ALBERT

Description: 〈h3〉Abstract〈/h3〉 〈p〉Electrical impedance tomography (EIT) is a technique that is able to deliver valuable information on conductivity distribution in a process. An extension to conventional EIT is electrical impedance spectro-tomography (EIST) which is offering spectroscopic information. Along with methodological advancement of EIST and to meet practical requirements, there are ongoing researches to develop proper high-quality EIST hardware. An important part of EIST hardware is voltage controlled current source which its design is appeared to be a challenging task. This paper addresses briefly the aspects of EIST system and discusses the effect of various factors and their joint impacts on commonly used Howland voltage controlled current sources for EIST application. To perform realistic simulations, practical load and electrode model are applied within sensing bandwidth. The simulation results revealed the considerable impact of resistors tolerance on reducing the output impedance of the current sources in low frequencies while increasing the operating frequency significantly reduces the impact of tolerance. The simulation also demonstrates that the load current relative error is directly proportional to the resistors’ tolerance and its value approaches the ideal state, with zero-tolerance resistors, by increasing operating frequency. This point can be employed to determine the minimum operating frequency in which the impact of resistors tolerance can be canceled. The results imply that by using resistors with a lower tolerance, it is possible to achieve wider bandwidth in EIST. It can also be concluded that dual op-amp Howland current source is a suitable choice for EIST application.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉In this paper, we performed an experiment to compare the observation performance of color CCD camera and the distance measurement characteristics of the eye-safe laser range scanner in a non-visible environment, which simulated atmospheric environments in the reactor building when a severe accident occurred at the nuclear power plant. For the comparison experiments, we fabricated a fog room. Fog particles are sprayed into the fog room until they reach a certain concentration at the time of the severe accident. In this paper, different methods are used to evaluate the performance of the camera and the eye-safe laser range scanner. To verify the observation performance of the color CCD camera under the dense aerosol environments, we calculated contrast from the observation image captured by the camera. And to evaluate characteristics of the eye-safe laser range scanner module, we reviewed the range measurement error under the same visibility constraint situations. In comparative tests, the CCD camera used is a general purpose model with a high magnification zoom lens. We evaluated a model with an effective measuring distance of 30 m, referring to the eye-safe laser range scanner module mounted on the Quince robot III. A visibility was introduced as quantitative evaluation parameter for comparing the monitoring performance of the color CCD camera and range measurement error of an eye-safe LRS module under aerosol environments where the fog particles are sprayed in the fog room. The visibility was calculated by attenuation of the He–Ne laser intensity in the fog room where the fog particles are being sprayed.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉The application of modern visualization techniques (e.g., isosurface rendering) has been beneficial for the 3D presentation of ground penetrating radar (GPR) datasets, compared to the conventional time-slice imaging method, in order to facilitate GPR data interpretation. In this case study, we concentrate on how to apply isosurface rendering technique to easily interpret and highlight the spatial variations within GPR data in archaeological applications. Isosurface renders are generated from surfaces with equal amplitudes, which are illuminated in the 3D data volume using artificial light that reflects off the surface, where the surfaces appear as 3D shaded images. In this paper, GPR data were acquired with a GSSI SIR2000 system using a 200 MHz shielded antenna and 0.5 m inline spacing in both x- and y-directions at Al-Nadura Temple archaeological site. The main objective of this case study was to image subsurface archaeological structures and potentially characterize cut-off ancient burial tombs inside ferrous sandstone situated at Al-Nadura temple, which date back to the Romans periods in Egypt. According to this case study, we highlighted the importance of isosurface rendering technique analyses towards more intensive GPR data interpretation and target identification. Moreover, by comparing extracted isosurface data with conventional presented time-slice, we demonstrate the radar important interpretability of the advanced visualization methods for precisely characterizing subsurface archaeological structures.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉Mechanical resonators are rarely reported to be used in liquid environment due to high damping. Most applications of mechanical resonators are related to their use in the gaseous environment. In this paper, a novel mass detective biosensor is proposed for Lab-On-a-Chip applications based on micromechanical resonator array. The sensor can be fabricated by MEMS technology such as standard Polysilicon Multi User MEMS Process (PolyMUMPs). The sensor consists of chained mechanically coupled resonators with electrostatic actuating elements and capacitive sensing elements on both sides. The sensor is suitable for digital microfluidic applications which an under test droplet can be led over the resonator array and affect the frequency or the phase delay of vibration. Thus, both sides of the array containing the actuating and sensing elements stay away from the droplet. It provides conditions that do not electrically interfere with the sample droplet. The vibration direction of resonators under liquid droplet is such that causes light slide film damping and does not cause strong squeeze film damping. The commercial finite element tool COMSOL Multiphysics is used to analyze the vibration behavior of proposed sensor and verify its functionality. The results are presented for the frequency response and mass sensitivity of the biosensor.