Rapid Algorithm Development Research Articles

Multi‐scale seismic reliability assessment of networks by centrality‐based selective recursive decomposition algorithm

AbstractAs lifeline networks such as transportation or electricity networks in modern societies are intricately interlocked, a small number of components damaged by natural or man‐made disasters can have a great impact on network performance. For this reason, it is essential to assure the capability of rapid assessment of network reliability to make prompt follow‐up measures. Despite the rapid development of various algorithms and computing power, the capability is still limited due to computational cost for analyzing the connectivity of a single origin and destination (O/D) node pair in large‐scale networks. Therefore, this paper introduces a new algorithm utilizing network centrality, termed “centrality‐based selective recursive decomposition algorithm” (CS‐RDA). By preferentially decomposing the node which is most likely to belong to the min‐cut identified based on the betweenness centrality, the convergence of the bounds on the O/D connectivity can be expedited significantly. This paper also introduces a new multi‐scale analysis approach termed “edge‐betweenness algorithm.” The algorithm groups components such that its modularity is maximized, by sequentially removing edges that have the highest level of betweenness centrality. As a result, the reliability of large‐scale networks can be accurately evaluated in a short time owing to the reduced complexity of the simplified network. The proposed methods are successfully demonstrated by a hypothetical network example, the highway bridge networks in San Jose and San Diego in California, USA.

Image‐based high‐throughput phenotyping for the estimation of persistence of perennial ryegrass (Lolium perenne L.)—A review

AbstractPerennial ryegrass (Lolium perenne L.) is considered the most important pasture species in temperate agriculture, with over six million hectares of sown area in Australia alone. However, perennial ryegrass has poor persistence in some environments because of low tolerance to a range of both abiotic and biotic stresses. To breed perennial ryegrass, cultivars with greater persistence and productivity may require evaluation of genotypes over a number of years. Persistence assessment in pasture breeding depends on manual ground cover estimation or counting the number of surviving plants or tillers in a known area. These methods are subjective and labour intensive, which may limit data collection in large‐scale breeding programs. With the rapid development of sensors and image processing algorithms, image‐based high‐throughput phenotyping (HTP) is becoming commonplace in the breeding of major food crops. Image‐based HTP approaches consist of the deployment of a wide range of sensors on ground‐based or airborne platforms and data analysed through image processing pipelines. Image‐based HTP shows high potential for use in pasture phenotyping in breeding programs and may be able to reduce timeframes for releasing new cultivars. Moreover, existing image‐based HTP approaches could be further developed to include precise tools for phenotyping pasture persistence traits such as pasture senescence, botanical composition, pathogen and pest resistance. In this study, we reviewed existing image‐based HTP approaches in precision agriculture and discussed their feasibility for perennial ryegrass persistence estimation in pasture breeding. Although the paper focuses on application in perennial ryegrass, the principles equally apply to other perennial forage species.

A Compact Adaptive Particle Swarm Optimization Algorithm in the Application of the Mobile Sensor Localization

The mobile sensor network can sense and collect the data information of the monitored object in real time in the monitoring area. However, the collected information is meaningful only if the location of the node is known. This paper mainly optimizes the Monte Carlo Localization (MCL) in mobile sensor positioning technology. In recent years, the rapid development of heuristic algorithms has provided solutions to many complex problems. This paper combines the compact strategy into the adaptive particle swarm algorithm and proposes a compact adaptive particle swarm algorithm (cAPSO). The compact strategy replaces the specific position of each particle by the distribution probability of the particle swarm, which greatly reduces the memory usage. The performance of cAPSO is tested on 28 test functions of CEC2013, and compared with some existing heuristic algorithms, it proves that cAPSO has a better performance. At the same time, cAPSO is applied to MCL technology to improve the accuracy of node localization, and compared with other heuristic algorithms in the accuracy of MCL, the results show that cAPSO has a better performance.

Athlete Behavior Recognition Technology Based on Siamese-RPN Tracker Model.

