Search

Nonlinear process modeling is a primary task in intelligent manufacturing, aiming at extracting high-value features from massive process data for further process analysis like process monitoring. However, it is still a challenge to develop nonlinear process models with robust representation capability for diverse process faults. From the new perspective of the correlation between process variables, this paper develops a nonlinear process modeling algorithm to adaptively preserve the features of both global and local inter-variable structures, in order to fully exploit inter-variable features for enhancing the nonlinear representation of process operating conditions. Specifically, a unidimensional convolutional operation with a self-attention mechanism is proposed to simultaneously extract global and local inter-variable structures, wherein different attentions can be adaptively adjusted to these two structures for the final aggregation of them. Besides, cooperating with a two-dimensional dynamic data extension, the unidimensional convolutional operation can represent the overall temporal relationship between process samples. Through stacking a collection of these convolutional operations, a ResNet-style convolutional neural network then is constructed to extract high-order nonlinear features. Experiments on the Tennessee Eastman process validate the effectiveness of the proposed algorithm for two vital process monitoring problems-fault detection and fault identification.

The emergence of fog computing has improved the performance and efficiency of intelligent manufacturing. In a fog computing platform, fog nodes can provide the distributed and proximity resources for the tasks related to intelligent manufacturing. Due to the heterogeneous characteristics of the fog nodes and the space–time characteristics of intelligent manufacturing, the effective deployment of the fog nodes is a precondition for the fog computing implementation and it is also a key to provide high-performance services with the fog computing platform. However, traditional deployment strategies cannot satisfy the performance requirements of intelligent manufacturing systems. In this article, the problem of the fog nodes deployment is studied and a fog nodes deployment strategy based on the space–time characteristics (TSBP) is proposed. A fog nodes deployment system model is set up and the objective function of the fog nodes deployment is built up, in which the optimization goal is to minimize the computing response time and realize the load balancing of the fog nodes. In addition, the discrete differential evolution algorithm is applied to search the optimal fog nodes deployment solution. Finally, the effectiveness of the proposed TSBP strategy is verified using a candy packaging intelligent production line prototype platform.

The advent of convolutional neural networks has led to remarkable progress in dense stereo labeling problem, achieving superior performance over the traditional methods. However, the ill-posed nature of stereo matching makes noise (outliers) in winner-takes-all (WTA) disparity maps inevitable. This paper presents a robust statistical approach to noise detection and refinement of WTA disparity maps. In the context of noise detection, the input noisy WTA disparity map is segmented into regular grid cells (regions) with the aim of leveraging Markov random field (MRF) to infer candidate disparity labels. However, there are two key problems: there can be large severe outliers in the regions; second, the regular partition process may produce regions with mixed disparity distributions. To overcome these problems, we optimize a robust objective function over the segmented disparity map. By obtaining the optimal solution of the objective function through a maximum a posteriori estimation in a probabilistic model, we are able to infer MRF candidate disparity labels. We then apply a soft-segmentation constraint on the estimated MRF candidate disparity labels to describe and detect outliers in the disparity map. Next, an edge-preserving statistical inference that leverages the joint statistics of the disparity map and its guidance reference image is used to select correct candidate disparity for each detected outlier. Finally, a weighted median filter is applied to remove small spikes and irregularities in the resulting disparity map. Rigorous and comprehensive experiments showed that the proposed method is distributionally robust and outlier resistant, and can effectively detect and correct outliers in disparity maps. Middlebury evaluation benchmark validated the competitive performance of the proposed method.

In this paper, we investigate different activation functions (AFs) on convergence performance of a gradient-based neural network (GNN) for solving linear matrix equation, AXB + X = C . It is observed that, by employing different AFs, i.e., linear, power-sigmoid, sign-power, and general sign-bi-power functions, the presented GNN model can achieve different convergence performance. More specifically, if linear function is employed, the GNN model can achieve exponential convergence; if the power-sigmoid function is employed, superior convergence can be achieved as compared to the linear case; while if the sign-power and general sign-bi-power functions are employed, the GNN model can achieve finite- and fixed-time convergence, respectively. Detailed theoretical proofs are offered to demonstrate these facts. Besides, the exponential convergence rate and the upper bounds of finite and fixed convergence time are also theoretically estimated. Finally, two illustrative examples are performed to further substantiate the aforementioned theoretical results and the effectiveness of the presented GNN model for solving the linear matrix equation.

