Finite Gamma mixture modelling using minimum message length inference: application to SAR image analysis

This paper discusses the unsupervised learning problem for finite mixtures of Gamma distributions. An important part of this problem is determining the number of clusters which best describes a set of data. We apply the Minimum Message Length (MML) criterion to the unsupervised learning problem in the case of finite mixtures of Gamma distributions. The MML and other criteria in the literature are compared in terms of their ability to estimate the number of clusters in a data set. The comparison utilizes synthetic and RADARSAT SAR images. The performance of our method is also tested by contextual evaluations involving SAR image segmentation and change detection.     

Evaluation using online support-vector-machines and fuzzy reasoning. Application to condition monitoring of speeds rolling process

A method for process condition monitoring and evaluation, which combines the online support vector machine (SVM) regression and the fuzzy sets methods, is proposed. To account for the time dependence, the proposed approach is based on moving windows in order to take into account the past and new data for the model’s adaptation. The fuzzy analysis is then applied to the generated residual data to give an evaluation of the condition monitoring. The proposed approach is applied to hot rolling for constructing a complementary condition monitoring system, which permits an online quality evaluation in the rolling process. Simulation results based on residual data show that the new approach is easily implementable.     

An automatic procedure for identifying alternative automobile ingress movements in young and elderly populations with or without prostheses

This paper presents an automatic numerical procedure for classifying the automobile ingress movements in young and elderly populations with or without prostheses. This procedure proceeds in two stages. The first stage works to identify the main movement strategies by classifying the subjects according to new indices, called 3BJCV (3 Bases Joint Contribution Vector). These indices make it possible to take the complexity of the automobile ingress movement into account. They quantify the contribution of the lower limbs chains angles' and those of the chain gathering the trunk and head in the final positioning of the midpoint between the two hips (MPH). The second stage specifies the main movement strategies obtained in stage 1 by classifying the subjects according to the relevant articular variables of the ingress movement.This two-stage procedure was applied to the ingress movements extracted from the national HANDIMAN (RNTS, 2004) project, measured in experiments carried out with a small vehicle and a large vehicle (i.e., a minivan). The two main ingress movement strategies were identified: a one-foot ingress strategy and a two-foot ingress strategy. In all, 6 ingress sub-strategies were highlighted automatically, without requiring any special expertise. Five of these sub-strategies had already been observed, based on the same experiments, in the past. The 6th sub-strategy was used by subjects who were divided into 2 other sub-strategy groups while of the observation.

Relevance to industry

The procedure proposed in this study can help ergonomists and car manufacturers to quantify and classify the automobile ingress movements without requiring any a priori expertise on the movement. It also helps to improve any given numerical simulation by providing numerical data characterizing the different ingress movement profiles.     

Active contours for video object tracking using region, boundary and shape information

In this paper, we propose a robust model for tracking in video sequences with non-static backgrounds. The object boundaries are tracked on each frame of the sequence by minimizing an energy functional that combines region, boundary and shape information. The region information is formulated by minimizing the symmetric Kullback–Leibler (KL) distance between the local and global statistics of the objects versus the background. The boundary information is formulated using a color and texture edge map of the video frames. The shape information is calculated adaptively to the dynamic of the moving objects and permits tracking that is robust to background distractions and occlusions. Minimization of the energy functional is implemented using the level set method. We show the effectiveness of the approach for object tracking in color, infrared (IR), and fused color-infrared sequences.     

Enhanced Dynamic Adhesion in Nematic Liquid Crystal Elastomers

Smart adhesives that undergo reversible detachment in response to external stimuli enable a wide range of applications in household products, medical devices, or manufacturing. Here, a new model system for the design of smart soft adhesives that dynamically respond to their environment is presented. By exploiting the effect of dynamic soft elasticity in nematic liquid crystal elastomers (LCE), the temperature‐dependent control of adhesion to a solid glass surface is demonstrated. The adhesion strength of LCE is more than double in the nematic phase, in comparison to the isotropic phase, further increasing at higher detachment rates. The static work of adhesion, related to the interfacial energy of adhesive contact, is shown to change very little within the explored temperature range. Accordingly, the observed enhanced adhesion in the nematic phase is primarily attributable to the increased internal energy dissipation during the detachment process. This adhesion effect is correlated with the inherent bulk dynamic‐mechanical response in the nematic LCE. The reported enhanced dynamic adhesion can lead to the development of a new class of stimuli‐responsive adhesives.     

Hybrid method combining superpixel,supervised learning,and random walk for glioma segmentation

Currently, the analysis of magnetic resonance imaging (MRI) brain images of pathological patients is performed manually, both for the recognition of brain structures or lesions and for their characterization. Physicians sometimes encounter difficulties in interpreting these images for a reliable diagnosis of the patient's condition. This is due to the difficulty of detecting the nature of the lesions, particularly glioma. Glioma is one of the most common tumors, and one of the most difficult to detect because of its shape, irregularities, and ambiguous limits. The segmentation of these tumors is one of the most crucial steps for their classification and surgical planning. This article presents a new, accurate, and automatic approach for the precise segmentation of early gliomas (benign tumors), combining the random walk (RW) algorithm and the simple linear iterative clustering algorithm. The study was carried out in four steps. The first step consisted of decomposing the image into superpixels to obtain an initial outline of the tumor. The superpixels were generated using the SLIC algorithm. In the second step, for each superpixel, a set of statistical and multifractal characteristics were calculated (gray‐level co‐occurrence matrix, multifractal detrending moving average). In the third step, the superpixels were classified using a supervised random forest (RF) type classier into healthy or tumorous brain tissue. In the final step, the contour of the detected tumor was enhanced using the customized RW algorithm. The proposed method was evaluated using the Brain Tumor Image Segmentation Challenge 2013 database. The results obtained are competitive compared to other existing methods.     