With the rapid development of deep learning algorithms, it is gradually applied in UAV (Unmanned Aerial Vehicle) driving, visual recognition, target tracking, behavior recognition, and other fields. In the field of sports, many scientists put forward the research of target tracking and recognition technology based on deep learning algorithms for athletes' trajectory and behavior capture. Based on the target tracking algorithm, a regional proposal network RPN algorithm combined with the twin regional proposal network Siamese algorithm is proposed to study the tracking and recognition technology of athletes' behavior. Then, the adaptive updating network is used to track the behavior target of athletes, and the simulation model of behavior recognition is established. This algorithm is different from the traditional twin network algorithm. It can accurately take the athlete's behavior as the target candidate box in model training and reduce the interference of environment and other factors on model recognition. The results show that the Siamese-RPN algorithm can reduce the interference from the background and environment when tracking the athletes' target behavior trajectory. This algorithm can improve the training behavior recognition model, ignore the background interference elements of the behavior image, and improve the accuracy and overall performance of the model. Compared with the traditional twin network method for sports behavior recognition, the Siamese-RPN algorithm studied in this paper can perform offline operations and distinguish the interference factors of athletes' background environment. It can quickly capture the characteristic points of athletes' behavior as the data input of the tracking model, so it has excellent popularization and application value.

Metagenomic Assembly: Reconstructing Genomes from Metagenomes.

Assembly of metagenomic sequence data into microbial genomes is of critical importance for disentangling community complexity and unraveling the functional capacity of microorganisms. The rapid development of sequencing technology and novel assembly algorithms have made it possible to reliably reconstruct hundreds to thousands of microbial genomes from raw sequencing reads through metagenomic assembly. In this chapter, we introduce a routinely used metagenomic assembly workflow including read quality filtering, assembly, contig/scaffold binning, and postassembly check for genome completeness and contamination. We also describe a case study to reconstruct near-complete microbial genomes from metagenomes using our workflow.

An Integrated Method for Tracking and Monitoring Stomata Dynamics from Microscope Videos.

Patchy stomata are a common and characteristic phenomenon in plants. Understanding and studying the regulation mechanism of patchy stomata are of great significance to further supplement and improve the stomatal theory. Currently, the common methods for stomatal behavior observation are based on static images, which makes it difficult to reflect dynamic changes of stomata. With the rapid development of portable microscopes and computer vision algorithms, it brings new chances for stomatal movement observation. In this study, a stomatal behavior observation system (SBOS) was proposed for real-time observation and automatic analysis of each single stoma in wheat leaf using object tracking and semantic segmentation methods. The SBOS includes two modules: the real-time observation module and the automatic analysis module. The real-time observation module can shoot videos of stomatal dynamic changes. In the automatic analysis module, object tracking locates every single stoma accurately to obtain stomatal pictures arranged in time-series; semantic segmentation can precisely quantify the stomatal opening area (SOA), with a mean pixel accuracy (MPA) of 0.8305 and a mean intersection over union (MIoU) of 0.5590 in the testing set. Moreover, we designed a graphical user interface (GUI) so that researchers could use this automatic analysis module smoothly. To verify the performance of the SBOS, the dynamic changes of stomata were observed and analyzed under chilling. Finally, we analyzed the correlation between gas exchange and SOA under drought stress, and the correlation coefficients between mean SOA and net photosynthetic rate (Pn), intercellular CO2 concentration (Ci), stomatal conductance (Gs), and transpiration rate (Tr) are 0.93, 0.96, 0.96, and 0.97.