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉This paper discusses a novel theoretical treatment for facilitating displacement of a micro-scale liquid marble. It can be beneficial in designing microsensors as liquid-state micro accelerometers. In this work, the marble has been placed on the substrate under which the finger electrodes are located. The marble has been bounded with another rigid plate above it. To overcome the complexities of the marble displacement, initially, it has been deformed utilizing the electric field. This deforming electric field is generated via applying a sufficient DC voltage to the top, and bottom fixed plates. The electrostatic manipulating force exerting on the marble opposes the marble’s gravity, reduces its contact zone area and eventually causes the marble to get spherical form approximately. For displacing the marble, another electric and considerably weaker field than deforming one is applied to the marble via finger electrodes. The displacing electric field, which causes the marble to roll, is created via applying a DC voltage to the finger electrodes underneath the substrate. The displacing potentials are set to be on and off regularly to accelerate the movement of the marble. It has been shown that for a liquid marble with an effective radius of 1 μm, applying the approximate value of 42 V can deform the marble to a spherical, shaped one. It was also indicated that the air friction force exerting on the marble during its displacement leads to the constant-velocity motion of the marble eventually.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉Demosaicking refers to the image processing of reconstruction full color image from the raw incomplete samples recorded by a single-chip image sensor. For most available demosaicking methods, the image edges are important cues to design the interpolation scheme, and the accuracy of edge estimation has great influence on the reconstructed image quality. In this paper, a block edge estimation method is proposed by considering all the color channels comprehensively. Based on the novel edge estimation method, the proposed algorithm firstly interpolates the G channel pixels with the guidance of the color correlation and edge directions. The interpolated G channel is further used to help the R and B channel interpolations in sequence. In addition, a border pixels interpolation strategy is also presented to overcome the difficulties of the gradient estimation at border positions. Both Kodak and IMAX data set are used to test the validation of the proposed methods. The experimental results demonstrate that the proposed algorithm provides superior performances in terms of both objective evaluate and subjective visual quality.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉A novel feature dimensionality reduction strategy based on kernel regularized relevance weighted discriminant analysis is proposed in this paper with some interesting characteristics. First, the proposed method has shown its effectiveness in dealing with a small sample size problem when using the regularized linear discriminant analysis (RLDA) technique and Kernel theory. Second, a new computation method is proposed to solve the complicated and inefficient computation procedure problem in the traditional RLDA technique while using the cross-validation method. The experimental results indicate that the proposed algorithm shows better performance than the other methods.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉This paper presents the design and fabrication of a prototype microwave patch antenna system for non-destructive corrosion monitoring of embedded steel rebar in civil structures. A planar rectangular patch for 2.4 GHz resonant frequency is modeled and loaded with electromagnetic bandgap (EBG) structure having an aperture slot of 2 × 10 cm〈sup〉2〈/sup〉 in the center. The EBG plane provides confined radiation from its aperture and suppresses any destructive interference in the sensing area. The proposed patch sensor is fabrication using RO4450B substrate and tested on autoclaved aerated concrete and concrete block. The EBG superstrated patch is mounted on the test specimen and magnitude of reflection coefficient (S〈sub〉11〈/sub〉) is measured at the near-field distance. A change in the intensity of scattered fields because of the oxide layer formed over the rebar by corrosion process is observed and measured. Analysis of weight loss of rebar due to corrosion is correlated with scattering parameter and equivalent DC voltage signaling. The measured results show that the proposed sensing technique is useful in corrosion detection of rebar inside the civil structures.