Selecting visually overlapping image pairs without any prior information is an essential task of large-scale structure from motion (SfM) pipelines. To address this problem, many state-of-the-art image retrieval systems adopt the idea of bag of visual words (BoVW) for computing image-pair similarity. In this paper, we present a method for improving the image pair selection using BoVW. Our method combines a conventional vector-based approach and a set-based approach. For the set similarity, we introduce a modified version of the Simpson (m-Simpson) coefficient. We show the advantage of this measure over three typical set similarity measures and demonstrate that the combination of vector similarity and the m-Simpson coefficient effectively reduces false positives and increases accuracy. To discuss the choice of vocabulary construction, we prepared both a sampled vocabulary on an evaluation dataset and a basic pre-trained vocabulary on a training dataset. In addition, we tested our method on vocabularies of different sizes. Our experimental results show that the proposed method dramatically improves precision scores especially on the sampled vocabulary and performs better than the state-of-the-art methods that use pre-trained vocabularies. We further introduce a method to determine the k value of top-k relevant searches for each image and show that it obtains higher precision at the same recall.

Self-review is essential to improving presentation, particularly for novice/unskilled researchers. In general, they could record a video of their presentation, and then check it out for self-review. However, they would be quite uncomfortable due to their appearance and voice in the video. They also struggle with in-depth self-review. To address these issues, we designed a presentation avatar that reproduces presentation made by researchers. The presentation avatar intends to increase self-awareness through self-reviewing. We also designed a checklist to aid in a detailed self-review, which includes points to be reviewed. This paper also demonstrates presentation avatar systems that use a virtual character and a robot, to allow novice/unskilled researchers as learners to self-review their own presentation using the checklist. The results of case studies with the systems indicate that the presentation avatar systems have the potential to promote self-review. In particular, we found that robot avatar promoted engagement in self-reviewing presentation.

Graph layouts reveal global or local structures of graph data. However, there are few studies on assisting readers in better reconstructing a graph from a layout. This paper attempts to generate a layout whose edges can be reestablished. We reformulate the graph layout problem as an edge classification problem. The inputs are the vertex pairs, and the outputs are the edge existences. The trainable parameters are the laid-out coordinates of the vertices. We propose a binary classification-based graph layout (BCGL) framework in this paper. This layout aims to preserve the local structure of the graph and does not require the total similarity relationships of the vertices. We implement two concrete algorithms under the BCGL framework, evaluate our approach on a wide variety of datasets, and draw comparisons with several other methods. The evaluations verify the ability of the BCGL in local neighborhood preservation and its visual quality with some classic metrics.

In order to further reduce the transmission rate of multi-channel satellite broadcast signals, whose carrier-to-noise ratio (CNR fluctuates due to rainfall attenuation, we propose a novel digitized radio-over-fiber (DRoF) -based optical re-transmission system based on adaptive combination compression for ultra-high definition (UHD) broadcasting satellite (BS)/communications satellite (CS) broadcast signals. The proposed system reduces the optical re-transmission rate of BS/CS signals as much as possible while handling input CNR fluctuations. Therefore, the transmission rate of communication signals in time-division multiplexing (TDM) transmission is ensured, and network sharing of communication signals and broadcast signals via passive optical network (PON) is realized. Based on the ITU-R P.618-13 prediction model, an experimental evaluation is performed using estimates of the long-term statistics of attenuation due to rainfall. The attenuation is evaluated as a percentage of the time that long-term re-transmission service is available. It is shown that the proposed system is able to accommodate a wide range of rainfall attenuation and achieve a 99.988% time percentage for the duration of service provision. In order to show the rate reduction effect of the proposed system, the quantization bit reduction effect as a function of the input CNR, which depends on rainfall attenuation, is experimentally confirmed. Experiments show that service operation time of 99.978% can be achieved by 3-bit transmission. This means a 62.5% reduction in transmission rate is realized compared to conventional fixed quantization. Furthermore, the average quantization bit number in our system for service operation times is 3.000, indicating that most service operation times are covered by just 3-bit transmission.