Smartphone-Based Indoor Tracking in Multiple-Floor Scenarios

With the rapid development of location-based services (LBSs), efficient and mobile-friendly localization algorithms should be designed for users to deliver a reliable LBS. In this paper, we present an algorithm with the corresponding smartphone app that enables users to calculate their locations based on representative infrastructures, such as nearby Wi-Fi access points and Bluetooth low energy (BLE) beacons subject to low-cost, rapid system deployment, and competitive location accuracy. Working under indoor multiple-floor scenarios, our app has three prominent features for estimating user locations. First, we establish a feature identifier to detect the current floor and the feasible area in which the user may walk. Second, owing to the structures of the indoor environment and the presence of different obstacles, the unpredictable variation of the received signal strength (RSS) in indoor environments is considered in the RSS-distance relationship to provide accurate location estimates. Third, with the prevalence of smartphones, we extract smartphone-inertial measurement units to learn users’ behavior preferences, while collecting reference signals (e.g., Wi-Fi/BLE readings) along the pathway and input to the tracking algorithm. Then, the user’s current location is displayed on the app. With this solution, we can provide an accurate location estimate with relatively low computational complexity regarding mobile device capability, while reducing labor costs from traditional fingerprint deployments. Finally, we test our tracking app in real-time multiple floor scenarios and evaluate the collected tracking data. Experimental results show that our proposed scheme achieves an average localization accuracy of more than 80% within a 2-m error bound in multiple-floor scenarios, while all areas (i.e., corridors, rooms, and stairs) were successfully identified.

Research on detection algorithm of lithium battery surface defects based on embedded machine vision

In the production process of lithium battery, the quality inspection requirements of lithium battery are very high. At present, most of the work is done manually. Aiming at the problem of large manual inspection workload and large error, the robot visual inspection technology is applied to the production of lithium battery. In recent years, with the rapid development and progress of science and technology, the rapid development of visual detection hardware and algorithms, making it possible to screen defective products through visual detection algorithms. This paper takes lithium battery as the research object, and studies its vision detection algorithm. As a common commodity, the quality of lithium battery is the key for users to choose. With the increasing requirements of users for battery quality, how to produce high-quality battery is the key problem to be solved by manufacturers. However, at present, the defects of battery surface are mostly carried out manually. There are low efficiency and low detection rate in the process of manual detection. In this paper, the visual detection algorithm is studied to detect the defects such as pits, rust marks and broken skin on the surface of lithium battery, specifically to design the imaging experimental platform of lithium battery; use different lighting schemes to design different battery positioning and extraction algorithms; use Hough detection method to locate the battery surface, and design the battery defect algorithm for this, and compare the algorithm through experiments.

Commentary: Targeted screening for effective detection of vision threatening diabetic retinopathy.

Various techniques of retinal examination and different types of screening programs have been devised over the last decade for detection of diabetic retinopathy (DR).[1] The success of any such program is based on how effective it has been in detecting and treating vision threatening diabetic retinopathy (VTDR) and thereby reducing the blinding complication of DR. In this context, the article by Sagnik et al.[2] is very relevant and informative and should help us to devise new strategies for effective screening of DR. The authors performed a subgroup analysis of the data on all enrolled participants above the age of 40 years from four major population-based studies from India.[3456] They concluded that persons with diabetes between the age of 50–69 years and those with systolic blood pressure ≥140 mm Hg have a higher risk of VTDR. Therefore, targeted screening of this vulnerable population alone can result in identifying 93% of all patients with VTDR among persons with diabetic retinopathy. However, the authors did not look into both micro and macroalbuminuria, two important risk factors for presence of retinopathy and severity of retinopathy.[7] How do we change our strategy and approach to screening programs based on these observations? The diabetic retinopathy screening guidelines in India should emphasize and give importance to identification of VTDR and to target the vulnerable population. The present guidelines fail to address this issue.[8] Similarly, the National Task Force on Diabetic Retinopathy should create awareness among primary care physicians and endocrinologists on the importance of identifying the vulnerable population and getting them examined for treatable diabetic retinopathy. All major departments of internal medicine in district hospitals, medical colleges, and large multispecialty hospitals including dedicated diabetic centers should have facilities for remote screening of diabetic retinopathy either through telemedicine or through the cloud format. Remote screening refers to the transmission of images to a central reading center using a network. A recent report from Japan demonstrates how successful remote screening is in detecting treatable diabetic retinopathy using nonmydriatic wide-field imaging.[9] Finally, the rapid development of artificial intelligence and deep learning algorithms for detection of diabetic retinopathy may make it easier to identify patients at a higher risk of losing vision.[10]’

Digital nanophotonics: the highway to the integration of subwavelength-scale photonics

