Search

Many information technology courses frequently use the learning by doing strategy in vocational high schools. Particularly, learning computer application operating skills is essential for students because excellent computer application operating skills can help them attain good jobs. However, when fostering students’ computer application operating skills by teaching in vocational high school using the learning by doing strategy, a teacher learns that helping all students, evaluating their learning problems, and providing feedback to correct their mistakes are challenging. After investigating the challenge, a machine-learning-based peer tutor recommender system (MPTRS) with automated assessment was proposed to enhance students’ learning performance in computer application operating skills. The advanced automated assessment system (AAS) used computer vision technology to evaluate student assignments and instantly return feedback. The recommendation mechanism of the MPTRS enhanced mutual help among students based on their social relationships, learning performance, and recommendation feedback. Furthermore, machine-learning techniques were used to improve recommendations. In the experiment, the experimental group used the proposed system, and the control group used a conventional commercial automated grading system. From the experimental results, the learning performance of the experimental group significantly improved between the pretest and post-test. Students can correct and complete more assignments using the advanced AAS and students who behind in learning also can use the peer tutor recommender function for asking help. Participants were also satisfied with the proposed advanced AAS and MPTRS. It is worth to promote the proposed system to teachers of adopting the learning-by-doing strategy in computer application classes.

Visual food recognition is emerging as an important application in dietary monitoring and management in recent years. Existing works use large backbone networks to achieve good performance. However, these networks are not able to be deployed on personal portable devices due to large size and computation cost. Some compact networks have been developed, however, their performance are usually lower than the large backbone networks. In view of this, this paper proposes a joint distillation framework that targets to achieve a high visual food recognition accuracy using a compact network. As opposed to the more traditional one-directional knowledge distillation methods, the proposed knowledge distillation framework trains both the large teacher network and the compact student network simultaneously. The framework introduces a new Multi-Layer Distillation (MLD) for simultaneous teacher-student learning at multiple layers of different abstraction. A novel Instance Activation Mapping (IAM) is proposed to jointly train the teacher and student networks using generated instance-level activation map that incorporates label information for each training image. Experimental results on the two benchmark datasets UECFood-256 and Food-101 show that the trained compact student network achieves state-of-the-art performance at 83.5% and 90.4%, respectively, while achieving more than 4 times deduction regarding network model size.

Wheelset bearings are crucial mechanical components of high-speed trains. Wheelset-bearing fault detection is of great significance to ensure the safety of high-speed train service. Convolution sparse representations (CSRs) provide an excellent framework for extracting impulse responses induced by bearing faults. However, the performance of CSR on extracting impulse responses is fairly sensitive to inappropriate selection of method-related parameters, and a convolution model for representing the impulse responses has not been discussed. In view of these two unsolved problems, a convolutional representation model of the impulse response series is developed. A novel fault detection method, named adaptive CSR (ACSR), is then proposed based on combinations of CSR and methods for estimating three parameters related to CSR. Finally, the effectiveness of the proposed ACSR method is validated via simulation, bench testing, and a real-life running test employing a high-speed train.

The impulsive components induced by bearing faults are key features for assessing gear-box bearing faults. However, because of heavy background noise and the interferences of other vibrations, it is difficult to extract these impulsive components caused by faults, particularly early faults, from the measured vibration signals. To capture the high-level structure of impulsive components embedded in measured vibration signals, a dictionary learning method called shift-invariant K-means singular value decomposition (SI-K-SVD) dictionary learning is used to detect the early faults of gear-box bearings. Although SI-K-SVD is more flexible and adaptable than existing methods, the improper selection of two SI-K-SVD-related parameters, namely, the number of iterations and the pattern lengths, has an adverse influence on fault detection performance. Therefore, the sparsity of the envelope spectrum (SES) and the kurtosis of the envelope spectrum (KES) are used to select these two key parameters, respectively. SI-K-SVD with the two selected optimal parameter values, referred to as optimal parameter SI-K-SVD (OP-SI-K-SVD), is proposed to detect gear-box bearing faults. The proposed method is verified by both simulations and an experiment. Compared to the state-of-the-art methods, namely, empirical model decomposition, wavelet transform and K-SVD, OP-SI-K-SVD has better performance in diagnosing the early faults of a gear-box bearing.

Aiming at the cyclic mechanical behavior of SiCp/2009Al at room temperature, uniaxial tensile test and low cycle fatigue test were carried out to obtain its uniaxial tensile properties and cyclic hysteresis properties. Based on the B-P viscoplastic constitutive theory, the cyclic mechanical behavior of the material is described, and the prediction results of the steady-state hysteretic response are compared with the experimental results. The results of the study indicate that the uniaxial tensile properties of SiCp/2009Al exhibit the characteristic which is independent of strain rate at room temperature, and T6 heat treatment can improve the strength of the material effectively; The steady state hysteresis curve which is described by the Bodner-Partom constitutive theory is close to the test curve, and the prediction error of the peak stress is less than 9%. It is reasonable to establish the cyclic constitutive model of SiCp/2009Al based on the B-P constitutive theory.