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉As an efficient non-destructive checking technique, scanning acoustic microscopy has been rapidly developed and widely used in nondestructive detection for composites, chip manufacturing, bio-pharmaceuticals and so on. Scanning acoustic microscopy can obtain the two-dimensional image and reconstruct three-dimensional model of the internal structure inside the sample, compared with the optical microscope. Due to the geometrical features of high frequency transducer, the precision of focus has already become the key factor for the digital imaging and defect detecting. Especially for the current 3D packaging chips and multi-layer composite materials, thickness of layer has reached the level of nm. Different focus plane will provide a completely different digital image. However, the process of focusing has been relying on manual operation without great development until now. To improve the speed and accuracy of the focusing, we proposed a quick auto-focus method on scanning acoustic microscopy. This method can be divided into three steps. The first one is fast auto-focusing on upper surface, the second other one is auto-focusing on the interlayer base on 〈span〉 〈span〉\(V(z)\)〈/span〉 〈/span〉 curve, and the last one is to improve the A-Scan signal of points in the 〈span〉 〈span〉\(V(z)\)〈/span〉 〈/span〉 curve with spars signal representation. This method is integrated into Scanning Acoustic Microscopy imaging software and has been validated in the experiment with Olympus transducers at 30\50 MHz. It can be seen that this method can effectively improve the accuracy of focus and shorten the focus time of Scanning Acoustic Microscopy through experiments.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉The change detection system plays an active role in remote sensing images by detecting the changes in the image environment. Changes may be occur frequently due to the seasonal factors such as deforestation, natural disorders, etc. The objective of this paper is to develop an effective approach to find the changes that were experienced over a time period. The changes detected by difference image (DI) method is not efficient because it has the drawbacks in speckle noise suppression and classification, also the quality of DI affects the change map. This paper proposes a sample selection-multiple dilated convolutional neural network for the detection of changed and unchanged areas. In this, multiple dilation rates are incorporated, that overcome the existing gridding problems in network. The feature map determined at fully connected network extracts the global information by widening the receptive field that covers all the regions without any missing portions in the image. Hence, this type of architecture improves the learning in deep network by detecting the changes accurately. Finally, the trained network model achieves a feature map from the two images with the classified result of changed and unchanged pixels. The accuracy of the proposed change detection result provides improved results as compared with the existing algorithms. The performance metrics like false positive, false negative, overall error, the percentage of correct classification and Kappa are used for the estimation of the proposed image change detection task. Experiments are conducted on seven real datasets, and the result shows the better detection result than the other existing approaches.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉Characters in the historical seals is valuable for the research of corresponding documents. Character extraction for these seals is challenging because of the special characteristics of the seals such as the different carving type (rilievi or diaglyph) and variant border form. Thus most existing character extraction methods do not work well with the images of the historical seals. In this paper, a new character extraction method is proposed based on novel features of the scanned seal images. First, the border width feature and structure feature of each scanned seal image are extracted by analysis of the probabilistic density distribution of border width and brightness of each column. Meanwhile, the optimal border width for rilievi or a starting position of border for diaglyph are generated for subsequent border removal and character extraction. The tri-border feature is used to differentiate between diaglyph and tri-border diaglyph. Then a decision tree is built to classify the seal images into rilievi and diaglyph. The classification result is used to divide background, border and character. After the border is removed with the optimal border width estimated from border width feature, the characters are finally extracted from the seal image. Experimental results on the real data set show that the proposed method classifies the scanned seal image accurately and extracts the characters effectively.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉Existing product data, presented as description documents, product images, etc., are very significant references for designing a new proposal. However, traditionally, the data collected manually only contain the product images without description documents. Obtaining reasonable description of product is a challenging task due to the high cost and labor-intensive process of human annotation. In this paper, a new approach is introduced to solve this problem and improve the efficiency of description by exploiting the potential information of product images. We propose a robust framework with multi-label learning to annotate each product image with several labels, which makes a brief description of the product from different aspects. An efficient method is proposed to accomplish product data acquisition, arrangement and analysis task to construct new data in images collection step. Furthermore, for the newly structured data, robust algorithms are employed to accomplish the data processing. Based on the processed data, automatic and semi-automatic multi-label annotation methods are applied to generate the labels for product images. We present the prediction results by several state-of-the-art multi-label learning classifiers based on the features extracted from different convolution neural framework. The results are evaluated by effective measures to validate the quality of the labeling result, as well as the predictive performance with respect to the number of the training examples. Experimental results indicate that the quality of the annotation is reasonable, and the proposed method can achieve excellent prediction accuracy based on the annotated data, generating satisfactory description for the products.