Deep penetration of personal computing devices and high-speed Internet has enabled everyone to be a broadcaster. In this crowdsourced live streaming service, numerous amateur broadcasters lively stream their video contents to viewers around the world. Consequently, these broadcasters generate a massive amount of video data. The set of video sources and recipients are big as well, so are demand for the storage and computational resources. Transcoding becomes a must to better service these viewers with different network and device configurations. However, the massive amount of video data contributed by countless channels even makes cloud significantly expensive for providing transcoding services to the whole community. In this paper, inspired by the paradigm of Edge Computing, we propose a Cloud-edge collaborative system which combines the idle end-viewers’ resources with the cloud to transcode the massive amount of videos at scale. Specifically, we put forward tailored viewer selection algorithms after empirically analyses the viewer behavior data. In the meantime, we propose auction-based payment schemes to motivate these viewers participating in the transcoding. Large-scale trace-driven simulations demonstrate the superiority of our approach in cost reduction and service stability. We further implement a prototype in PlanetLab to prove the feasibility of our design.

Support vector machine (SVM) is a powerful technique in pattern classification, but its performance is highly dependent on its parameters. In this paper, a new SVM optimized by a novel differential evolution (DE) with a hybrid parameter setting strategy and a population size adaptation method is proposed and simplified as FDE-PS-SVM. In the hybrid parameter setting strategy, the SVM parameter offspring are generated by DE operators with evolutionary parameters that are fixed or with the ones generated by fuzzy logic inference (FLI) according to a given probability. In the population size adaptation method, the population size is shrunk gradually during the search, which tries to balance the diversity and concentration ability of the algorithm to find better SVM parameters. Some benchmark data sets are used to evaluate the proposed algorithm. Experimental results show that the two proposed strategies are effective to search for better SVM parameters while the proposed FDE-PS-SVM algorithm outperforms other algorithms published in other literature.

As a typical representative of the recurrent neural network (RNN), Zhang neural network (ZNN) has been proved as a powerful parallel-processing neural solver for time-varying matrix problems. Recent studies have shown that, in the absence of noise, the ZNN model activated by a combined activation function (CAF), which is the linear combination of sign-bi-power (SBP) and linear functions (termed CAF-ZNN), can achieve much better finite-time convergence compared with other ZNN models for time-varying matrix pseudoinversion. This paper investigates for the first time the influence of noises on such a CAF-ZNN model. We not only present the upper bound of steady-state solution error synthesized by the noise-polluted CAF-ZNN model for time-varying matrix pseudoinversion but also derive the finite convergence time for the solution error reaching the upper bound. Both the theoretical analyses and simulation observation reveal that the solution error can be made arbitrarily small by choosing some design parameters of CAF-ZNN. At last, a robotic application is successfully performed to further illustrate the feasibility of the CAF-ZNN model to kinematic control of redundant robot manipulator.

In this paper, based on the difference of color depth and gray scale of wafer dopant surface, a regional variational level set model is proposed to apply to the distribution extraction problem of wafer dopants. First, for the interior of the level set evolution curve, the weighted sum of the mean value of the cluster center point after K-means clustering and the median filtered pixel is used as the internal fitting value of the evolution curve. Second, for the outside of the evolution curve, the image entropy is introduced as a supplement to the gray information. Third, the maximum absolute mean difference is used to adaptively adjust the weight ratio both inside and outside the level set evolution curve. Finally, the experimental results show that compared with the existing level set segmentation methods, the proposed method makes the segmented images have higher segmentation accuracy.

