Published Online

Display Method:
Computational Intelligence in Remote Sensing Image Registration: A survey
Yue Wu, Jun-Wei Liu, Chen-Zhuo Zhu, Zhuang-Fei Bai, Qi-Guang Miao, Wen-Ping Ma, Mao-Guo Gong
doi: 10.1007/s11633-020-1248-x
In recent years, computational intelligence has been widely used in many fields and achieved remarkable performance. Evolutionary computing and deep learning are important branches of computational intelligence. Many methods based on evolutionary computation and deep learning have achieved good performance in remote sensing image registration. This paper introduces the application of computational intelligence in remote sensing image registration from the two directions of evolutionary computing and deep learning. In the part of remote sensing image registration based on evolutionary calculation, the principles of evolutionary algorithms and swarm intelligence algorithms are elaborated and their application in remote sensing image registration is discussed. The application of deep learning in remote sensing image registration is also discussed. At the same time, the development status and future of remote sensing image registration are summarized and their prospects are examined.
Research Article
A Performance Evaluation of Classic Convolutional Neural Networks for 2D and 3D Palmprint and Palm Vein Recognition
Wei Jia, Jian Gao, Wei Xia, Yang Zhao, Hai Min, Jing-Ting Lu
doi: 10.1007/s11633-020-1257-9
Palmprint recognition and palm vein recognition are two emerging biometrics technologies. In the past two decades, many traditional methods have been proposed for palmprint recognition and palm vein recognition, and have achieved impressive results. However, the research on deep learning-based palmprint recognition and palm vein recognition is still very preliminary. In this paper, in order to investigate the problem of deep learning based 2D and 3D palmprint recognition and palm vein recognition in-depth, we conduct performance evaluation of seventeen representative and classic convolutional neural networks (CNNs) on one 3D palmprint database, five 2D palmprint databases and two palm vein databases. A lot of experiments have been carried out in the conditions of different network structures, different learning rates, and different numbers of network layers. We have also conducted experiments on both separate data mode and mixed data mode. Experimental results show that these classic CNNs can achieve promising recognition results, and the recognition performance of recently proposed CNNs is better. Particularly, among classic CNNs, one of the recently proposed classic CNNs, i.e., EfficientNet achieves the best recognition accuracy. However, the recognition performance of classic CNNs is still slightly worse than that of some traditional recognition methods.
Camera-based Basketball Scoring Detection Using Convolutional Neural Network
Xu-Bo Fu, Shao-Long Yue, De-Yun Pan
doi: 10.1007/s11633-020-1259-7
Recently, deep learning methods have been applied in many real scenarios with the development of convolutional neural networks (CNNs). In this paper, we introduce a camera-based basketball scoring detection (BSD) method with CNN based object detection and frame difference-based motion detection. In the proposed BSD method, the videos of the basketball court are taken as inputs. Afterwards, the real-time object detection, i.e., you only look once (YOLO) model, is implemented to locate the position of the basketball hoop. Then, the motion detection based on frame difference is utilized to detect whether there is any object motion in the area of the hoop to determine the basketball scoring condition. The proposed BSD method runs in real-time with satisfactory basketball scoring detection accuracy. Our experiments on the collected real scenario basketball court videos show the accuracy of the proposed BSD method. Furthermore, several intelligent basketball analysis systems based on the proposed method have been installed at multiple basketball courts in Beijing, and they provide good performance.
Type Synthesis and Dynamics Performance Evaluation of a Class of 5-DOF Redundantly Actuated Parallel Mechanisms
Bing-Shan Jiang, Hai-Rong Fang, Hai-Qiang Zhang
doi: 10.1007/s11633-020-1255-y
This paper presents a five degree of freedom (5-DOF) redundantly actuated parallel mechanism (PM) for the parallel machining head of a machine tool. A 5-DOF single kinematic chain is evolved into a secondary kinematic chain based on Lie group theory and a configuration evolution method. The evolutional chain and four 6-DOF kinematic chain SPS (S represents spherical joint and P represents prismatic joint) or UPS (U represents universal joint) can be combined into four classes of 5-DOF redundantly actuated parallel mechanisms. That SPS-(2UPR)R (R represents revolute joint) redundantly actuated parallel mechanism is selected and is applied to the parallel machining head of the machine tool. All formulas of the 4SPS-(2UPR)R mechanism are deduced. The dynamic model of the mechanism is shown to be correct by Matlab and automatic dynamic analysis of mechanical systems (ADAMS) under no-load conditions. The dynamic performance evaluation indexes including energy transmission efficiency and acceleration performance evaluation are analyzed. The results show that the 4SPS-(2UPR)R mechanism can be applied to a parallel machining head and have good dynamic performance.