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉The deaf community relies on sign language as the primary means of communication. For the millions of people around the world who suffer from hearing loss, interaction with hearing people is quite difficult. The main objective of sign language recognition (SLR) is the development of automatic SLR systems to facilitate communication with the deaf community. Arabic SLR (ArSLR) specifically did not receive much attention until recent years. This work presents a comprehensive comparison between two different recognition techniques for continuous ArSLR, namely a Modified k-Nearest Neighbor which is suitable for sequential data and Hidden Markov Models (HMMs) techniques based on two different toolkits. Additionally, in this work, two new ArSL datasets composed of 40 Arabic sentences are collected using Polhemus G4 motion tracker and a camera. An existing glove-based dataset is employed in this work as well. The three datasets are made publicly available to the research community. The advantages and disadvantages of each data acquisition approach and classification technique are discussed in this paper. In the experimental results section, it is shown that classification accuracy for sign sentences acquired using a motion tracker are very similar the classification accuracy for sentences acquired using sensor gloves. The modified KNN solution is inferior to HMMs in terms of the computational time required for classification.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉In the field of welding process control, on-line monitoring of welding quality based on multi-sensor information fusion has attracted more attention. In order to recognize the penetration state of the Keyhole mode Tungsten Inert Gas welded joint in real time, an acoustic and visual sensing system was established in this paper. The acoustic and visual features that characterize the penetration state of the welded joints in 34 dimensions were extracted and the variation of the acoustic signal and the keyhole geometry were analyzed. In addition, the weighted scoring criterion based on the Fisher distance and the maximum information coefficient (Fisher–MIC) and Support Vector Machine (SVM) model based on cross-validation (CV) are designed as the feature selection method. The feature selection method can evaluate the penetration recognition accuracy of different feature subsets. The experiment results show that the maximum recognition accuracy was 97.1655%, which was performed by the 10-dimension optimal feature subset and the CV–SVM model with particle swarm optimization (PSO–CV–SVM). It is proved that the selected acoustic and visual features can well characterize the penetration state of the welded joints, and the feature selection method and PSO–CV–SVM model have superior performance.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉In the process of spectral segmentation, it is crucial to compute a reliable affinity matrix with different features of an image. In this paper, we present a method of constructing the affinity matrix based on multi-resolution features extracted from the original features. A pointwise multi-resolution feature descriptor (PMFD) is designed based on spectral graph wavelets, which characterize the topology of the image centered at different pixels. After choosing the scales of interest in our descriptor, a new affinity matrix is constructed based on the extracted features. For large-size affinity matrixes, it is difficult to compute the proposed PMFD for all pixels of an image. Therefore, an approximate algorithm is proposed to compute the PMFD. To demonstrate the effectiveness of our method, a series of experiments on the Berkeley image segmentation dataset are implemented using the PMFD-based spectral segmentation algorithm. A comparison with other image segmentation techniques demonstrates that our method offers significantly improved pointwise spectral segmentation performance.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉In Earth, the most crucial illness for cause of death is ischemic stroke. Ischemic stroke arises as a result of an obstacle within a blood vessel supplying blood to the brain. In this paper, for sub-acute ischemic stroke lesion segmentation, we utilize an effective meta-heuristic feature selection technique along with hybrid Naive Bayes (NB) and sample weighted random forest (SWRF) classification approach. Initially, the features are extracted from the pre-processed image, after that, the feature selection is done by using the multi-objective enhanced firefly algorithm. To improve the classification performance, the dimensionality of the feature vectors and errors are reduced by eliminating such irrelevant and redundant features. After the feature selection process, an ensemble of NB and SWRF classifiers is used for segmenting the image. Here the NB classifier is trained and applied to estimate the weights of training samples. Then, the training samples with estimated weights are utilized to train SWRF. In our work stroke lesion segmentation is formulated as a binary classification problem where every local region is classified as either affected or non-affected area.