Quality monitoring is important for farmland protection. Here, high-resolution remote sensing data obtained by unmanned aerial vehicles (UAVs) and long-term ground sensing data, obtained by wireless sensor networks (WSNs), are uniquely suited for assessing spatial and temporal changes in farmland quality. However, existing UAV-WSN systems are unable to fully integrate the data obtained from these two monitoring systems. This work addresses this problem by designing an improved UAV-WSN monitoring system that can collect both high-resolution UAV images and long-term WSN data during a single-flight mission. This is facilitated by a newly proposed data transmission optimization routing protocol (DTORP) that selects the communication node within a cluster of the WSN to maximize the quantity of data that can be efficiently transmitted, additionally combining individual scheduling algorithms and routing algorithms appropriate for three different distance scales to reduce the energy consumption incurred during data transmission between the nodes in a cluster. The performance of the proposed system is evaluated based on Monte Carlo simulations by comparisons with that obtained by a conventional system using the low-energy adaptive clustering hierarchy (LEACH) protocol. The results demonstrate that the proposed system provides a greater total volume of transmitted data, greater energy utilization efficiency, and a larger maximum revisit period than the conventional system. This implies that the proposed UAV-WSN monitoring system offers better overall performance and enhanced potential for conducting long-term farmland quality data collection over large areas in comparison to existing systems.

Recently, a large number of physical devices, together with distributed information systems, deployed in internet of things (IoT), are collecting more and more images. Such collected images recognition poses an important challenge on optimization in internet of things. Specially, most of existing methods only adopt shallow learning models to integrate various features of images for recognition limiting classification accuracy. In this paper, we propose a multimodal deep learning (MMDL) approach to integrate heterogeneous visual features by considering each type of visual feature as one modality for image recognition optimization in internet of things. In our scheme, we extract the high-level abstraction of each modality by a stacked autoencoders. Furthermore, we design a back propagation algorithm with shared weights learned from a softmax layer to update the pretrained parameters of multiple stacked autoencoders simultaneously. The integration is performed by concatenating the last hidden layers of the multimodal stacked autoencoders architecture. Extensive experiments are carried out on three datasets i.e. Animal with Attributes, NUS-WIDE-OBJECT, and Handwritten Numerals, by comparison with SVM, SAE, and AMMSS. Results demonstrate that our scheme has superior performance on heterogeneous visual features integration for image recognition optimization in internet of things.

Vision inspection has been extensively used in the field of defect measurement field as a non-contact and nondestructive measurement technique. The conventional methods of visual defect detection rely heavily on the standard templates and features of image, but the standard templates are difficult to obtain or even don’t exist in many application fields especially for the field of flexible Integrated Circuit (IC) substrate defect detection based on micro vision. To solve the above problems, an algorithm of oxidation defects detection based on topology mapping. First, the filter template of weighted average neighborhood closed curve based on topology mapping is used to remove the image noise. Second, an image segmentation method is proposed by using the idea of maximum variance between classes to find out the threshold, then the spatially filtered is used to remove the normal surface texture and the structure of the substrate which is introduced by metallographic microscope enlargement. Finally, experiments are performed to show that the filter template of weighted average neighborhood closed curve is better than the existing filters and effectively removes image noise, then the segmentation method in this paper works better.

This article describes how with the rapid increasing of multimedia content on the Internet, the need for effective cross-modal retrieval has attracted much attention recently. Many related works ignore the latent semantic correlations of modalities in the non-linear space and the extraction of high-level modality features, which only focuses on the semantic mapping of modalities in linear space and the use of low-level artificial features as modality feature representation. To solve these issues, the authors first utilizes convolutional neural networks and topic modal to obtain a high-level semantic feature of various modalities. Sequentially, they propose a supervised learning algorithm based on a kernel with partial least squares that can capture semantic correlations across modalities. Finally, the joint model of different modalities is learnt by the training set. Extensive experiments are conducted on three benchmark datasets that include Wikipedia, Pascal and MIRFlickr. The results show that the proposed approach achieves better retrieval performance over several state-of-the-art approaches.