Abstract Nanophotonic devices with high densities are extremely attractive because they can potentially merge photonics and electronics at the nanoscale. However, traditional integrated photonic circuits are designed primarily by manually selecting parameters or employing semi-analytical models. Limited by the small parameter search space, the designed nanophotonic devices generally have a single function, and the footprints reach hundreds of microns. Recently, novel ultra-compact nanophotonic devices with digital structures were proposed. By applying inverse design algorithms, which can search the full parameter space, the proposed devices show extremely compact footprints of a few microns. The results from many groups imply that digital nanophotonics can achieve not only ultra-compact single-function devices but also miniaturized multi-function devices and complex functions such as artificial intelligence operations at the nanoscale. Furthermore, to balance the performance and fabrication tolerances of such devices, researchers have developed various solutions, such as adding regularization constraints to digital structures. We believe that with the rapid development of inverse design algorithms and continuous improvements to the nanofabrication process, digital nanophotonics will play a key role in promoting the performance of nanophotonic integration. In this review, we uncover the exciting developments and challenges in this field, analyse and explore potential solutions to these challenges and provide comments on future directions in this field.

Development platform for artificial pancreas algorithms

Assessing algorithms of artificial pancreas systems is critical in developing automated and fault-tolerant solutions that work outside clinical settings. The development and evaluation of algorithms can be facilitated with a platform that conducts virtual clinical trials. We present in this paper a clinically validated cloud-based distributed platform that supports the development and comprehensive testing of single and dual-hormone algorithms for type 1 diabetes mellitus (T1DM). The platform is built on principles of object-oriented design and runs user algorithms in real-time virtual clinical trials utilizing a multi-threaded environment enabled by concurrent execution over a cloud infrastructure. The platform architecture isolates user algorithms located on personal machines from proprietary patient data running on the cloud. Users import a plugin into their algorithms (Matlab, Python, or Java) to connect to the platform. Once connected, users interact with a graphical interface to design experimental protocols for their trials. Protocols include trial duration in days, mealtimes and amounts, variability in mealtimes and amounts, carbohydrate counting errors, snacks, and onboard insulin levels. The platform facilitates development by solving the ODE model in the cloud on large CPU-optimized machines, providing a 62% improvement in memory, speed and CPU utilization. Users can easily debug & modify code, test multiple strategies, and generate detailed clinical performance reports. We validated and integrated into the platform a glucoregulatory system of ordinary differential equations (ODEs) parameterized with clinical data to mimic the inter and intra-day variability of glucose responses of 15 T1DM patients. The platform utilizes the validated patient model to conduct virtual clinical trials for the rapid development and testing of closed-loop algorithms for T1DM.

Immersed boundary method for multiphase transport phenomena

Abstract Multiphase flows with momentum, heat, and mass transfer exist widely in a variety of industrial applications. With the rapid development of numerical algorithms and computer capacity, advanced numerical simulation has become a promising tool in investigating multiphase transport problems. Immersed boundary (IB) method has recently emerged as such a popular interface capturing method for efficient simulations of multiphase flows, and significant achievements have been obtained. In this review, we attempt to give an overview of recent progresses on IB method for multiphase transport phenomena. Firstly, the governing equations, the basic ideas, and different boundary conditions for the IB methods are introduced. This is followed by numerical strategies, from which the IB methods are classified into two types, namely the artificial boundary method and the authentic boundary method. Discussions on the implementation of various boundary conditions at the interphase surface with momentum, heat, and mass transfer for different IB methods are then presented, together with a summary. Then, the state-of-the-art applications of IB methods to multiphase flows, including the isothermal flows, the heat transfer flows, and the mass transfer problems are outlined, with particular emphasis on the latter two topics. Finally, the conclusions and future challenges are identified.

