product data in a single data warehouse. In order to broaden the range of application data able to be stored and shared using the GPM data warehouse, this paper presents a methodology and a translator that allow management data to be included in a GPM data warehouse. Given the fact that most management data are stored in Enterprise Resource Planning (ERP) software, and that the latter allows extraction of database contents into Excel files, we propose a translator that handles Enterprise Resource Planning management data that are already extracted into Excel Format.The proposed framework enables data management contained in flat Excel Files to be translated into structured GPM data. Translation methodologies are given for the translation of mass production and customized product data through two case studies, one in the computer retail sector and the other in the extrusion machine industry.     

Using LiDAR and quickbird data to model plant production and quantify uncertainties associated with wetland detection and land cover generalizations

Spatiotemporal data from satellite remote sensing and surface meteorology networks have made it possible to continuously monitor global plant production, and to identify global trends associated with land cover/use and climate change. Gross primary production (GPP) and net primary production (NPP) are routinely derived from the Moderate Resolution Imaging Spectroradiometer (MODIS) onboard satellites Terra and Aqua, and estimates generally agree with independent measurements at validation sites across the globe. However, the accuracy of GPP and NPP estimates in some regions may be limited by the quality of model input variables and heterogeneity at fine spatial scales. We developed new methods for deriving model inputs (i.e., land cover, leaf area, and photosynthetically active radiation absorbed by plant canopies) from airborne laser altimetry (LiDAR) and Quickbird multispectral data at resolutions ranging from about 30 m to 1 km. In addition, LiDAR-derived biomass was used as a means for computing carbon-use efficiency. Spatial variables were used with temporal data from ground-based monitoring stations to compute a six-year GPP and NPP time series for a 3600 ha study site in the Great Lakes region of North America. Model results compared favorably with independent observations from a 400 m flux tower and a process-based ecosystem model (BIOME-BGC), but only after removing vapor pressure deficit as a constraint on photosynthesis from the MODIS global algorithm. Fine-resolution inputs captured more of the spatial variability, but estimates were similar to coarse-resolution data when integrated across the entire landscape. Failure to account for wetlands had little impact on landscape-scale estimates, because vegetation structure, composition, and conversion efficiencies were similar to upland plant communities. Plant productivity estimates were noticeably improved using LiDAR-derived variables, while uncertainties associated with land cover generalizations and wetlands in this largely forested landscape were considered less important.     

A method for least-cost QoS multicast routing based on genetic simulated annealing algorithm

A method for least-cost QoS multicast routing based on genetic simulated annealing algorithm (NGSA) is presented. Genetic algorithm and simulated annealing algorithm are combined to improve the computing performance in this method. The chromosomes of the multicast tree are represented by tree structure coding to save the time of conversion between encoding space and solution space. A new population initialization method is used to make sure that every chromosome in initial population is reasonable multicast tree without loops. The adaptive crossover probability is used to improve the evolutionary efficiency. The simulation results show that this method has high speed of convergence and search capability. And the problem of least-cost QoS multicast routing is solved effectively.     

A new process for fabricating tip-shaped polymer microstructure array with patterned metallic coatings

In this paper, we present a new process for fabricating tip-shaped polymer microstructure array coated by patterned metal layer. This new process involves three techniques including: micro-molding, patterned metal layer transfer, and electrochemical-based sacrificial layer. As we know, it is very difficult to remove the extra photoresist (PR) in the concave areas in traditional micro-fabrication technology, which hinders patterning metal layers on three-dimensional (3D) microstructures. The electrochemical-based sacrificial layer technique can effectively resolve this problem, which is verified by scanning electron microscopy (SEM) characterization. Comparative study between the 3D metal-coated polyimide microstructures fabricated with and without the electrochemical-based sacrificial layer step is also performed and SEM images proved the effect of the presented process. The applicability of the developed process is also demonstrated with the successful fabrication of a pyramid-shaped polyimide microelectrodes array for neural stimulation.     

Affine feature extraction: A generalization of the Fukunaga–Koontz transformation

Dimension reduction methods are often applied in machine learning and data mining problems. Linear subspace methods are the commonly used ones, such as principal component analysis (PCA), Fisher's linear discriminant analysis (FDA), common spatial pattern (CSP), et al. In this paper, we describe a novel feature extraction method for binary classification problems. Instead of finding linear subspaces, our method finds lower-dimensional affine subspaces satisfying a generalization of the Fukunaga–Koontz transformation (FKT). The proposed method has a closed-form solution and thus can be solved very efficiently. Under normality assumption, our method can be seen as finding an optimal truncated spectrum of the Kullback–Leibler divergence. Also we show that FDA and CSP are special cases of our proposed method under normality assumption. Experiments on simulated data show that our method performs better than PCA and FDA on data that is distributed on two cylinders, even one within the other. We also show that, on several real data sets, our method provides statistically significant improvement on test set accuracy over FDA, CSP and FKT. Therefore the proposed method can be used as another preliminary data-exploring tool to help solve machine learning and data mining problems.     

Determination of number of kanbans in a supply chain system via Memetic algorithm

In this research multi-stage supply chain system which is controlled by kanban system, is evaluated. In kanban system, decision making is based on determination of the number of kanbans as well as batch sizes. This paper attempts to model supply chain system with regard to costs under just-in-time (JIT) production philosophy. Since adopted model is of mixed integer non-linear programming (MINLP) type and solving it by exact algorithm such as branch and bound (B&B) takes a lot of time, a heuristic method via Memetic algorithm (MA) is presented. Some problems are solved by our proposed MA to illustrate its performance.     

A survey of energy-efficient context recognition systems using wearable sensors for healthcare applications

Human context recognition (HCR) from on-body sensor networks is an important and challenging task for many healthcare applications because it offers continuous monitoring capability of both personal and environmental parameters. However, these systems still face a major energy issue that prevent their wide adoption. Indeed, in healthcare applications, sensors are used to capture data during daily life or extended stays in hospital. Thus, continuous sampling and communication tasks quickly deplete sensors’ battery reserves, and frequent battery replacement is not convenient. Therefore, there is a need to develop energy-efficient solutions for long-term monitoring applications in order to foster the acceptance of these technologies by the patients. In this paper, we survey existing energy-efficient approaches designed for HCR based on wearable sensor networks. We propose a new classification of the energy-efficient mechanisms for health-related human context recognition applications and we review the related works in detail. Moreover, we provide a qualitative comparison of these solutions in terms of energy-consumption, recognition accuracy and latency. Finally, we discuss open research issue and give directions for future works.