Transreceiving of encrypted medical image – a cognitive approach

Multimedia Tools and Applications - Recently, there is an increasing demand for efficient and secure transreception of medical images in telemedicine applications. Though a fixed spectrum is...     

Early detection of human actions—A hybrid approach

Early detection of human actions is essential in a wide spectrum of applications ranging from video surveillance to health-care. While human action recognition has been extensively studied, little attention is paid to the problem of detecting ongoing human action early, i.e. detecting an action as soon as it begins, but before it finishes. This study aims at training a detector to be capable of recognizing a human action when only partial action sample is seen. To do so, a hybrid technique is proposed in this work which combines the benefits of computer vision as well as fuzzy set theory based on the fuzzy Bandler and Kohout's sub-triangle product (BK subproduct). The novelty lies in the construction of a frame-by-frame membership function for each kind of possible movement. Detection is triggered when a pre-defined threshold is reached in a suitable way. Experimental results on a publicly available dataset demonstrate the benefits and effectiveness of the proposed method.     

Integrating information — a task-orientated approach

To date, there have been few attempts at integrating the increasing amounts of information that are available to users of computer systems. Using published literature, this paper collates some relevant principles (or heuristics) for integrating information based on task orientation. These are then applied to the electronic fishing aids in an in-shore fishing vessel's wheelhouse with the purpose of reducing the number of screens used therein. However, the application raises some issues; in particular that of the designer's knowledge of the tasks so that unique information is not lost.     

An adaptive gravitational search optimization (AGSO)-based convolutional neural network—long short-term memory (CNN-LSTM) approach for face recognition and classification

Face recognition and classification have gained increasing attraction in the recent decades due to their widespread adoption in real time application systems. Most of the conventional research efforts focused on developing face recognition frameworks using enhanced optimization-based classification methods, they are hampered by issues such as computational complexity, increased overhead, limited capacity to handle large datasets, and lengthy processing time. The novel contribution of this paper is to develop a highly competent and precise face recognition methodology through an innovative mechanism. In this framework, the initial step involves face detection from input images using an analytical face parts detection methodology. Subsequently, the tutor face filtering (TFF) technique is applied to preprocess the face image, enhancing its quality and filtering out noise content. Following this preprocessing step, features are extracted from the processed image using the direction-based pattern extraction (DBPE) model. To improve classifier accuracy, a novel adaptive gravitational search optimization (AGSO) technique is employed to select the optimal features during model training. Finally, an integrated deep learning model, referred to as convolutional neural network — long short-term memory (LSTM), is utilized for accurate face image recognition based on the selected optimal features. To assess and compare the system's performance, various metrics are employed in the results analysis to demonstrate the superiority of the proposed approach.     

Voice recognition — still a long way to go

The article describes the state of voice recognition and discusses the implementation of a ‘hybrid’ voice input system combining voice input with a traditional means of data input, e.g. a keyboard, in a system for the disabled.     

VIPS — a digital image processing algorithm development environment

The major requirements of an image processing algorithm development system are presented. VIPS, a Vax-based image processing system developed at the University of Canterbury, New Zealand, is described and discussed in terms of algorithm development. Some of the applications of VIPS are listed.     

Improving enterprise system support—a case-based approach

Many organisations have implemented or are implementing large enterprise systems, like ERP and PDM, for integrating their business functions and streamlining the flow of information. Implementing such systems is very complex, however. In many companies the results envisioned have not or have only partly been achieved. In the literature many explanations can be found for insufficient or failing implementation efforts. A large part of the problems encountered appears to be human and organisational in nature, while it has been stressed that implementing an enterprise system requires and involves organisational change.Success and failure factors and do's and don'ts, as can be found in the literature, provide necessary, but not sufficient, preconditions for starting an implementation project. The complexity of such a project makes full anticipation and control of potential problems impossible. An organisation needs to be prepared to encounter disturbances and take corresponding actions. Knowledge on the dynamics of enterprise system implementation processes is, however, scarce and scattered. Moreover, to what extent the context in which an enterprise system is implemented influences implementation is not yet fully known.In this article, an approach is presented to gather knowledge on implementation process dynamics. The approach builds on theoretical and practical contributions in search for a structured human and organisational approach. The knowledge gathered has been analysed by means of a comprehensive reference framework. After this step a tool has been developed to support a consultant in assessing the maturity of a company to start an enterprise system implementation project. The knowledge, used directly in the tool, makes it possible to suggest improvement actions sensitive to the situation and context. Through statistical analysis, interesting context differences have been identified, which may also help in further refining the advice. Initial validation results are promising.     