Computational Decision Support System for ADHD Identification
Senuri De Silva, Sanuwani Dayarathna, Gangani Ariyarathne, Dulani Meedeniya, Sampath Jayarathna, Anne M. P. Michalek
doi: 10.1007/s11633-020-1252-1
Attention deficit/hyperactivity disorder (ADHD) is a common disorder among children. ADHD often prevails into adulthood, unless proper treatments are facilitated to engage self-regulatory systems. Thus, there is a need for effective and reliable mechanisms for the early identification of ADHD. This paper presents a decision support system for the ADHD identification process. The proposed system uses both functional magnetic resonance imaging (fMRI) data and eye movement data. The classification processes contain enhanced pipelines, and consist of pre-processing, feature extraction, and feature selection mechanisms. fMRI data are processed by extracting seed-based correlation features in default mode network (DMN) and eye movement data using aggregated features of fixations and saccades. For the classification using eye movement data, an ensemble model is obtained with 81% overall accuracy. For the fMRI classification, a convolutional neural network (CNN) is used with 82% accuracy for the ADHD identification. Both ensemble models are proved for overfitting avoidance.
Knowing Your Dog Breed: Identifying a Dog Breed with Deep Learning
Punyanuch Borwarnginn, Worapan Kusakunniran, Sarattha Karnjanapreechakorn, Kittikhun Thongkanchorn
doi: 10.1007/s11633-020-1261-0
Dog breed identification is essential for many reasons, particularly for understanding individual breeds′ conditions, health concerns, interaction behavior, and natural instinct. This paper presents a solution for identifying dog breeds using their images of their faces. The proposed method applies a deep learning based approach in order to recognize their breeds. The method begins with a transfer learning by retraining existing pre-trained convolutional neural networks (CNNs) on the public dog breed dataset. Then, the image augmentation with various settings is also applied on the training dataset, in order to improve the classification performance. The proposed method is evaluated using three different CNNs with various augmentation settings and comprehensive experimental comparisons. The proposed model achieves a promising accuracy of 89.92% on the published dataset with 133 dog breeds.
Image Inpainting Based on Structural Tensor Edge Intensity Model
Jing Wang, Yan-Hong Zhou, Hai-Feng Sima, Zhan-Qiang Huo, Ai-Zhong Mi
doi: 10.1007/s11633-020-1256-x
In the exemplar-based image inpainting approach, there are usually two major problems: the unreasonable calculation of priority and only considering the color features in the patch lookup strategy. In this paper, we propose an image inpainting approach based on the structural tensor edge intensity model. First, we use the progressive scanning inpainting method to avoid the image filling order being affected by the priority function. Then, we use the edge intensity model to build the patches similarity function for correctly identifying the local image structure. Finally, the balance operator is used to restrict the excessive propagation of structural information to ensure the correct structural reconstruction. The experimental results show that the our approach is comparable and even superior to some state-of-the-art inpainting algorithms.
Event-triggered Control of Positive Switched Systems with Actuator Saturation and Time-delay
Jun-Feng Zhang, Lai-You Liu, Shi-Zhou Fu, Shuo Li
doi: 10.1007/s11633-020-1245-0
This paper investigates the event-triggered control of positive switched systems with randomly occurring actuator saturation and time-delay, where the actuator saturation and time-delay obey different Bernoulli distributions. First, an event-triggering condition is constructed based on a 1-norm inequality. Under the presented event-triggering scheme, an interval estimation method is utilized to deal with the error term of the systems. Using a co-positive Lyapunov functional, the event-triggered controller and the cone attraction domain gain matrices are designed via matrix decomposition techniques. The positivity and stability of the resulting closed-loop systems are reached by guaranteeing the positivity of the lower bound of the systems and the stability of the upper bound of the systems, respectively. The proposed approach is developed for interval and polytopic uncertain systems, respectively. Finally, two examples are provided to illustrate the effectiveness of the theoretical findings.
Rail Detection Based on LSD and the Least Square Curve Fitting
Yun-Shui Zheng, Yan-Wei Jin, Yu Dong
doi: 10.1007/s11633-020-1241-4
It is necessary to rely on the rail gauge to determine whether the object beside the track will affect train operation safety or not. A convenient and fast method based on line segment detector (LSD) and the least square curve fitting to identify the rail in the image is proposed in this paper. The image in front of the train can be obtained through the camera on-board. After preprocessing, it will be divided equally along the longitudinal axis. Utilizing the characteristics of the LSD algorithm, the edges are approximated into multiple line segments. After screening the terminals of the line segments, it can generate the mathematical model of the rail in the image based on the least square. Experiments show that the algorithm in this paper can fit the rail curve accurately and has good applicability and robustness.