Given a set of n disjoint intervals on a line and an integer k, we want to find k points in the intervals so that the minimum pairwise distance of the k points is maximized. Intuitively, given a set of n disjoint time intervals on a timeline, each of which is a time span we are allowed to check something, and an integer k, which is the number of times we will check something, we plan k checking times so that the checks occur at equal time intervals as much as possible, that is, we want to maximize the minimum time interval between the k checking times. We call the problem the k-dispersion problem on intervals. If we need to choose exactly one point in each interval, so k=n, and the disjoint intervals are given in the sorted order on the line, then two O(n) time algorithms to solve the problem are known. In this paper we give the first O(n) time algorithm to solve the problem for any constant k. Our algorithm works even if the disjoint intervals are given in any (not sorted) order. If the disjoint intervals are given in the sorted order on the line, then, by slightly modifying the algorithm, one can solve the problem in O(log n) time. This is the first sublinear time algorithm to solve the problem. Also we show some results on the k-dispersion problem on disks, including an FPTAS.

Running IoT applications on edge computing infrastructures has the benefits of low response times and efficient bandwidth usage. System verification on a testbed is required to deploy IoT applications in production environments. In a testbed, Docker containers are preferable for a smooth transition of tested application programs to production environments. In addition, the round-trip times (RTT) of Docker containers to clients must be ensured, according to the target application's response time requirements. However, in existing testbed systems, the RTTs between Docker containers and clients are not ensured. Thus, we must undergo a large amount of configuration data including RTTs between all pairs of wireless base station nodes and servers to set up a testbed environment. In this paper, we present an edge computing testbed system with simple application programming interfaces (API) for testbed users that ensures RTTs between Docker containers and clients. The proposed system automatically determines which servers to place Docker containers on according to virtual regions and the RTTs specified by the testbed users through APIs. The virtual regions provide reduced size information about the RTTs in a network. In the proposed system, the configuration data size is reduced to one divided by the number of the servers and the command arguments length is reduced to approximately one-third or less, whereas the increased system running time is 4.3s.

In this letter, we present an adaptive weighted transfer subspace learning (AWTSL) method for cross-database speech emotion recognition (SER), which can efficiently eliminate the discrepancy between source and target databases. Specifically, on one hand, a subspace projection matrix is first learned to project the cross-database features into a common subspace. At the same time, each target sample can be represented by the source samples by using a sparse reconstruction matrix. On the other hand, we design an adaptive weighted matrix learning strategy, which can improve the reconstruction contribution of important features and eliminate the negative influence of redundant features. Finally, we conduct extensive experiments on four benchmark databases, and the experimental results demonstrate the efficacy of the proposed method.

An interpretation method of inversion phenomena is newly proposed for backward transient scattered field components for both E- and H-polarizations when an ultra-wideband (UWB) pulse wave radiated from a line source is incident on a two-dimensional metal cylinder covered with a lossless dielectric medium layer (coated metal cylinder). A time-domain (TD) asymptotic solution, which is referred to as a TD saddle point technique (TD-SPT), is derived by applying the SPT in evaluating a backward transient scattered field which is expressed by an integral form. The TD-SPT is represented by a combination of a direct geometric optical ray (DGO) and a reflected GO (RGO) series, thereby being able to extract and calculate any backward transient scattered field component from a response waveform. The TD-SPT is useful in understanding the response waveform of a backward transient scattered field by a coated metal cylinder because it can give us the peak value and arrival time of any field component, namely DGO and RGO components, and interpret analytically inversion phenomenon of any field component. The accuracy, validity, and practicality of the TD-SPT are clarified by comparing it with two kinds of reference solutions.