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉The impetus of the study is to present a novel micro electro mechanical system based tunable gyroscope with the possibility of sensitivity enhancement using appropriate electrostatic hopping voltages. The proposed model is a silicon-based clamped–clamped micro beam sandwiched with two piezoelectric layers throughout the entire length. The nonlinear electrostatic forces are applied to the micro beam along its sense and drive mode directions (either lateral sides). The drive mode actuation is a combination of a direct current (DC) and an alternating current voltage; whereas the sense mode actuation is a pure DC voltage. The micro beam oscillates along the drive mode due to the harmonic drive mode excitation; as the micro beam undergoes base rotation, the Coriolis force induces another motion in the direction of the sense mode which is perpendicular to the drive mode direction. The more is the amplitude of the base rotation, the more is the sense mode amplitude. The sense mode amplitude is directly attributed to the magnitude of base rotation. The piezoelectric layers are actuated by a DC voltage which leads to an axial force proportional to the applied DC voltage. Exciting the piezoelectric layers changes the overall stiffness of the micro beam and as a result the operating frequency of the gyroscope becomes tunable. The partial differential equation of the motion is derived using Hamiltonian principle and discretized into two nonlinear ordinary differential equations along the drive and sense mode directions. The shooting method is used to capture the periodic motion orbits and accordingly the frequency response curves. By using Floquet theory the stability of the periodic orbits is determined. Due to the nonlinearity of the governing equations in the vicinity of the primary resonance, the gyroscope exhibits multi-response solution; Applying appropriate hopping voltages, the micro beam is pushed into the attraction basin of the response with higher amplitude and accordingly the sensitivity of the gyroscope is enhanced. The proposed gyroscope not only has the capability of having improved sensitivity but also its operating frequency can be tuned both in forward and backward directions by means of applying appropriate piezoelectric voltage with an appropriate polarity. 〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉Electromagnetic tomography (EMT) has been developed for visualizing the conductivity distribution of materials with multi-coil electromagnetic sensors. It is crucial to design an optimized EMT sensor for the improvement of reconstructed images quality. The selection of the number of coils is discussed in this paper due to its great importance to system performance and system complexity. It is commonly believed that more coils in EMT sensor would obtain better performance of reconstructed images. In order to study the impact of number of coils in EMT sensors on quality of reconstructed images, five kinds of sensors with different number of coils including 4, 8, 12, 16 and 20, are involved to conduct numerical simulations. EMT forward problem can be solved through implementing finite element method, then measurements and sensitivity matrix are obtained, which can be used to solve EMT inverse problem with proper image reconstruction algorithms. Five typical conductivity distributions are used to verify the performance of EMT sensors with different number of coils. The sensitivity matrices of different EMT sensors are analyzed to further explain the essential reason of these numerical simulation results using singular value decomposition. It can be concluded that EMT sensor with 16 coils produces good-enough image reconstruction results for most of the typical conductivity distributions. Limited improvement can be obtained in the quality of reconstructed images when the number of coils is more than 16.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉The convergence of spectroscopy and imaging technologies, emerge into a single sensing technology i.e., provides spatial and spectral information of the objects under investigation. The hyperspectral technique is one of the popular techniques used in numerous fields of study to determine size, shape, texture, material composition, morphology and external defects. The main advantage of this sensing technology lies in the fact that it is capable not only makes direct assessment of the material under study but also it can indicate the spatial distribution of the selected parameters. The aim of this paper is to present a detailed outline of the principles, background, acquisition methods, component descriptions and recent advances of the hyperspectral imaging systems for laboratory and industrial environments.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉In this study, we developed a logical and complete mathematical analysis of the relationship between the discrete Fourier transform (DFT) magnitude and periodic feature of a signal. The physical meaning of the DFT magnitude index corresponds to the periodic signal feature; further analysis makes clear the relationship among DFT magnitude, magnitude index, and number of samples. The proposed analysis method also elucidates the relationship between alternating current magnitude index and frequency, by the way, the unit of image frequency will be given. The proposed analysis method is suitable for the machine vision and image processing in general.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉Handwritten characters recognition is a challenging task due to the diversity of writing styles. In the present paper, a new offline recognition system based on deep convolutional neural networks (CNNs) is designed. It uses CNNs to extract features from raw pixels which makes the proposed technique more flexible than the other conventional ones that require a set of additional preprocessing steps to extract the desired invariants features. The proposed system is tested on the AMHCD data set and achieved the best recognition accuracy (99.10%) compared to the state-of-the-art results.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉Traditionally, design schemes were usually drawn manually or with the help of graphics software. These methods are usually labor-intensive and time-consuming. The design field calls for advanced product design, which will be highly efficient and meet more individualized demands for customized manufacturing. In this paper, a new design approach is introduced to assist designers to improve the efficiency of product design with automatic sample generation algorithms. This approach mainly consists of two parts: design scheme generation and sketch inversion. For design scheme generation, we adopt the generative adversarial networks to extract features from the existing product images and generate new design schemes based on these features. This is followed by post-processing including wire frame design and sketch design. Meanwhile, for sketch inversion, the sketches are generated and input along with the corresponding color images to train the sketch inversion model. With this model, hand-drawn sketches are transferred into color design schemes. We take watch designing as an example to validate the effectiveness of the proposed method on design scheme generation and sketch inversion. Experimental results demonstrate that the proposed design scheme generation method can generate new product design schemes, and further sketch inversion process can transfer them into color design schemes with high quality.〈/p〉