Purpose This study aims to describe case studies of two models, LibQUAL+ and a model based on the analytic hierarchy process (AHP), used to assess library service quality and compare the two assessment models. Design/methodology/approach First, an AHP-based evaluation model is set up to assess library service quality, following which the model is adapted to assess the service quality of Nanjing Agricultural University Library. Second, a modified LibQUAL+ model is created and implemented in the form of a questionnaire survey to assess the library service quality of the same library. Finally, the two assessment models are compared. Findings The results of the two case studies indicate that LibQUAL+ is suitable for assessing Chinese university libraries after modified, and the library AHP model is more suitable for comparing libraries in China between themselves. Originality/value This paper uses two methods for assessing library service quality; this should prove to be helpful for the librarians interested in assessing the service quality of Chinese university libraries.

In recent years, palmprint recognition has made great progress and many methods have been put forward. The extraction of robust orientation features and finding efficient matching strategies are two key points for palmprint recognition. Traditional coding methods usually only use a dominant filter response to extract orientation features of palmprint images while not taking into account the other useful filter responses. Without increasing the number of filers, this study presents a modified Competitive Code to extract orientation features more accurately, which makes use of the relation between the filter responses. Besides, a distinctive extended eight-pixel neighbourhood method is proposed to select the sample points for matching by extracting the local features. At the matching stage, an effective fusion matching scheme with a double-layer image pyramid is designed to calculate the similarity between two palmprint images. Extensive experiments on four types of public palmprint databases show that the proposed method has excellent performance compared with the other state-of-the-art algorithms.

Floating foreign objects are one of the biggest safety hazards to overhead transmission lines. The existing tools and methods of removing foreign objects are difficult to meet the requirements for economical, rapid and safe removal. To solve this problem, a method of using laser to remotely remove foreign objects on overhead lines is proposed in this paper. After selecting a suitable type of laser, the appropriate power density is determined through the irradiation experiment on the typical foreign objects. Then, the influence of the laser on the main components of the transmission line is studied. The result shows that this certain power density laser has no damage to the line. On this basis, a laser equipment for removing foreign objects on transmission lines is developed, and simulation experiments and field tests are applied. The results prove that the removal equipment using fiber laser can remove foreign objects on live lines remotely, quickly, and safely.

A wheelset bearing is a crucial energy transmission element in high-speed trains. Any parts of the wheelset bearing that have faults may endanger the safety of the railway service. Therefore, it is important to monitor the running condition of a wheelset bearing. The multifault on a wheelset bearing is very common, and these impulsive components generated by different types of faults may interact with each other, which increases the difficulty of entirely identifying those faults. To solve the multifault problem, this paper proposed a hierarchical shift-invariant K-means singular value decomposition (H-SI-K-SVD) to hierarchically separate those multifault impulsive components based on their fault power levels. Each of the separated impulse signals contains only one fault impulse, and the fault information could be highlighted both in time domain and frequency domain. In addition, the sparsity of envelope spectrum (SES) is introduced as an indicator to adaptively tune a key parameter in this method. The effectiveness of the proposed method is verified by both simulation and experimental signals. Compared with ensemble empirical model decomposition (EEMD), the proposed method exhibits better performance in separating the multifault impulsive components and detecting the faults of a wheelset bearing.

Ethnicity information is an integral part of human identity, and a useful identifier for various applications ranging from video surveillance, targeted advertisement to social media profiling. In recent years, Convolutional Neural Networks (CNNs) have shown state-of-the-art performance in many visual recognition problems. Currently, there are a few CNN-based approaches on ethnicity classification [1], [2]. However, the approaches suffer from the following limitations: (i) most face datasets do not include ethnicity information, and those with ethnicity information are typically small to medium in size, thereby they do not provide sufficient samples for training of CNNs from the scratch, and (ii) the CNN methods often treat ethnicity classification as a multi-class classification where the likelihood of each class label is generated. However, it does not utilize the intermediate activation functions of CNNs which provide rich hierarchical features to assist in ethnicity classification. In view of this, this paper proposes a new hybrid supervised learning method to perform ethnicity classification that uses both the strength of CNN as well as the rich features obtained from the network. The method combines the soft likelihood of CNN classification output with an image ranking engine that leverages on matching of the hierarchical features between the query and dataset images. A supervised Support Vector Machine (SVM) hybrid learning is developed to train the combined feature vectors to perform ethnicity classification. The performance of the proposed method is evaluated using a dataset consisting of Bangladeshi, Chinese and Indian ethnicity groups, and it outperforms the state-of-the-art methods [2], [3] by up to 3% in recognition accuracy.