Missing values are common in cyber-physical systems (CPS) for a variety of reasons, such as sensor faults, communication malfunctions, environmental interferences, and human errors. An accurate missing value imputation is crucial to promote the data quality for data mining and statistical analysis tasks. Unfortunately, most of the existing methods take use of the whole data set to impute a missing value, which could have unfavorable influences and impacts (low accuracy or high complexity) on the imputed results caused by irrelevant records. Aiming at this problem, this paper develops a novel local similarity imputation method that estimates missing data based on fast clustering and top $k$ -nearest neighbors. To improve the imputation accuracy, a two-layer stacked autoencoder combined with distinctive imputation is applied to locate the principal features of a dataset for clustering. Then, the top $k$ -nearest neighbor hybrid distance weighted imputation is approached to fill in missing values in clusters. The proposed method is evaluated on five popular University of California Irvine datasets as well as one air quality monitoring dataset collected from CPS through comparison with four high-quality existing imputation methods. Empirical results present that the proposed scheme can impute the missing data values effectively and efficiently, especially for the incomplete data with local characteristic in CPS.

With the rapidly increasing popularity of economic activities, a large amount of economic data is being collected. Although such data offers super opportunities for economic analysis, its low-quality, high-dimensionality and huge-volume pose great challenges on efficient analysis of economic big data. The existing methods have primarily analyzed economic data from the perspective of econometrics, which involves limited indicators and demands prior knowledge of economists. When embracing large varieties of economic factors, these methods tend to yield unsatisfactory performance. To address the challenges, this paper presents a new framework for efficient analysis of high-dimensional economic big data based on innovative distributed feature selection. Specifically, the framework combines the methods of economic feature selection and econometric model construction to reveal the hidden patterns for economic development. The functionality rests on three pillars: (i) novel data pre-processing techniques to prepare high-quality economic data, (ii) an innovative distributed feature identification solution to locate important and representative economic indicators from multidimensional data sets, and (iii) new econometric models to capture the hidden patterns for economic development. The experimental results on the economic data collected in Dalian, China, demonstrate that our proposed framework and methods have superior performance in analyzing enormous economic data.

Recent years have witnessed a surge of interests in cross-modal ranking. To bridge the gap between heterogeneous modalities, many projection based methods have been studied to learn common subspace where the correlation across different modalities can be directly measured. However, these methods generally consider pair-wise relationship merely, while ignoring the high-order relationship. In this paper, a combinative hypergraph learning in subspace for cross-modal ranking (CHLS) is proposed to enhance the performance of cross-modal ranking by capturing high-order relationship. We formulate the cross-modal ranking as a hypergraph learning problem in latent subspace where the high-order relationship among ranking instances can be captured. Furthermore, we propose a combinative hypergraph based on fused similarity information to encode both the intra-similarity in each modality and the inter-similarity across different modalities into the compact subspace representation, which can further enhance the performance of cross-modal ranking. Experiments on three representative cross-modal datasets show the effectiveness of the proposed method for cross-modal ranking. Furthermore, the ranking results achieved by the proposed CHLS can recall 80% of the relevant cross-modal instances at a much earlier stage compared against state-of-the-art methods for both cross-modal ranking tasks, i.e. image query text and text query image.

There has been significant interest in the use of HTTP adaptive streaming for live or on-demand video over the Internet in recent years. To mitigate the streaming transmission delay and reduce the networking overhead, an effective and critical approach is to utilize cache services between the origin servers and the heterogeneous clients. As the underlying protocol for web transactions, HTTP has great potentials to explore the resources within state-of-the-art CDNs for caching; yet distinct challenges arise in the HTTP adaptive streaming context. After examining a long-term and large-scale adaptive streaming dataset as well as statistical analysis, we demonstrate that the switching requests among the different qualities frequently emerge and constitute a significant portion in a per-day view. Consequently, they have substantially affected the performance of cache servers and Quality-of-Experience (QoE) of viewers. In this paper, we propose a novel cache model that captures the dependency among the segments in the cache server for adaptive HTTP streaming. Our work does not assume any specific selection algorithm on the client’s side and hence can be easily incorporated into existing streaming cache systems. Its centralized nature is also well accommodated by the latest DASH specification. Moreover, we extend our work to the multi-server caching context and present a similarity-aware allocation mechanism to enhance the caching efficiency. The performance evaluation shows our dependency- and similarity-aware strategy can significantly improve the cache hit-ratio and QoE of HTTP streaming as compared to previous approaches.