Lidar sheds new light on plant phenomics for plant breeding and management: Recent advances and future prospects

Plant phenomics is a new avenue for linking plant genomics and environmental studies, thereby improving plant breeding and management. Remote sensing techniques have improved high-throughput plant phenotyping. However, the accuracy, efficiency, and applicability of three-dimensional (3D) phenotyping are still challenging, especially in field environments. Light detection and ranging (lidar) provides a powerful new tool for 3D phenotyping with the rapid development of facilities and algorithms. Numerous efforts have been devoted to studying static and dynamic changes of structural and functional phenotypes using lidar in agriculture. These progresses also improve 3D plant modeling across different spatial–temporal scales and disciplines, providing easier and less expensive association with genes and analysis of environmental practices and affords new insights into breeding and management. Beyond agriculture phenotyping, lidar shows great potential in forestry, horticultural, and grass phenotyping. Although lidar has resulted in remarkable improvements in plant phenotyping and modeling, the synthetization of lidar-based phenotyping for breeding and management has not been fully explored. We identify three main challenges in lidar-based phenotyping development: 1) developing low cost, high spatial–temporal, and hyperspectral lidar facilities, 2) moving into multi-dimensional phenotyping with an endeavor to generate new algorithms and models, and 3) embracing open source and big data.

Review of the Research Landscape of Multi-Criteria Evaluation and Benchmarking Processes for Many-Objective Optimization Methods: Coherent Taxonomy, Challenges and Recommended Solution

Evaluation and benchmarking of many-objective optimization (MaOO) methods are complicated. The rapid development of new optimization algorithms for solving problems with many objectives has increased the necessity of developing performance indicators or metrics for evaluating the performance quality and comparing the competing optimization algorithms fairly. Further investigations are required to highlight the limitations of how criteria/metrics are determined and the consistency of the procedures with the evaluation and benchmarking processes of MaOO. A review is conducted in this study to map the research landscape of multi-criteria evaluation and benchmarking processes for MaOO into a coherent taxonomy. Then contentious and challenging issues related to evaluation are highlighted, and the performance of optimization algorithms for MaOO is benchmarked. The methodological aspects of the evaluation and selection of MaOO algorithms are presented as the recommended solution on the basis of four distinct and successive phases. First, in the determination phase, the evaluation criteria of MaOO are collected, classified and grouped for testing experts’ consensus on the most suitable criteria. Second, the identification phase involves the process of establishing a decision matrix via a crossover of the ‘evaluation criteria’ and MaOO’, and the level of importance of each selective criteria and sub-criteria from phase one is computed to identify its weight value by using the best–worst method (BWM). Third, the development phase involves the creation of a decision matrix for MaOO selection on the basis of the integrated BWM and VIKOR method. Last, the validation phase involves the validation of the proposed solution.

Research on Optimization of Improved Gray Wolf Optimization-Extreme Learning Machine Algorithm in Vehicle Route Planning

With the rapid development of intelligent transportation, intelligent algorithms and path planning have become effective methods to relieve traffic pressure. Intelligent algorithm can realize the priority selection mode in realizing traffic optimization efficiency. However, there is local optimization in intelligence and it is difficult to realize global optimization. In this paper, the antilearning model is used to solve the problem that the gray wolf algorithm falls into local optimization. The positions of different wolves are updated. When falling into local optimization, the current position is optimized to realize global optimization. Extreme Learning Machine (ELM) algorithm model is introduced to accelerate Improved Gray Wolf Optimization (IGWO) optimization and improve convergence speed. Finally, the experiment proves that IGWO-ELM algorithm is compared in path planning, and the algorithm has an ideal effect and high efficiency.

SOS 2.0: an evolutionary approach for SOS algorithm

With the shortcomings on the solution given for most-recent optimization problems, decision-makers from different fields yearn the existence of tenacious breakthrough. In fact, they all shared the same obligation to optimize work efficiency, whether to minimize cost, consumption or to maximize the profit acquirement. Metaheuristic search is the more-advanced method proven to be useful for difficult optimization tasks. Moreover, development records also signalized rapid development of these algorithms, contributing several notable and powerful optimization algorithms. Among them, Symbiotic Organisms Search (SOS) received noticeable attention due to its simplicity and also its parameter-less nature. Nonetheless, several considerable issues are still challenging for further development. For instance, local optima and premature convergence issues found from any improper and inefficiency computational procedure on higher dimensional problems. Also, exploitation and exploration trade-off is another essential issue involving stability for optimal performance. In that case, this work proposed a new evolutionary approach named SOS 2.0. There are two distinct features associated with the evolution: Self-Parameter-Updating (SPU) technique and chaotic maps sequencing. Both features are integrated for a better balance of exploration and exploitation in which SPU focuses on exploration and chaotic map focuses on exploitation instead. This work also applied benchmarks function tests and engineering design optimization problem in advance for validation purpose of the performance. The experimental results showed that SOS 2.0 delivers not only better performance from its predecessor and also several recent SOS modifications which can be concluded as one successive approach for better SOS algorithm, but also enhances the computation efficiency and capability of searching optimal solution.