Brand plays a significant role in fashion clothing. Consumers are brand conscious during the clothing search and purchase. Existing visual fashion search methods [1], [8], [17]–[19] often do not explicitly consider the brand information such as logos. Brand logo in clothing images are quite small and often suffer various deformations, and hence pose a significant challenge for branded clothing search. In view of this, this paper presents a new Brand-Aware Fashion Search (BAFS) framework that explores the brand information during the visual search. We construct a new brand fashion dataset which consists of 10K images of branded clothing with trademark logos. The proposed framework first jointly detects the brand logo and clothing items in the images. Next, to extract the rich visual information from the clothing images, we propose a new deep feature encoding known as Principal Component Maximum Activation of Convolutions (PMAC) that leverages hierarchies of CNN activations. The PMAC feature aims to capture both low-level visual information and high-level global abstraction from the images. The proposed method further uses a brand-aware re-ranking technique to improve the search. Experiments conducted on the brand fashion dataset shows that the proposed framework achieves superior performance to other comparative methods.

Palmprint deformation, including rotation, displacement, swelling and so on, is a crucial factor for the palmprint recognition performance. The traditional algorithms mainly focus on the feature-based matching and the block-based matching. In this paper, a novel block-based matching palmprint recognition scheme is proposed. The sampling points on the template image are built by the harris corner detection algorithm. Furthermore, the corresponding sampling points on the query palmprint image are estimated by image pyramids algorithm for reducing the impact of distortion. At last, a matching method based on statistical offset is proposed to calculate the final matching score. Extensive experiments on public palmprint database shows that our method can achieve excellent recognition performance compared with many other classic methods.

Some high school students lack a motivation to review the English vocabulary learning in second language after school. However, many studies have pointed out that the game-based learning (GBL) can improve students' learning motivations and effectiveness. We present a digital game-based learning system for high school students' after-school learning in English vocabulary. Teachers should only guide students playing the game and no teaching styles will be changed. In the whole process, students design the learning flash cards on Web by themselves, play learning flash card game on mobile device and they will reward for passing the test with the game cards. They have to play the mobile game with collecting or exchanging the game cards with other team members to finish the puzzle and get reward cards. Reward cards are used to exchange real pocket money from their parents or ask the teacher for scores.

A numerical simulation and an experiment on adaptive optics (AO) in 2.5Gbps laser communication system are introduced in this paper. The simulation results show that the signal fading is mitigated at a large part with AO correction. The experiment results show that the average bit-error-rate (BER) is 6×10−3 without AO correction. The average BER will be reduced to 1×10−4 with AO tip-tilt correction. And there is no bit-error in 120s with AO high order correction.

Game-based Learning, to date, is an emergent solution to raise the learning motivation by involving the entertainment factors into original learning process. However, experiences revealed that learners paid more emphasis on the game-playing than on the pedagogical content. Thus, in this study, the issue to discover the fundamental factors that may cause the increase or loss of motivation will be discussed. The Problem-based strategy was adopted to be the subject of game design and development while the content was based on the historical knowledge in specific area. The ADDIE is adopted to make improvement of the convention of historical knowledge via a Game-based strategy. The issues, both pedagogy and entertainment, are involved to bring up the learning motivation of students of specific domain.

The carry-free property of modified signed-digit addition is discussed, and a space-position–logic-encoding scheme is proposed, which not only makes best use of the convenience of binary (0, 1) logic operation but is also suitable for the trinary (1,0,1¯) property of modified signed-digit digits. Based on the space-position–logic-encoding scheme, a fully parallel modified signed-digit adder and subtracter is built by use of optoelectronic switch modules and butterfly interconnections; thus an effective combination of a parallel algorithm and a parallel architecture is implemented. The effectiveness of this architecture is verified by both simulation and experimental results.

It is reported that Ada95 is a full object-oriented programming language. Ada95's task type mechanism could be considered as a model of a concurrent object-oriented problem solution. However, there is still a gap between the language constructions of Ada95 and the general O-O concept as well as description. The complexity of Loeper's (1997) template of Ada95 codes for concurrent object is convincing evidence. Based on Loeper's concurrent object template, the authors present a user description of concurrent object classes in the general O-O concept and the implementation of automatic generation of Ada95 codes from the description.

