Mapping urban land cover based on spatial-spectral classification of hyperspectral remote-sensing data

In this article, an innovative classification framework for hyperspectral image data, based on both spectral and spatial information, is proposed. The main objective of this method is to improve the accuracy and efficiency of high-resolution land-cover mapping in urban areas. The spatial information is obtained by an enhanced marker-based minimum spanning forest (MMSF) algorithm. A pixel-based support vector machine (SVM) algorithm is first used to classify the hyperspectral image data, then the enhanced MMSF algorithm is applied in order to increase the accuracy of less accurately classified land-cover types. The enhanced MMSF algorithm is used as a binary classifier. These two classes are the low-accuracy class and remaining classes. Finally, the SVM algorithm is trained for classes with acceptable accuracy. In the proposed approach, namely MSF-SVM, the markers are extracted from the classification maps obtained by both SVM and watershed segmentation algorithms, and are then used to build the MSF. Three benchmark hyperspectral data sets are used for the assessment: Berlin, Washington DC Mall, and Quebec City. Experimental results demonstrate the superiority of the proposed approach compared with SVM and the original MMSF algorithms. It achieves approximately 5, 6, and 7% higher rates in kappa coefficients of agreement in comparison with the original MMSF algorithm for the Berlin, Washington DC Mall, and Quebec City data sets, respectively.     

Artificial neural network training utilizing the smooth variable structure filter estimation strategy

A multilayered neural network is a multi-input, multi-output nonlinear system in which network weights can be trained by using parameter estimation algorithms. In this paper, a novel training method is proposed. This method is based on the relatively new smooth variable structure filter (SVSF) and is formulated for feed-forward multilayer perceptron training. The SVSF is a state and parameter estimation that is based on the sliding mode concept and works in a predictor–corrector fashion. The SVSF training performance is tested on three benchmark pattern classification problems. Furthermore, a study is presented comparing the popular back-propagation method, the extended Kalman filter, and the SVSF.     

Effects of slots on thermoelastic quality factor of a vertical beam MEMS resonator

Thermoelastic damping is one of the dominant mechanisms of structural damping in vacuum-operated microresonators. A three dimensional numerical model based on the finite element method is used for simulating thermoelastic damping in clamped–clamped microelectromechanical beam resonators. In this regards, both simple and slotted beam are considered. To understand the effect of slot positions and sizes on the resonator performance, resonant frequency and thermoelastic quality factor are calculated for both simple and slotted beams for a wide range of beam length from 10 to 400 µm. Punching slots in the resonator beam reduces the stiffness and mass of the beam which affect the resonant frequency. In addition thermo-mechanical coupling mechanisms of the resonator are affected by the slots which improve the thermoelastic quality factor. For most of the beam lengths, it is shown that the slots at the beam-anchor interface region, where the strain is high, are more effectively enhanced the thermoelastic quality factor than one at the centre of the beam region. However, the highest resonance frequency is achieved with the slots at the center region.     

Information fusion between short term learning and long term learning in content based image retrieval systems

Content based image retrieval (CBIR) systems could provide more precise results by taking the user’s feedbacks into account. Two types of the relevance feedback learning paradigms are short term learning (STL) and long term learning (LTL). By using both STL and LTL, a collaborative CBIR system is proposed in this paper. The proposed system introduced three fusion methods: including fusion in retrieved images, fusion in ranks, and fusion in similarities to make cooperation between STL and LTL. The proposed fusion methods are examined in a CBIR system equipped with a proposed statistical semantic clustering (SSC) method of LTL. The SSC method works based on the concept of semantic categories of the images by clustering techniques and constructing a relevancy matrix between images and semantic categories. The results of the SSC method with the suggested fusion methods are compared with two state-of-the-art LTL methods, namely virtual feature based method and dynamic semantic clustering. Comparative results confirm the efficiency of the proposed method. Furthermore, experimental results demonstrate that for a unique LTL method, various fusion methods lead to different results.     

Multiple kernel representation and classification of multivariate satellite-image time-series for crop mapping

Multivariate satellite-image time-series (MSITS) are a valuable source of information for a wide range of agricultural applications. Image classification, one of the main applications of this type of data, is a challenging task. It is mainly because MSITS are generated by a complex interaction among several sources of information, which are known as the factors of variation. These factors contain different information with different levels of relevance to a classification task. Thus, a proper representation of MSITS data is required in order to extract and model the most useful information from these factors for classification purpose. To this end, this article proposes three multiple kernel representations of MSITS data. These representations extract the most classification-related information from these data through combining the basis kernels constructed from different factors of variation of the MSITS data. In the proposed representations, the combination of the basis kernels was achieved by using the multiple kernel learning algorithms. The efficiency of the proposed multiple kernel representations was evaluated based both on analysing the relevance of their kernels to the classification task and their classification performances. Two different MSITS data sets composed of 10 RapidEye imageries of an agricultural area were used to evaluate the performances of the proposed methods. In addition, the classification results of both MSITS using a single kernel were considered as the baseline for comparison. The results showed an increase of up to 14% in overall accuracy of the classification maps by using the multiple kernel representations. Moreover, these particular representations for classification of time-series observations were able to handle the undesirable effects in image data such as the presence of clouds and their shadows.     