Towards machine learning analytics for jet substructure

The past few years have seen a rapid development of machine-learning algorithms. While surely augmenting performance, these complex tools are often treated as black-boxes and may impair our understanding of the physical processes under study. The aim of this paper is to move a first step into the direction of applying expert-knowledge in particle physics to calculate the optimal decision function and test whether it is achieved by standard training, thus making the aforementioned black-box more transparent. In particular, we consider the binary classification problem of discriminating quark-initiated jets from gluon-initiated ones. We construct a new version of the widely used N-subjettiness, which features a simpler theoretical behaviour than the original one, while maintaining, if not exceeding, the discrimination power. We input these new observables to the simplest possible neural network, i.e. the one made by a single neuron, or perceptron, and we analytically study the network behaviour at leading logarithmic accuracy. We are able to determine under which circumstances the perceptron achieves optimal performance. We also compare our analytic findings to an actual implementation of a perceptron and to a more realistic neural network and find very good agreement.

Hybrid wind power forecasting based on extreme learning machine and improved TLBO algorithm

Wind power forecasting (WPF) plays an important role in the planning, efficient operation, and security maintenance of power systems. A large number of hybrid models have been applied to WPF in the past two decades. Due to the rapid development of swarm intelligence algorithms, there is great potential for forecasting performance improvements by combining them with basic data-driven models for parameter optimization. In this study, a hybrid WPF method is proposed, which combines an extreme learning machine (ELM) and improved teaching-learning-based optimization (iTLBO), and incorporates a recursive feature elimination (RFE) method for feature selection. For WPF, appropriate feature combination is recognized from original input data using the RFE method, which helps facilitate understanding of the data pattern and defy the curse of dimensionality. To enhance the convergence speed and learning ability of the basic TLBO, four improvements are performed, and the obtained iTLBO algorithm is applied to optimize the parameters of the ELM model. Case study data came from a wind farm in Yunnan, China. The ERMSE, EMAE, and MAPE values of the proposed hybrid method are all lower than those of the comparison methods. The results demonstrate the superior forecasting performance that makes the hybrid method more applicable in real WPF applications.

Continuous Model Adaptation Using Online Meta-Learning for Smart Grid Application.

The rapid development of deep learning algorithms provides us an opportunity to better understand the complexity in engineering systems, such as the smart grid. Most of the existing data-driven predictive models are trained using historical data and fixed during the execution stage, which cannot adapt well to real-time data. In this research, we propose a novel online meta-learning (OML) algorithm to continuously adapt pretrained base-learner through efficiently digesting real-time data to adaptively control the base-learner parameters using meta-optimizer. The simulation results show that: 1) both ML and OML can perform significantly better than online base learning. 2) OML can perform better than ML and online base learning when the training data are limited, or the training and real-time data have very different time-variant patterns.

A Two-layer Mechanism Identification Method for VIN Adversarial Examples

With the rapid development of machine learning algorithms, the security problems are gradually emerging. Most existing machine learning algorithms may be attacked by adversarial examples. Especially in the domain of path planning, the adversarial maps may result in multiple harmful consequences on the paths predicted by Deep Reinforcement Learning (DRL) algorithms. However, there is no suitable approach to automatically identify them. To our knowledge, all previous work used manual observation method to identify the adversarial maps, which is time-consuming. Therefore, this paper explores a two-layer mechanism method to automatically identify the adversarial examples in Value Iteration Networks (VIN). We define the four categories of attack results and identify them by combining path feature comparison and path image classification. Experiments show that this method can achieve an effective identification on VIN adversarial examples.