Simultaneous decoupling and disturbance rejection—a structural approach

The combined problem of disturbance rejection and decoupling for linear systems is considered. New necessary and sufficient conditions are derived for the existence of a state feedback controller that rejects the disturbances and simultaneously decouples the closed-loop system. These conditions are simple, intuitively appealing and are expressed in terms of system structural invariants such as infinite zero orders. Then we consider the disturbance rejection with simultaneous decoupling problem for structured linear systems. Necessary and sufficient conditions for the generic solvability of this problem are provided. They are expressed in terms of input-output path length comparisons on the system associated graph.     

Control of robotic manipulators—a neural network approach

A neural-network-based scheme is used for the control of a robotic manipulator. The main idea is that, by using a neural network to learn the characteristics of the robot system (or specifically its inverse dynamics), accurate trajectory following and good performance results are obtained. However, the traditional back-propagation algorithm commonly used for control and identification of nonlinear systems suffers from a slow rate of convergence. We investigate the effect of adusting the slope of the activation function (the node nonlinearity) on the performance of a back-propagation algorithm. It is shown that learning speed is increased significantly by making the slope of non-linearity adaptive. The results demonstrate that the proposed method gives better error minimization and faster convergence. The suggested method is applied to a two-link robotic manipulator. The resulting controller is sufficiently robust with respect to the changing conditions.     

Computer aided training in a renal dialysis ward—a novel approach

This paper describes a novel approach to training renal dialysis patients in the use and management of life-saving kidney dialysis machines.The product of this approach has been the development of a low-cost sophisticated, audio-enhanced, hardware/software tool. It can be used by teachers and workers in many areas and disciplines, although the tool was originally developed for the world of renal medicine.The tool provides a structured environment in which lessons can be created quickly and cheaply. The dubbing of audio onto this lession is achieved in a similar structured manner.     

V-DOM and P-XML—towards a valid programming of XML-based applications

Many WWW applications generate hypertext markup language or extensible markup language (XML) documents dynamically. Current tools, however, like languages using document object model (DOM) or JAVA Server Pages do not allow to check the validity of the generated documents statically. Instead, validity has to be ‘checked’ dynamically by appropriate test runs. This paper addresses this problem by introducing a new distinct class for each element type of a document type definition (DTD). Each class extends the Element-class of the DOM. The resulting object model is called validating DOM (V-DOM). Parametric XML (P-XML) is an extension of XML allowing to insert values of the newly defined classes in places, where the corresponding element types are allowed according to the underlying DTD. Like V-DOM, P-XML guarantees the validity of all XML documents generated by using P-XML. V-DOM and P-XML are illustrated by several examples.     

Efficient magnetohydrodynamic simulations on distributed multi-GPU systems using a novel GPU Direct–MPI hybrid approach