Description: 〈span〉 〈h3〉Abstract〈/h3〉 〈p〉Detection methods and analytical devices have drawn increasing attention in recent years due to their direct impact on early detection, monitoring and diagnosis of disease in medical research. In this work, we describe a simple but so far unrecognized label-free method for surface-bound analyte detection, which could be applicable to a wide range of substances. In this respect, the feasibility and practical aspects of a micrometer-scale, poly-〈span〉l〈/span〉-lysine-based, solid-phase assay for label-free analyte detection with electrostatic read-out are investigated. Micropatterned poly-〈span〉l〈/span〉-lysine layers were produced using soft-lithography on mica and their electrostatic surface potential was determined using Kelvin-probe Force Microscopy. Ribose, a natural sugar, was used as analyte. Upon exposure to ribose, the surface potential changed from positive to negative in a reversible manner. We report for the first time the use of an electrostatic principle for assay read-out. This purely physical effect could be used to develop label- and marker-free assays for sugars, various other substances or, possibly, biosensors.〈/p〉 〈/span〉 〈span〉 〈h3〉Graphical Abstract〈/h3〉 〈p〉 〈span〉 〈span〉 〈img alt="" src="https://static-content.springer.com/image/MediaObjects/11220_2019_244_Figa_HTML.png"〉 〈/span〉 〈/span〉 〈/p〉 〈/span〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉Synthetic aperture imaging is widely applied to reconstruct the occluded objects. Unfortunately, due to the superimposition of foreground rays, blurriness occurs in reconstructed images, which leads to an undesirable quality of reconstruction result. Screening effective rays before synthetic aperture imaging become the key to improve the quality of reconstructed image. In this paper, we propose a method to remove interference from occlusion in synthetic aperture imaging by using image matting. By focusing on the occlusion plane, the focusing degree of synthetic aperture imaging result can roughly label occlusions and non-occlusions. These labels can serve as the scribbles of input image of image matting, which can distinguish occlusions and backgrounds more precisely. After that, the focusing at the desired depth is by averaging pixels without occlusions. The experimental results show that our method can effectively remove the blur in the reconstruction results. We demonstrate the superiority of our method by presenting experimental results as well as comparing our method with other’s methods.〈/p〉

Description: 〈p〉The original version of this article unfortunately contained a mistake. Figure 1 was partly missing at the bottom in the pdf and xml versions of the article.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉Diapers rash are harmful to the babies and the elderly; so, wetness detection technologies have been sophisticated by many research worker to assist in discovering these events. The study and survey offered here looked into a radio frequency-based microcontroller and GSM for wetness detection. One of the crucial concerns is not only the detection of humidity but the Health complications of diapers rash. This study utilizes a wetness detection system with different types of alarms: audio buzz, flashing light emitting diode, and Smartphone. Our proposed system consists of a wetness sensor, microcontroller, radio frequency transceiver microcontroller, and GSM modem, or computer. Two thin wires between diaper sheets or layers of medical dressing set up the wetness sensor. The two or four thin wires connect to a couple of pressing studs sticking out from one end of the diaper or gauze. When the wires are wetted, the resistance between the studs falls down a pre-established threshold, launching the alarm. Radio Frequency transceiver passes the warning alarm through Microcontroller to the GSM modem. Consequently, Smartphone and other devices forward the alarm to whom it may concern wetness or bleeding is occurring.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉Currently, the diagnosis of depression is largely based on clinical judgments due to the absence of objective biomarkers. There are increasing evidences that depression (DP) is associated with structural abnormalities. However, the previous analyses have a poor predictive power for individuals. To discriminate DP patients from normal controls (NCs) studied with structural magnetic resonance images using the method of support vector machine (SVM) combined with recursive feature elimination (RFE). In this study, 40 DP patients and 40 age- and sex-matched NCs were recruited from Guangzhou Brain Hospital and the local community, respectively. We calculated gray matter volume (GMV) and white matter volume (WMV) of 210 cortical and 36 subcortical regions, defined by the Human Brainnetome Atlas. The group differences between DP patients and NCs were compared. The method of SVM combined with RFE was applied into the discriminative analysis of DP patients from NCs, in which discriminative features were drawn from GMV and WMV. We found that the DP patients showed significant GMV reductions in eight brain regions and showed significant WMV reductions in ten brain regions. The classifier using GMV as input features achieved the best performance (an accuracy of 86.25%, a sensitivity of 85%, and a specificity of 87.5%) in the discriminative analyses between DP patients and NCs. These findings provided evidences that specific structural brain regions associated with DP patients might qualify as a potential biomarker for disease diagnosis, and the machine-learning method of SVM with RFE may reveal neurobiological mechanisms in distinguishing DP patients from NCs.