This article aims to solve the Moore–Penrose inverse of time-varying full-rank matrices in the presence of various noises in real time. For this purpose, two varying-parameter zeroing neural networks (VPZNNs) are proposed. Specifically, VPZNN-R and VPZNN-L models, which are based on a new design formula, are designed to solve the right and left Moore–Penrose inversion problems of time-varying full-rank matrices, respectively. The two VPZNN models are activated by two novel varying-parameter nonlinear activation functions. Detailed theoretical derivations are presented to show the desired finite-time convergence and outstanding robustness of the proposed VPZNN models under various kinds of noises. In addition, existing neural models, such as the original ZNN (OZNN) and the integration-enhanced ZNN (IEZNN), are compared with the VPZNN models. Simulation observations verify the advantages of the VPZNN models over the OZNN and IEZNN models in terms of convergence and robustness. The potential of the VPZNN models for robotic applications is then illustrated by an example of robot path tracking.

The conventional methods of visual defect detection heavily rely on the standard templates and features of an image. But the standard templates are difficult to obtain or even don’t exist in many application fields, especially for the flexible Integrated Circuit (IC) substrates based on micro vision. To find a solution for the above problems, the authors have proposed a novel defect detection approach based on topology mapping. Firstly, topological mapping on the pixel values of the image is performed to remove the noise. Then, a morphological operation is used to clear small spots. Furthermore, binarization is introduced to display the defect of corrosion and remove the black background and surface texture structure of the substrate image. The experiments are implemented to prove that the proposed method has higher Peak Signal to Noise Ratio and produces better quality of image, which confirms the efficiency of the proposed algorithm.

This paper focuses on an operation optimisation problem for a class of multi-head surface mounting machines in printed circuit board assembly lines. The problem involves five interrelated sub-problems: assigning nozzle types as well as components to heads, assigning feeders to slots and determining component pickup and placement sequences. According to the depth of making decisions, the sub-problems are first classified into two layers. Based on the classification, a two-stage mixed-integer linear programming (MILP) is developed to describe it and a two-stage problem-solving frame with a hybrid evolutionary algorithm (HEA) is proposed. In the first stage, a constructive heuristic is developed to determine the set of nozzle types assigned to each head and the total number of assembly cycles; in the second stage, constructive heuristics, an evolutionary algorithm with two evolutionary operators and a tabu search (TS) with multiple neighbourhoods are combined to solve all the sub-problems simultaneously, whe...

This paper is mainly devoted to a monotonically convergent iterative learning control (ILC) design for a class of uncertain discrete-time switched systems with state delay (UDTSDSs). By taking advantage of output error and state information, a hybrid ILC law for a class of UDTSDSs is proposed. After the ILC process is transformed into a 2D system, sufficient conditions in terms of linear matrix inequalities (LMIs) are derived by using a multiple Lyapunov–Krasovskii-like functional approach and a quadratic performance function. It is shown that if certain LMIs are met, the tracking error 2-norm converges monotonically to zero along the iteration direction, while the learning gains could be determined directly by solving the LMIs. The simulation results are provided to illustrate the theoretical analysis.

Containers underwent a whirlwind adoption across cloud providers recently, as the Open Container Initiative works toward standardizing the container format and configuration. The container could become the cloud infrastructure's "narrow waist," bridging a proliferation of existing and emerging cloud services.