Traffic Priority-Aware Medical Data Dissemination Scheme for IoT Based WBASN Healthcare Applications

Wireless Body Area Sensor Network (WBASN) is an automated system for remote health monitoring of patients. WBASN under umbrella of Internet of Things (IoT) is comprised of small Biomedical Sensor Nodes (BSNs) that can communicate with each other without human involvement. These BSNs can be placed on human body or inside the skin of the patients to regularly monitor their vital signs. The BSNs generate critical data as it is related to patient's health. The data traffic can be classified as Sensitive Data (SD) and Non-sensitive Data (ND) packets based on the value of vital signs. These data packets have different priority to deliver. The ND packets may tolerate some delay or packet loss whereas, the SD packets required to be delivered on time with minimized packet loss otherwise it can be life threating to the patients. In this research, we propose a Traffic Priority-aware Medical Data Dissemination (TPMD²) scheme for WBASN to deliver the data packets according to their priority based on the sensitivity of the data. The assessment of the proposed scheme is carried out in various experiments. The simulation results of the TPMD² scheme indicate a significant improvement in packets delivery, transmission delay and energy efficiency in comparison with the existing schemes.     

A Classification Approach for an Accurate Analog/RF BIST Evaluation Based on the Process Parameters

Specifications of Radio Frequency (RF) analog integrated circuits have increased strictly as their applications tend to be more complicated and high test cost demanding. This makes them very expensive due to an increased test time and to the use of sophisticated test equipment. Alternative test measures, extracted by means of Built-In Self Test (BIST) techniques, are useful approaches to replace standard specification-based tests. One way to evaluate the efficiency of the CUT measures at the design stage is by estimating the Test Escapes (T _E ) and the Yield Loss (Y _L ) at ppm level. Unfortunately, an important number of Monte Carlo simulations must be run in order to guarantee their accuracy. For certain types of circuits, this requires many months or even years to generate millions of circuits. To overcome this limitation, we present in this paper a new technique where a small number of simulations is sufficient to reach an important precision. This method is based on a classification using machine learning methods, such as SVM and Neural Networks based classifiers to determine pass/fail regions. The proposed approach requires a few number of simulations only to determine the region separating the process parameters generating good and faulty, or pass and fail circuits. Then only this region is needed to estimate the test metrics without running any additional simulation. The proposed methodology is illustrated for the evaluation of a filter BIST technique.     

Analytical Analysis of Applying Packet Fragmentation Mechanism on Both Basic and RTS/CTS Access Methods of the IEEE 802.11b DCF Network Under Imperfect Channel and Finite Load Conditions

The mathematical modeling and performance evaluation of the IEEE 802.11 network in all its various extensions (802.11b, 802.11a, 802.11g, 802.11e, 802.11n, etc.) have already been widely explored over the past years. However, the Packet Fragmentation Mechanism (PFM), which is proposed by the IEEE work group to enhance the MAC sub-layer of the IEEE 802.11 standard in an error-prone channel, has been missed in the available literature. Yet, the PFM is the only existing solution to reduce the influence of bit error rate and the length of data packets on the packet error rate, and consequently on the performances of IEEE 802.11 networks. In this paper, we propose a new three-dimensional Markov chain in order to model, for the first time in the literature, the PFM in both Basic and RTS/CTS access methods of the IEEE 802.11b DCF network under imperfect channel and finite load conditions. Then, we develop mathematical models to derive a variety of performance metrics, such as: the overall throughput, the average packet delay successfully transmitted, the average packet drop time, the delay jitter and the packet delay distribution. Performance analysis of applying PFM on both Basic and RTS/CTS access methods of the IEEE 802.11b DCF network under imperfect channel and finite load conditions shows original results and leads to new conclusions that could not be intuitively expected.     

Purification and characterisation of milk‐clotting and caseinolytic activities of pepsin isolated from adult sheep abomasa

Ovine pepsin was isolated and assessed for its milk‐clotting properties and caseinolytic activity in comparison with commercial chymosin. Ovine pepsin showed similar responses to variations in pH, temperature and CaCl₂ concentration of milk compared with chymosin, although its pH sensitivity was higher. SDS‐PAGE electrophoretic analysis of the casein fractions treated with ovine pepsin showed that alpha‐casein was more susceptible to proteolysis than beta‐casein, in contrast to chymosin. Curd‐firming properties of skim milk gels obtained with ovine pepsin and chymosin were evaluated by Gelograph under the same conditions. Curd produced using ovine pepsin was less firm than that made with chymosin.