Modern graphics processing units (GPUs) have been widely utilized in magnetohydrodynamic (MHD) simulations in recent years. Due to the limited memory of a single GPU, distributed multi-GPU systems are needed to be explored for large-scale MHD simulations. However, the data transfer between GPUs bottlenecks the efficiency of the simulations on such systems. In this paper we propose a novel GPU Direct–MPI hybrid approach to address this problem for overall performance enhancement. Our approach consists of two strategies: (1) We exploit GPU Direct 2.0 to speedup the data transfers between multiple GPUs in a single node and reduce the total number of message passing interface (MPI) communications; (2) We design Compute Unified Device Architecture (CUDA) kernels instead of using memory copy to speedup the fragmented data exchange in the three-dimensional (3D) decomposition. 3D decomposition is usually not preferable for distributed multi-GPU systems due to its low efficiency of the fragmented data exchange. Our approach has made a breakthrough to make 3D decomposition available on distributed multi-GPU systems. As a result, it can reduce the memory usage and computation time of each partition of the computational domain. Experiment results show twice the FLOPS comparing to common 2D decomposition MPI-only implementation method. The proposed approach has been developed in an efficient implementation for MHD simulations on distributed multi-GPU systems, called MGPU–MHD code. The code realizes the GPU parallelization of a total variation diminishing (TVD) algorithm for solving the multidimensional ideal MHD equations, extending our work from single GPU computation (Wong et al., 2011) to multiple GPUs. Numerical tests and performance measurements are conducted on the TSUBAME 2.0 supercomputer at the Tokyo Institute of Technology. Our code achieves 2 TFLOPS in double precision for the problem with 1200³ grid points using 216 GPUs.     

Certificate policies and certification practice statements — a practical approach

The IETF [1], RFC2527 [2] describes and defines the contents of documentation that is at the heart of the process-driven elements of a PKI. These are captured in the certificate policy (CP), which defines what is delivered, and the certification practice statement (CPS), which defines how it is delivered. The RFC has formed the basis for the contents of both documents. However, with the move of PKI from employment via service providers to deployment directly by organisations, and the extension in the use of PKI to a much wider and less knowledgeable potential market, the difficulties of creating an appropriate set of documentation have become apparent. While I do not suggest that RFC2527 is wrong, I argue that it appears to pose as many questions as it answers when it comes to implementation. In this article I attempt to consider the factors that influence the implementation of a documented infrastructure and suggest a pragmatic approach, based on RFC2527, to delivering a document structure that will support the delivery of PKI services and the understanding of trust that can be placed in those services.     

Velocity planning for a mobile robot to track a moving target — a potential field approach

Potential field method has been widely used for mobile robot path planning, but mostly in a static environment where the target and the obstacles are stationary. The path planning result is normally the direction of the robot motion. In this paper, the potential field method is applied for both path and speed planning, or the velocity planning, for a mobile robot in a dynamic environment where the target and the obstacles are moving. The robot’s planned velocity is determined by relative velocities as well as relative positions among robot, obstacles and targets. The implementation factors such as maximum linear and angular speed of the robot are also considered. The proposed approach guarantees that the robot tracks the moving target while avoiding moving obstacles. Simulation studies are provided to verify the effectiveness of the proposed approach.     

Real-time object recognition on image sequences with the adaptable time delay neural network algorithm — applications for autonomous vehicles

Within the framework of the vision-based “Intelligent Stop&Go” driver assistance system for both the motorway and the inner city environment, we present a system for segmentation-free detection of overtaking vehicles and estimation of ego-position on motorways as well as a system for the recognition of pedestrians in the inner city traffic scenario. Both systems are running in real-time in the test vehicle UTA of the DaimlerChrysler computer vision lab, relying on the adaptable time delay neural network (ATDNN) algorithm. For object recognition, this neural network processes complete image sequences at a time instead of single images, as it is the case in most conventional neural algorithms. The results are promising in that using the ATDNN algorithm, we are able to perform the described recognition tasks in a large variety of real-world scenarios in a computationally highly efficient and rather robust and reliable manner.     

The flooding sink — a new approach to local area networking

A new local area networking technology is presented. The approach is based on an old routing algorithm called flooding — forward messages to all neighbouring nodes. The problem with this algorithm is that the network is deluged with duplicate messaes. The solution is a simple device which uses local memory to detect and ignore redundant messages, thus also acting as a message sink. Networks based on this device can be more flexible and reliable than current networks. Flooding also has the advantage that any messages lost due to transmission errors are quickly replaced by one of the copies. This can make low-level protocols unnecessary. When the low-level protocols are omitted, significant performance improvements are achieved. Simulation results are presented which show that this flooding technology can perform better than current CSMA and ring technologies.