A New Algorithm to Design Minimal Multi‐Functional Observers for Linear Systems

Designing minimum possible order (minimal) observers for multi‐input multi‐output (MIMO) linear systems have always been an interesting subject. In this paper, a new methodology to design minimal multi‐functional observers for linear time invariant (LTI) systems is proposed. The approach is applicable, and it also helps in regulating the convergence rate of the observed functions. It is assumed that the system is functional observable or functional detectable, which is less conservative than assuming the observability or detectability of the system. To satisfy the minimality of the observer, a recursive algorithm is provided that increases the order of the observer by appending the minimum required auxiliary functions to the desired functions that are going to be estimated. The algorithm increases the number of functions such that the necessary and sufficient conditions for the existence of a functional observer are satisfied. Moreover, a new methodology to solve the observer design interconnected equations is elaborated. Our new algorithm has advantages with regard to the other available methods in designing minimal order functional observers. Specifically, it is compared with the most common schemes, which are transformation based. Using numerical examples it is shown that under special circumstances, the conventional methods have some drawbacks. The problem partly lies in the lack of sufficient numerical degrees of freedom proposed by the conventional methods. It is shown that our proposed algorithm can resolve this issue. A recursive algorithm is also proposed to summarize the observer design procedure. Several numerical examples and simulation results illustrate the efficacy, superiority and different aspects of the theoretical findings.     

Microstructural study of synthesized polyethylenes by homogeneous and heterogeneous nickel α-diimine catalysts

In this work, ethylene polymerization was investigated by using homogeneous and heterogenouse nickel α-diimine catalysts [1,4-bis(2,6-diisopropylphenyl) acenaphthene diimine nickel(II) dibromide]. Methyl aluminoxane (MAO) and triethyl aluminum (TEA) were used as cocatalysts in homogenous and heterogeneous polymerizations, respectively. The heterogeneous catalyst showed lower activity than its homogeneous equivalent. The influence of polymerization temperature and heterogenization conditions was studied on the microstructure properties of the prepared polymers. Increasing polymerization temperature (T _P) up to 50 °C decreased the activity of both homogenous (LN) and heterogeneous (LNS) nickel α-diimine catalysts. The highest activities were 1286 and 982 kg PE (mol Ni bar h)^?1 obtained at T _P = 30 °C for LN and LNS catalysts, respectively. The polymer samples obtained by supported catalyst (LNS) showed lower unsaturation contents. Moreover, DSC analysis did not show any melting peaks for polymers obtained by LN catalyst due to their amorphous structure, which was confirmed by XRD analysis. The microstructure of the prepared polymers was completed by successive self-nucleation annealing (SSA) and was investigated by ¹³C NMR studies. The SSA thermogram of samples made by LNS catalyst exhibited several crystal types with different lamella thicknesses. The branches in polyethylene samples produced by homogenous catalyst were higher and showed more diversity. The total methyl branch percentages for both LN and LNS catalysts were 13.1 and 3.4%, respectively.     

Effect of different additives and their mixtures on thermal stability of polyethylene synthesized using iron and titanocene hybrid catalysts

Traditionally, additives are introduced into a polymer matrix via extrusion process which consumes a high amount of energy. In this study, the use of different additives including antioxidants for the in‐reactor stabilization of polyethylene has been investigated in order to provide an energy saving system. Particular attention was dedicated to the efficiency of antioxidant influencing the catalysts activity and properties of polymers. The effect of the addition of Irganox 1330 and Irgafos 168 antioxidants and zinc stearate on the activities of metallocene, post‐metallocene, and their supported hybrid were studied. In addition, the effect of different additives on the thermal characteristics of the synthesized polymers and oxidative induction time (OIT) was evaluated. Our polymerization results exhibited that the factors such as chemical structure of antioxidant and its steric hindrance, type of catalysts, and their hybrid could affect the catalyst performance and OIT contents. The use of antioxidants mixture and hybrid of catalysts is a way that can increase oxidation resistance of polymers considerably. © 2017 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2017 , 134, 45482.