A multilayer incremental neural network architecture for classification

A new multilayer incremental neural network (MINN) architecture and its performance in classification of biomedical images is discussed. The MINN consists of an input layer, two hidden layers and an output layer. The first stage between the input and first hidden layer consists of perceptrons. The number of perceptrons and their weights are determined by defining a fitness function which is maximized by the genetic algorithm (GA). The second stage involves feature vectors which are the codewords obtained automaticaly after learning the first stage. The last stage consists of OR gates which combine the nodes of the second hidden layer representing the same class. The comparative performance results of the MINN and the backpropagation (BP) network indicates that the MINN results in faster learning, much simpler network and equal or better classification performance.     

Modelling of the spring-in phenomenon in curved parts made of a thermosetting composite

In this study, a simple two step finite element model is developed to predict the spring-in of C-sections parts made of AS4/8552 composite. The development of resin properties throughout the MRCC were derived using Group Interaction Modelling and the mechanical properties of the composite predicted by using micromechanical models. Important mechanisms during manufacturing that are effective in the formation of residual stresses and shape distortions are defined. The finite element method implemented is composed of two steps before and after the vitrification of the resin. Vitrification is treated as a point at which the material suddenly changes from the rubbery to glassy state with constant properties in each state. The spring-in angles predicted by the finite element analysis are compared to the angles measured on C-section specimens of various lay-ups and thicknesses. The correlation is good showing the validity of the assumptions adopted.     

The impact of a realistic packet traffic model on the performance of surveillance wireless sensor networks

It is quite common to see that classical periodic or Poisson packet traffic models are used for evaluating the performance of wireless sensor networks (WSNs). However, these models may not be appropriate for modeling the data traffic resulting from a particular application. Furthermore, they may be overestimating the performance of a WSN. In this paper, we show the significance of using a realistic and application-specific packet traffic model by comparing the performance of a well-known WSN protocol under the Surveillance WSN packet traffic model (SPTM), as well as under periodic and binomial traffic models. A packet traffic framework specific to surveillance applications is proposed which is then used for deriving SPTM analytically. In order to be adaptable and flexible, SPTM incorporates a probabilistic and parametric sensor detection model. Simulation results show that to employ an application-specific packet traffic model has significant impact on the performance evaluation of the WSN and ordinary traffic models may overestimate the capacity of the WSN.     

Co-Oxidative Transformation of Piperine to Piperonal and 3,4-Methylenedioxycinnamaldehyde by a Lipoxygenase from Pleurotus sapidus

The valuable aroma compound piperonal with its vanilla-like olfactory properties is of high interest for the fragrance and flavor industry. A lipoxygenase (LOX_Psa1) of the basidiomycete Pleurotus sapidus was identified to convert piperine, the abundant pungent principle of black pepper (Piper nigrum), to piperonal and a second volatile product, 3,4-methylenedioxycinnamaldehyde, with a vanilla-like odor through an alkene cleavage. The reaction principle was co-oxidation, as proven by its dependence on the presence of linoleic or α-linolenic acid, common substrates of lipoxygenases. Optimization of the reaction conditions (substrate concentrations, reaction temperature and time) led to a 24-fold and 15-fold increase of the piperonal and 3,4-methylenedioxycinnamaldehyde concentration using the recombinant enzyme. Monokaryotic strains showed different concentrations of and ratios between the two reaction products.     

Drilling detritus and the operating parameters of thermally stable PDC core bits

A single rock type was drilled using two types of thermally stable polycrystalline diamond compact (PDC) and an impregnated diamond core bit using a fully instrumented laboratory drilling rig at a fixed rotational speed, and over a range of weights on bit (WOB). Operating parameters of the bits such as WOB, rate of penetration (ROP), rotational speed (RPM), torque and drilling specific energy (SE) were continuously monitored during the drilling trials. The effects of these parameters on the performance of the bits were examined. Relations between the drilling variables are also described. At each set WOB for each bit, the drilling detritus were collected. The detritus sizes between 1180 and 53 microns were analysed using wet sieving and sub 53 micron was analysed using a Malvern particle size machine. Both data were combined to obtain a particle size distribution at set WOBs for each bit. The relations between the particle size of the drilling detritus and operating parameters, cutter size and wear of the PDC and impregnated diamond core bits, and original rock grain size were established. An increase in WOB, ROP, wear rates, pins or diamond size increased the drilling detritus size for the bits. No relation was found between the drilling detritus and the rock grain size due to the regrinding effect and the high proportion (> 30%) of matrix material in the rock.     

Fast algorithms for the discrete Fourier preprocessing transforms

It is pointed out that the two-stage Fourier algorithms are useful in a variety of applications in signal/image processing and recognition. The first stage, known as the DFPT (discrete Fourier preprocessing transform), has the potential of very fast and low-cost implementation, say, in a VLSI chip, as well as high-quality performance in applications without the second stage. The DFT (discrete Fourier transform), the RDFT (real DFT), or any other DFPT can be obtained if the second stage is completed. Since the second stage consists of independent blocks of circular correlations, they can be computed in fast sequential or parallel architectures. There are two groups of fast algorithms for the computation of DFPTs. The first group involves the representation of a DFPT in terms of skew-circular correlations (SCCs), which are then computed by fast SCC algorithms or parallel architectures such as semisystolic arrays. The second group involves the fast computation of a DFPT, including the DFT and the RDFT, by the implementationally simplest DFPT, such as class 2, case V DFPT     

Image coding with the discrete cosine-III transform

The discrete cosine-III transform (DC3T) is the same as the discrete symmetric cosine transform (DSCT) with a specific preprocessing of input data. It has less computational complexity than the discrete cosine transform (DCT) in terms of multiplications. The DC3T is also related to the DCT by a weighting matrix. The performance of the DC3T is compared to the DCT when compression is performed by adaptive coding. Experimental results show that a significant improvement in visual performance and mean square reconstruction error can be achieved over what is possible with the DCT. The better performance is attributed to the weighting of the DCT coefficients, which is indirectly achieved at reduced computational cost     

Evaluation of destruction specific energy of fly ash and slag admixed concrete interlocking paving blocks (CIPB)

Concrete interlocking paving blocks (CIPB) are widely used in flooring and covering applications such as city roads, pedestrian foots, gardening architecture and many surface coverings. The focus of this study is to evaluate destruction specific energy (SE_des) of CIPB containing fly ash and blast-furnace slag at different curing conditions. The SE_des is estimated from the area (integral) under stress–strain envelope for an unconfined compression test. This method provides a better understanding between performance and the main mechanical character of CIPB. Compressive strength, splitting tensile and flexural tensile strengths, rebound hardness and abrasion resistance tests at 3, 7, 28, 90 and 180 days were carried out. The test results showed that fly ash (FA) and blast-furnace slag (BFS) can replace the cement by amounts of 10–20%, 30–40% of mass, respectively. This helps to produce CIPB with lower cost and better engineering properties. The relationship between SE_des, strength, hardness and abrasion are defined for CIPB.