The rapid development of the Internet of Things causes the dramatic growth of data, which poses an important challenge on the storage and quick retrieval of big data. As an effective representation model, RDF receives the most attention. More and more storage and index schemes have been developed for RDF model. For the large-scale RDF data, most of them suffer from a large number of self-joins, high storage cost, and many intermediate results. In this paper, we propose a scalable two-level index scheme (STLIS) for RDF data. In the first level, we devise a compressed path template tree (CPTT) index based on S-tree to retrieve the candidate sets of full path. In the second level, we create a hierarchical edge index (HEI) and a node-predicate (NP) index to accelerate the match. Extensive experiments are executed on two representative RDF benchmarks and one real RDF dataset in IoT by comparison with three representative index schemes, that is, RDF-3X, Bitmat, and TripleBit. Results demonstrate that our proposed scheme can respond to the complex query in real time and save much storage space compared with RDF-3X and Bitmat.

Effective fault classification is of great significance to isolate and eliminate faults in the smart manufacturing process, especially for chemical industry. However, incipient faults and limited labeled samples make it hard to classify faults accurately. This motivates the formulation of a semi-supervised convolutional ladder network with local and global feature fusion. In the algorithm, a convolutional ladder network is developed to capture higher-order correlations from both labeled and unlabeled samples simultaneously to overcome the problem caused by limited labeled samples, skipped connections are embedded within which to make a balance between supervised and unsupervised feature learning for further identifying faults. To improve the classification performance on incipient faults, a local and global feature fusion strategy is proposed to enhance the representation of incipient faults. Furthermore, a semi-supervised dynamic data representation strategy is introduced to jointly deal with labeled and unlabeled process samples, which enables the proposed method to handle process dynamics by characterizing temporal information of process variables. Experiments on the Tennessee Eastman process show that the proposed method is effective for process fault classification when labeled samples are limited compared to the state-of-the-art algorithms.

A fast hierarchical algorithm based on saliency analysis for 3D-object recognition is proposed to eliminate the large and complex computation of the present pattern recognition system based on RGB-D image classification method. Firstly, the process of establishing the key point set of the 3D-object is described. Secondly, a hierarchical recognition algorithm based on the proper four description operators is presented to sketch the process requirements and calculation rules. The time complexity of the four descriptors contained in the proposed hierarchical algorithm is also analyzed to illustrate the rapidity of this algorithm. Finally, experiments are performed to show that the accuracy and rapidity of the algorithm improved greatly, compared with the general recognition algorithm based on PFH descriptor or S-NARF descriptor alone.

Virtual tour applications have gained popularity in recent years by providing immersed user experience and visual navigation hints. Typically, a virtual tour contains both image data and inter-scene spatial connections, and is usually presented as a constellation of images. The construction of virtual tours originally requires sophisticated devices and time-consuming site survey. Lately, the technology advance in smartphones facilitates the virtual tour capture on such versatile end devices. In this work, we introduce a crowdsourced approach toenable large-scale panoramic virtual tour construction with the active participation of smartphone users. We have designed and implemented the framework, as well as proposed further optimizations on constellation construction. Extensive experiments are conducted to evaluate the overall performance, including image quality, constellation accuracy, user experience, and energyefficiency.

Sensor data processing plays an important role on the development of the wireless sensor networks in the big data era. Owning to the existence of a large number of incomplete data in wireless sensor networks, fuzzy c-means clustering algorithm (FCM) finds it difficult to produce an appropriate cluster result. The paper proposes a distributed weighted fuzzy c-means algorithm based on incomplete data imputation for big sensor data (DWFCM). DWFCM improves Affinity Propagation (AP) clustering algorithm by designing a new similarity metrics for imputing incomplete sensor data, and then proposes a weighted FCM (wFCM) by assigning a lower weighted value to the incomplete data object for improving the cluster accuracy. Finally, we validate the proposed weighted FCM algorithm on the dataset collected from the smart WSN lab. Experiments demonstrate that the weighted FCM algorithm could fill the missing values very accurately and improve the clustering results effectively.