An algorithm for quickly training the BP neural network system identifier (BPNNSI) of the chaotic optical systems is presented in this paper. The ability of this algorithm, termed as the chaos speedup BP algorithm (CSBPA), has been demonstrated with the computer simulation of identifying the Bragg diffraction acousto-optic system (BDAOS) in which a 1:4:1 BP network was employed in identification. Taking the normalized output time series of the BDAOS as the training series, the BPNNSI was trained with the CSBPA as follows: (1) trained the BPNNSI to learn a chaotic state of the BDAOS with the BP algorithm where the initial weight distribution was set randomly; (2) took the final weight distribution obtained in (1) as the initial weight distribution for the other states of the BDAOS to be identified; (3) trained the BPNNSI to learn the other states still with the BP algorithm but with the initial weight distribution obtained in (2).© (1996) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

In this paper, a space-position-logic-encoding scheme is proposed, which not only makes best use of the convenience of binary logic operation, but is also suitable for the trinary property of modified signed-digit numbers. Based on the space-position-logic-encoding scheme, a fully parallel modified signed-digit adder and subtracter is built by use of optoelectronic switch and microstructure interconnect technologies. Thus an effective combination of a parallel algorithm and a parallel architecture is implemented. Finally, both simulation results and experimental results are provided.© (1995) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

In this paper, the carry-free property of modified-signed-digit (MSD) addition is analyzed with a space position logic encoding scheme. On this basis, MSD multiplication is discussed and a fast MSD multiplication system composed of optoelectronic logic technology and a multilayer optical interconnection architecture is propsed and studied. Finally, the effectivity of the fast MSD multiplication system is demostrated by using a 2X2 bit multiplication example and experimental results are given.© (1994) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

In this paper, we study the constructions and mathematical expressions of butterfly interconnections and networks. Because 2-D and 3-D networks have different constructions, we discuss the transform from 3-D forms to 2-D ones. We start with optical Fourier transforms to derive the discrete transforms of 2-D and 3-D butterfly interconnect networks for implementing 1-D and 2-D fast Fourier transforms, respectively, which are the basic operations of intelligent optical computings.© (1992) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

We present a new architecture of an adaptive optical processor, constructed according to adaptive learning and adaptive resonance architecture. It has four main properties.

A space-position-logic-encoding scheme is proposed and demonstrated, This encoding scheme not only makes the best use of the convenience of binary logic operation, but is also suitable for the trinary property of modified signed-digit (MSD) numbers. Based on the space-position-logic-encoding scheme, a fully parallel modified signed-digit adder and subtracter is built using optoelectronic switch technologies in conjunction with fiber-multistage 3-D optoelectronic interconnects, Thus an effective combination of a parallel algorithm and a parallel architecture is implemented. In addition, the performance of the optoelectronic switches used in this system is experimentally studied and verified. Both the 3-bit experimental model and the experimental results of a parallel addition and a parallel subtraction are provided and discussed. Finally, the speed ratio between the MSD adder and binary adders is discussed and the advantage of the MSD in operating speed is demonstrated. (C) 1996 Society of Photo-Optical instrumentation Engineers.

As modern optical information processing has developed, research on massive and parallel rapid computing and processing has attracted more attention. In this paper, butterfly networks and a variety of types of optical information processing are studied and discussed. For a basis, one- and two-dimensional butterfly interconnection networks are studied in constructions, and the relationship and the transformation between them are provided. Algorithms for both the one- and two-dimensional fast Fourier transforms are analyzed, one- and two-dimensional butterfly networks for implementing the algorithms are built, and computer-simulation results are attained. Finally, an underlying optical network system is suggested and studied in respect to its architecture and advantages; it is a new optical butterfly network hardware system consisting of two-dimensional binary phase diffraction gratings, which perform a variety of types of fast-Fourier-transform-based optical information processing.

Free-space optical interconnections are important in both massive digital optical computing and communication systems. The architectural features of three interconnection networks are analyzed and compared, and the optical butterfly interconnection is shown to have many advantages over other interconnections in implementing various basic logic functions such as addition, subtraction, multiplication, and fast Fourier transforms (FFTs). Starting with conventional Karnaugh maps and Boolean algebra, the characteristics of full addition and full subtraction are analyzed and compared. An n-bit parallel calculator that can implement both ripple carry full additions and ripple borrow full subtractions using multilayer butterfly interconnection networks is designed. Then the schematic and architecture of the full adder/subtractor, interconnection networks, and the patterns of key devices such as masks to implement AND and OR operations in this calculation are described in detail. The correct simulation results of several groups of multibit digits are provided. Finally, the development of the interconnections in implementing logic operations is discussed.

Free-space optical interconnections are important both in massive digital optical computing and in communication systems. The optical butterfly interconnection has many advantages over other interconnections in implementing various basic logic functions such as addition, subtraction, multiplication. This paper starts with the conventional Karnaugh maps and Boolean algebra to implement a parallel n-bit ripple carry full adder by the use of multilayer butterfly interconnection networks. Then we describe in detail the design and architecture of the full adder and provide accurate interconnection networks and the structures or patterns of key devices such as the masks to implement AND and OR operations in this calculation. Finally, we discuss development of the interconnection in implementing logic operations.