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉Because of the large structure and long training time, the development cycle of the common depth model is prolonged. How to speed up training is a problem deserving of study. In order to accelerate training, K-means clustering optimizing deep stacked sparse autoencoder (K-means sparse SAE) is presented in this paper. First, the input features are divided into K small subsets by K-means clustering, then each subset is input into corresponding autoencoder model for training, which only has fewer nodes in the hidden layer than traditional models. After training, each autoencoder’s trained weights and biases is merged to obtain the next layer’s input features by feedforward network. The above steps are repeated till the softmax layer, then fine-tuning is carried out. Using MNIST-Rotation datasets to train the network that has three hidden layers and each layer has 800 nodes, the improved model has higher classification accuracy and shorter training time when K = 10. With K increasing, the training time is reduced to almost the same as the fine-tuning time but the recognition ability is descended. Compared with the recently stacked denoising sparse autoencoder, the recognition accuracy is improved by 1%, not only the noise factor is not selected but also the training speed is significantly increased. The trained filters from the improved model is also used to train convolutional autoencoder, and it performs better than traditional models. We find that pre-training stage doesn’t need large samples simultaneously, and small samples parallel training reduces the probability of falling into the local minimum.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉For the quantitative analysis of medical images in clinical research, diagnosis and treatment, a reliable basic framework for developing a normal human ear atlas of voxel-based computed tomography (CT) images was proposed. We annotated 10 precise ear structures with different labels from 64 patients with normal ear structures. Paired-samples 〈em〉t〈/em〉 test, Pearson’s test and descriptive statistics were carried on the volume and coordinate data, which were first obtained from annotation to verify the correlation and difference. In addition, we constructed a three dimensional (3D) model of the standard human ear atlas with six views for presentation. Through a series of statistical analyses, a standard 3D normal human ear atlas containing volume and spatial data was obtained from voxel-based CT images. There was a significant negative correlation exists between age and the volume of the incus, and no correlation with other structures. There was no significant correlation between slice thickness and the volume of 10 structures. The volume of most structures on both sides is significantly correlated and there was no significant difference in the volume of most structures on both sides except for the jugular foramen. Besides, the coordinate range of the bilateral structures is relatively consistent. The specific volume and spatial data for the human ear atlas are helpful in the diagnosis of abnormalities, and this 3D normal human ear atlas will provide new insights for radiologists in clinical research.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉This paper explores the possibility of the precise determination of mechanical vibrations using metamaterial split ring resonator (SRR) structure. The amplitude of the interacting electromagnetic wave in the range of GHz frequency is directly varied in accordance with the amplitude of mechanical vibration, using a Broadside Coupled SRR (BCSRR) acting as a vibration sensor. Dependence of the spacing between the two rings on the resonance frequency of the BCSRR is used for the detection of vibration and it is achieved by allowing the spacing to change in accordance with the amplitude of mechanical vibration. For the effective sensing of mechanical vibration, the electromagnetic wave frequency is chosen at the center of the linear portion of the rising or the falling slope of the resonant curve of the BCSRR. By properly choosing the parameters of the BCSRR along with the effective tuning of the operating frequency, it is possible to detect even very weak vibrations. The chances of various distortions in the detected vibration waveform in connection with selection of the operating frequency and intensity of vibrations are also analyzed. The qualitative formulation of the detection process along with its experimental verification is presented. This novel method may find applications in the detection of mechanical vibrations caused by various man made and natural sources and may find manifold possibilities in the field of communication and instrumentation.