For a class of linear discrete-time uncertain systems, a feedback feed-forward iterative learning control ILC scheme is proposed, which is comprised of an iterative learning controller and two current iteration feedback controllers. The iterative learning controller is used to improve the performance along the iteration direction and the feedback controllers are used to improve the performance along the time direction. First of all, the uncertain feedback feed-forward ILC system is presented by an uncertain two-dimensional Roesser model system. Then, two robust control schemes are proposed. One can ensure that the feedback feed-forward ILC system is bounded-input bounded-output stable along time direction, and the other can ensure that the feedback feed-forward ILC system is asymptotically stable along time direction. Both schemes can guarantee the system is robust monotonically convergent along the iteration direction. Third, the robust convergent sufficient conditions are given, which contains a linear matrix inequality LMI. Moreover, the LMI can be used to determine the gain matrix of the feedback feed-forward iterative learning controller. Finally, the simulation results are presented to demonstrate the effectiveness of the proposed schemes.

Rape is a non-grain feedstock with several characteristics that make it suitable for biofuel production, such as high oil yields and low environmental impact. Planting rape on winter fallow fields could allow for seasonal rotation between the bioenergy production and agricultural production. In the present study, the Global Agro-Ecological Zones model was used to estimate the production potential of rapeseed on winter fallow fields in the Yangtze River region from 2010 to 2015. Life cycle assessment was then conducted to calculate energy efficiency and greenhouse gas emissions from the entire energy-producing process and to estimate the number of bioenergy enterprises and their spatial distribution. The results indicated that the total area available of winter fallow fields in the Yangtze River region was 24.93 million ha, accounting for 34.2% of the total cultivated land area. The total yield of winter rapeseed was up to 46.41 million tons (an average yield of 1.86 tons/ha). Thus, the study area could potentially produce net energy of up to 3564 million MJ with energy return on energy investment (EROEI) as high as 1.52–1.84. The total reduction of greenhouse gas emissions may be up to 23.28 million tons, and the total number of bioenergy enterprises could be 589, from which the total biodiesel output could be 167.5 million tons. Hubei, Henan, Anhui and Jiangxi should be prioritized for generating biofuels from rapeseed planted on winter fallow fields.

Fast generalized fuzzy C-means (FGFCM) is one of the popular image segmentation algorithms. However, a single FGFCM algorithm is easy to get stuck at local optimum and the choices of the initial cluster centers affect segmentation consequences seriously. Furthermore, conventional FGFCM ignores the relationship between the selections of neighbors and edges so that some image details are not considered. In order to solve these problems, this paper presents an improved image segmentation approach using FGFCM with an edges-based neighbor selection strategy and PSO. PSO is used to search initial cluster centers scientifically for FGFCM and helps to avoid local optimum. A new additional edges-based neighbor selection strategy is proposed for the first time, aiming at making more contributions to the preservation of image details. Experiments performed on synthetic and real-world images prove the proposed algorithm is more effective and can preserve more image details.

Traditional real estate appraisal methods obtain estimates of real estate by using mathematical modeling to analyze the existing sample data. However, the information of sample data sometimes cannot fully reflect the real-time quotes. For example, in a thin real estate market, the correlated sample data for estimated object is lacking, which limits the estimates of these traditional methods. In this paper, an optimal rubrics-based approach to real estate appraisal is proposed, which brings in crowdsourcing. The valuation estimate can serve as a market indication for the potential real estate buyers or sellers. It is not only based on the information of the existing sample data (just like these traditional methods), but also on the extra real-time market information from online crowdsourcing feedback, which makes the estimated result close to that of the market. The proposed method constructs the rubrics model from sample data. Based on this, the cosine similarity function is used to calculate the similarity between each rubric for selecting the optimal rubrics. The selected optimal rubrics and the estimated point are posted on a crowdsourcing platform. After comparing the information of the estimated point with the optimal rubrics on the crowdsourcing platform, those users who are connected with the estimated object complete the appraisal with their knowledge of the real estate market. The experiment results show that the average accuracy of the proposed approach is over 70%; the maximum accuracy is 90%. This supports that the proposed method can easily provide a valuable market reference for the potential real estate buyers or sellers, and is an attempt to use the human-computer interaction in the real estate appraisal field.