Register pushdown system (RPDS) is an extension of pushdown system (PDS) that has registers for dealing with data values. An LTL model checking method for RPDS with regular valuations has been proposed in previous work; however, the method requires the register automata (RA) used for defining a regular valuation to be backward-deterministic. This paper proposes another approach to the same problem, in which the model checking problem for RPDS is reduced to that problem for PDS by constructing a PDS bisimulation equivalent to a given RPDS. This construction is simpler than the previous model checking method and does not require RAs deterministic or backward-deterministic, and the bisimulation equivalence clearly guarantees the correctness of the reduction. On the other hand, the proposed method requires every RPDS (and RA) to have the freshness property, in which whenever the RPDS updates a register with a data value not stored in any register or the stack top, the value should be fresh. This paper also shows that the model checking problem with regular valuations defined by general RA is undecidable, and thus the freshness constraint is essential in the proposed method.

We propose a penalty-based and constraint Bayesian optimization methods with an agent-based supply-chain (SC) simulator as a new Monte Carlo optimization approach for multi-echelon inventory management to improve key performance indicators such as inventory cost and sales opportunity loss. First, we formulate the multi-echelon inventory problem and introduce an agent-based SC simulator architecture for the optimization. Second, we define the optimization framework for the formulation. Finally, we discuss the evaluation of the effectiveness of the proposed methods by benchmarking it against the most commonly used genetic algorithm (GA) in simulation-based inventory optimization. Our results indicate that the constraint Bayesian optimization can minimize SC inventory cost with lower sales opportunity loss rates and converge to the optimal solution 22 times faster than GA in the best case.

In the online removable knapsack problem, a sequence of items, each labeled with its value and its size, is given one by one. At each arrival of an item, a player has to decide whether to put it into a knapsack or to discard it. The player is also allowed to discard some of the items that are already in the knapsack. The objective is to maximize the total value of the knapsack. Iwama and Taketomi gave an optimal algorithm for the case where the value of each item is equal to its size. In this paper we consider a case with an additional constraint that the capacity of the knapsack is a positive integer N and that the sizes of items are all integral. For each positive integer N, we design an algorithm and prove its optimality. It is revealed that the competitive ratio is not monotonic with respect to N.

This paper proposes a novel interference cancellation technique that prevents radio receivers from degrading due to periodic interference signals caused by electromagnetic waves emitted from high power circuits. The proposed technique cancels periodic interference signals in the frequency domain, even if the periodic interference signals drift in the time domain. We propose a drift estimation based on a super resolution technique such as ESPRIT. Moreover, we propose a sequential drift estimation to enhance the drift estimation performance. The proposed technique employs a linear filter based on the minimum mean square error criterion with assistance of the estimated drifts for the interference cancellation. The performance of the proposed technique is confirmed by computer simulation. The proposed technique achieves a gain of more than 40dB at the higher frequency part in the band. The proposed canceler achieves such superior performance, if the parameter sets are carefully selected. The proposed sequential drift estimation relaxes the parameter constraints, and enables the proposed cancellation to achieve the performance upper bound.

The top-K sorting is a variant of sorting used heavily in applications such as database management systems. Recently, the use of field programmable gate arrays (FPGAs) to accelerate sorting operation has attracted the interest of researchers. However, existing hardware top-K sorting algorithms are either resource-intensive or of low throughput. In this paper, we present a resource-efficient top-K sorting architecture that is composed of L cascading sorting units, and each sorting unit is composed of P sorting cells. K=PL largest elements are produced when a variable length input sequence is processed. This architecture can operate at a high frequency while consuming fewer resources. The experimental results show that our architecture achieved a maximum 1.2x throughput-to-resource improvement compared to previous studies.

Accurately describing user behaviors with appropriate sensors is always important when developing computing cost-effective systems. This paper employs datasets recorded for fine-grained reading detection using the J!NS MEME, an eye-wear device with electrooculography (EOG), accelerometer, and gyroscope sensors. We generate models for all possible combinations of the three sensors and employ self-supervised learning and supervised learning in order to gain an understanding of optimal sensor settings. The results show that only the EOG sensor performs roughly as well as the best performing combination of other sensors. This gives an insight into selecting the appropriate sensors for fine-grained reading detection, enabling cost-effective computation.