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉Image super-resolution reconstruction is the current research focus and important field of digital image processing, and widely applied in public security, medical diagnosis, etc. Inspired by the idea of granular computing, we explore the granular computing models and algorithms of image super-resolution reconstruction integrating theories and methods of computer science, mathematics, and cognitive psychology. The research includes the following aspects. (1) The granulation method of image is proposed to transform the image space into granularity space. (2) The join operator and the meet operator between two granules are designed for the fuzzy inclusion measure mu and sigma between two granules, are used to realize the transformation between two granularity spaces with different granularities, to obtain the prior knowledge to guide the design of image super-resolution reconstruction algorithms. (3) According to up–down and bottom-up computing models, the granular computing algorithms of image super-resolution reconstruction are designed to realize the transformation from granularity space to image space in terms of the prior knowledge. Research can be summarized for learning process and reconstruction process, the learning process obtains the prior knowledge of image by granulation and forms granular computing model, and the reconstruction process reconstructs the high-resolution image of the given low-resolution image by granulation and the obtained prior knowledge. The feasibility of the preliminary study was verified by experiments. The research aims to build image super-resolution reconstruction models and algorithms satisfying human cognition.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉This paper describes the use of shallow trench isolation as a modified trench capacitor in CMOS image sensor. MTC helps to tune the conversion gain at the floating diffusion (FD) node. The use of MTC in pixel improves the sensitivity of the pixel without affecting the dynamic range (DR) of the pixel. The proposed pixel structure uses an existing in-built isolation trench as a capacitor. During low illumination, the MTC is separated from the floating diffusion (FD) node thus makes it easier to detect the low light signals owing to the lower FD node capacitance. However, in the event of high illumination, MTC is connected in parallel to the FD node increasing the capacity to store more charges from the photodiode. This attributed to the improved DR performance of the pixel. Using 2D numerical simulations, we show that an overall 60 dB enhancement in the DR as compared to the conventional pixel. Moreover, the proposed modification does not affect the other pixel characteristics such as dark current and crosstalk. 〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉It is well known that under blue LED/laser illumination a class of synthetic conjugated polymers can create fluorescence, and after absorbing a certain class of explosive vapors the fluorescence will be quenched thus the explosives can be detected. The unsolved problems are how can we detect drug, TATP and nitro aromatic explosives with the same device, how can we distinguish different explosives and drugs, and how can we avoid false alarms; solving these problems are crucial for world security. We have developed a new synthetic conjugated polymer with single molecule layer and coated on porous silicon with large surface area to increase quenching signal at least one order, based on this new film a small handheld explosive detector with sensitivities of 0.1 pg for TNT and 0.1 ng for black gun powder are obtained. Many explosives and drugs (such as TNT, RDX, PETN, black gun powder, cocaine, etc.) have been tested, from their quenching curves and our data base we are able to distinguish them qualitatively and avoid false alarms. Therefore we believe our detector can be used not only for battle field but also for airports, government entrances and other security check points. Last year, after face to face competition, our device was selected as the only security detector for the G20 summit held in Hangzhou, China. Here, we will give our detailed designs to let readers develop detectors by themselves to counter terrorism in the world and save lives. 〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉Obtaining a desired transient performance of closed-loop system is the most important issues in impractical and industrial applications, in general. It is to note that the small overshoot and acceptable settling time of the system response are almost two typical characteristics in the transient performance. There are a number of contradictions between these performances; and therefore, a tradeoff between these ones should deeply be considered. In a word, a so-called composite nonlinear feedback (CNF) method is an efficient and simple technique which is employed to overcome the contradiction of simultaneous achievement of the mentioned transient performances. CNF based sliding mode control (SMC) for nonlinear systems with known upper bound uncertainty are considered in this paper. Unlike the existing results for the linear systems, the proposed CNF based SMC comprises two nonlinear parts. The first term assures the stability of the closed-loop nonlinear system and provides a fast convergence response. The second term reduces the overshoot of the response. Finally, to show the merits of the proposed approach, it is applied to two nonlinear Genesio’s chaotic system and a nonlinear helicopter system. The investigated results verify the effectiveness of the proposed approach.〈/p〉

Description: 〈h3〉Abstract〈/h3〉 〈p〉Samarium doped chalcogenide core, perfluorinated polymer clad with Ag metal and plasmonic 2D materials based plasmonic fiber-optic sensor is simulated and analyzed in near-infrared (NIR) wavelength regime. Proposed sensor is directed at the detection of the malignancy stages of liver tissues. The performance analysis (in terms of figure-of-merit, 〈em〉i.e.〈/em〉, FOM) was carried out taking into account the MoS〈sub〉2〈/sub〉 and graphene monolayers as performance enhancing 2D materials. The analysis suggests that FOM values of MoS〈sub〉2〈/sub〉-based sensor probe are better than graphene-based probe. Further, a comparative study shows that fiber-optic probe is able to provide much better performance than prism-based probes. The FOM gets better for longer wavelength. The specificity of the biosensor can be improved by employing a suitable buffer layer (1–15 nm) as a bio-recognition element.〈/p〉