Clustering is a fundamental and important technique under many circumstances including data mining, pattern recognition, image processing and other industrial applications. During the past decades, many clustering algorithms have been developed, such as DBSCAN, AP and CFS. As the latest clustering algorithm proposed in Science magazine in 2014, clustering by fast search and find of density peaks, named as CFS, is a simple and outstanding algorithm for its promising performance on data sets of arbitrary shape. However, CFS's performance is usually affected by the cutoff distance dc, the density peaks, the selection of cluster centers and the allocation strategy of data points. In this paper, we propose an improved algorithm (ICFS) to deal with the several weaknesses of it. Unlike CFS, the proposed algorithm designs a formula for the cutoff distance calculation and a method for cluster centers selection to improve its robustness. Moreover, a new non-center point's allocation strategy and the cluster merging and splitting processes are developed to adapt to the density peaks and adjust the clusters dynamically, which can improve the clustering accuracy and scalability. The ICFS method is evaluated on several datasets by comparison with the original CFS algorithm. Results demonstrate the effectiveness of the proposed method.

A rapid 3D reconstruction method for IC chips based on color fringe projection is proposed in this paper. In the defect inspection of IC chips, the present methods of 3D reconstruction can't meet the requirement of high speed and high accuracy. In this paper, periodic color fringe patterns are designed for the projection. The relationship between fringe pattern offset and surface height of the object can be obtained from the mathematical model of the color fringe projection system. A new method using the duty ratio is proposed to calculate the fringe pattern offset. Only a single fringe image is needed to realize the 3D reconstruction of IC chips. Experimental results show that the proposed method can realize real time 3D reconstruction and obtain the 3D information accurately.

Vision inspection has been widely used in the field of defect measurement as a noncontact, nondestructive measurement technique. Conventional visual defect detection methods rely heavily on standard templates and image features, but the standard templates are either difficult to obtain or do not exist in many application fields, especially in flexible integrated circuit (IC) substrate defect detection at a microscopic level. To solve the above problems, a rapid online oxidation defect detection algorithm for micron-level flexible IC substrate oxidation defects, based on the differential geometric approach, is proposed in this paper. First, the sample fitting model is established by using minimal generalized variance as well as the surface fitting model. Second, the defect area of the substrate is determined by analyzing the similarity and membership, which is coded among the pixels by considering the spatial topological structure of the image. Finally, experiments are performed to show that the algorithm proposed in this paper not only has high precision, but also meets industrial real-time requirements for the rapid production of a flexible substrate. Since the method does not depend on the standard template of the original image, it can achieve vision detection effects at high speed and with high precision.

Summary The continuous growth of resource description framework (RDF) data poses an important challenge on RDF data partitioning that is a vital technique for effective cloud storage. Recently, many partitioning algorithms for large RDF data have been developed, and most of them are based on graph partitioning. However, existing graph partitioning methods could not partition asymmetric RDF data effectively, resulting in a lower performance for cloud storage. This paper proposes a balanced RDF graph partitioning algorithm for storing massive RDF data on cloud. We first devise a modularity-based multi-level label propagation algorithm (MMLP) to partition RDF graph roughly and then use a balanced K-mediods clustering algorithm for final k-way partitioning. Balanced RDF graph partitioning algorithm designs an effective label update rule and a balanced modification strategy to achieve a high quality coarsening result and make the partition as equilibrium as possible. Experiments are carried on two representative RDF benchmarks and one real RDF dataset by comparison with two representative graph partitioning methods, that is, METIS and MLP+METIS. Results demonstrate that our proposed scheme can produce a high-quality partition for massive RDF data storage on cloud. Copyright © 2016 John Wiley & Sons, Ltd.