The performance of conventional direction of arrival (DOA) methods is susceptible to the uncertainty of acoustic velocity in the underwater environment. To solve this problem, an underwater DOA estimation method with L-shaped array for wide-band signals under unknown acoustic velocity is proposed in this paper. The proposed method refers to the idea of incoherent signal subspace method and Root-MUSIC to obtain two sets of average roots corresponding to the subarray of the L-shaped array. And the geometric relationship between two vertical linear arrays is employed to derive the expression of DOA estimation with respect to the two average roots. The acoustic velocity variable in the DOA estimation expression can be eliminated in the proposed method. The simulation results demonstrate that the proposed method is more accurate and robust than other methods in an unknown acoustic velocity environment.

In this paper, we address the problem of finding a representation of a subtree distance, which is an extension of a tree metric. We show that a minimal representation is uniquely determined by a given subtree distance, and give an O(n2) time algorithm that finds such a representation, where n is the size of the ground set. Since a lower bound of the problem is Ω(n2), our algorithm achieves the optimal time complexity.

This paper investigates the joint problem of user association and spectrum allocation in satellite-terrestrial integrated networks (STINs), where a low earth orbit (LEO) satellite access network cooperating with terrestrial networks constitutes a heterogeneous network, which is beneficial in terms of both providing seamless coverage as well as improving the backhaul capacity for the dense network scenario. However, the orbital movement of satellites results in the dynamic change of accessible satellites and the backhaul capacities. Moreover, spectrum sharing may be faced with severe co-channel interferences (CCIs) caused by overlapping coverage of multiple access points (APs). This paper aims to maximize the total sum rate considering the influences of the dynamic feature of STIN, backhaul capacity limitation and interference management. The optimization problem is then decomposed into two subproblems: resource allocation for terrestrial communications and satellite communications, which are both solved by matching algorithms. Finally, simulation results show the effectiveness of our proposed scheme in terms of STIN's sum rate and spectrum efficiency.

This letter studies a biobjective optimization problem in binary associative memories characterized by ternary connection parameters. First, we introduce a condition of parameters that guarantees storage of any desired memories and suppression of oscillatory behavior. Second, we define a biobjective problem based on two objectives that evaluate uniform stability of desired memories and sparsity of connection parameters. Performing precise numerical analysis for typical examples, we have clarified existence of a trade-off between the two objectives.

The altered fingerprints help criminals escape from police and cause great harm to the society. In this letter, an altered fingerprint detection method is proposed. The method is constructed by two deep convolutional neural networks to train the time-domain and frequency-domain features. A spectral attention module is added to connect two networks. After the extraction network, a feature fusion module is then used to exploit relationship of two network features. We make ablation experiments and add the module proposed in some popular architectures. Results show the proposed method can improve the performance of altered fingerprint detection compared with the recent neural networks.

To accommodate an increasing amount of traffic efficiently, elastic optical networks (EON) that can use optical spectrum resources flexibly have been studied. We implement multi-path routing in case we cannot allocate the spectrum with single-path routing. However, multi-path routing requires more guard bands to avoid interference between two adjacent optical paths when compared with single-path routing in EON. A multi-path routing algorithm with traffic grooming technology has been proposed. The researchers assumed that a uniform modulation level was adopted, and so they did not consider the impact of path length on the resources needed. In this paper, we propose a multi-path routing method with traffic grooming considering path lengths. Our proposed method establishes an optical multi-path considering path length, fiber utilization, and the use of traffic grooming. Simulations show we can decrease the call-blocking probability by approximately 24.8% in NSFNET. We also demonstrate the effectiveness of traffic grooming and the improvement in the utilization ratio of optical spectrum resources.