Design of an Executable ANFIS-based Control System to Improve the Attitude and Altitude Performances of a Quadcopter Drone
Mohammad Al-Fetyani, Mohammad Hayajneh, Adham Alsharkawi
doi: 10.1007/s11633-020-1251-2
Nowadays, quadcopters are presented in many life applications which require the performance of automatic takeoff, trajectory tracking, and automatic landing. Thus, researchers are aiming to enhance the performance of these vehicles through low-cost sensing solutions and the design of executable and robust control techniques. Due to high nonlinearities, strong couplings and under-actuation, the control design process of a quadcopter is a rather challenging task. Therefore, the main objective of this work is demonstrated through two main aspects. The first is the design of an adaptive neuro-fuzzy inference system (ANFIS) controller to develop the attitude and altitude of a quadcopter. The second is to create a systematic framework for implementing flight controllers in embedded systems. A suitable model of the quadcopter is also developed by taking into account aerodynamics effects. To show the effectiveness of the ANFIS approach, the performance of a well-trained ANFIS controller is compared to a classical proportional-derivative (PD) controller and a properly tuned fuzzy logic controller. The controllers are compared and tested under several different flight conditions including the capability to reject external disturbances. In the first stage, performance evaluation takes place in a nonlinear simulation environment. Then, the ANFIS-based controllers alongside attitude and position estimators, and precision landing algorithms are implemented for executions in a real-time autopilot. In precision landing systems, an IR-camera is used to detect an IR-beacon on the ground for precise positioning. Several flight tests of a quadcopter are conducted for results validation. Both simulations and experiments demonstrated superior results for quadcopter stability in different flight scenarios.
Hybrid Approach to Document Anomaly Detection: An Application to Facilitate RPA in Title Insurance
Abhijit Guha, Debabrata Samanta
doi: 10.1007/s11633-020-1247-y
Anomaly detection (AD) is an important aspect of various domains and title insurance (TI) is no exception. Robotic process automation (RPA) is taking over manual tasks in TI business processes, but it has its limitations without the support of artificial intelligence (AI) and machine learning (ML). With increasing data dimensionality and in composite population scenarios, the complexity of detecting anomalies increases and AD in automated document management systems (ADMS) is the least explored domain. Deep learning, being the fastest maturing technology can be combined along with traditional anomaly detectors to facilitate and improve the RPAs in TI. We present a hybrid model for AD, using autoencoders (AE) and a one-class support vector machine (OSVM). In the present study, OSVM receives input features representing real-time documents from the TI business, orchestrated and with dimensions reduced by AE. The results obtained from multiple experiments are comparable with traditional methods and within a business acceptable range, regarding accuracy and performance.
Saliency Detection via Manifold Ranking Based on Robust Foreground
Wei-Ping Ma, Wen-Xin Li, Jin-Chuan Sun, Peng-Xia Cao
doi: 10.1007/s11633-020-1246-z
The graph-based manifold ranking saliency detection only relies on the boundary background to extract foreground seeds, resulting in a poor saliency detection result, so a method that obtains robust foreground for manifold ranking is proposed in this paper. First, boundary connectivity is used to select the boundary background for manifold ranking to get a preliminary saliency map, and a foreground region is acquired by a binary segmentation of the map. Second, the feature points of the original image and the filtered image are obtained by using color boosting Harris corners to generate two different convex hulls. Calculating the intersection of these two convex hulls, a final convex hull is found. Finally, the foreground region and the final convex hull are combined to extract robust foreground seeds for manifold ranking and getting final saliency map. Experimental results on two public image datasets show that the proposed method gains improved performance compared with some other classic methods in three evaluation indicators: precision-recall curve, F-measure and mean absolute error.
A Position Synchronization Controller for Co-ordinated Links (COOL) Dual Robot Arm Based on Integral Sliding Mode: Design and Experimental Validation
Sumi Phukan, Chitralekha Mahanta
doi: 10.1007/s11633-020-1242-3
In this study, a simple position synchronization control algorithm based on an integral sliding mode is developed for dual-arm robotic manipulator systems. A first-order sliding surface is designed using cross-coupling error in order to ensure position synchronization of dual-arm manipulators. The design objective of the proposed controller is to ensure stability as well as to synchronize the movement of both arms while maintaining the trajectory as desired. The integral sliding mode eliminates the reaching phase and guarantees robustness throughout the whole operating period. Additionally, a low pass filter is used to smoothen the discontinuous element and minimize unwanted chattering. Lyapunov stability theory is utilized to prove the asymptotic stability of the controlled system. Simulation studies are performed to validate the proposed controller′s effectiveness. Also, to investigate the possibility of realizing the proposed dynamic control method in practical applications, experiments are conducted on a 14DoF coordinated links (COOL) dual-arm robotic manipulator system. Experimental evidence indicates adequate efficiency in trajectory tracking and guarantees robustness in the presence of parametric uncertainty and external disturbance.