scholarly journals Design of Decision Tree Structure with Improved BPNN Nodes for High-Accuracy Locomotion Mode Recognition Using a Single IMU

Sensors ◽  
2021 ◽  
Vol 21 (2) ◽  
pp. 526
Author(s):  
Yang Han ◽  
Chunbao Liu ◽  
Lingyun Yan ◽  
Lei Ren
Keyword(s):  
Decision Tree ◽  
Time Window ◽  
Threshold Value ◽  
Recognition System ◽  
Tree Structure ◽  
Classification Model ◽  
Measurement Unit ◽  
Local Optima ◽  
Mode Recognition ◽  
Locomotion Mode

Smart wearable robotic system, such as exoskeleton assist device and powered lower limb prostheses can rapidly and accurately realize man–machine interaction through locomotion mode recognition system. However, previous locomotion mode recognition studies usually adopted more sensors for higher accuracy and effective intelligent algorithms to recognize multiple locomotion modes simultaneously. To reduce the burden of sensors on users and recognize more locomotion modes, we design a novel decision tree structure (DTS) based on using an improved backpropagation neural network (IBPNN) as judgment nodes named IBPNN-DTS, after analyzing the experimental locomotion mode data using the original values with a 200-ms time window for a single inertial measurement unit to hierarchically identify nine common locomotion modes (level walking at three kinds of speeds, ramp ascent/descent, stair ascent/descent, Sit, and Stand). In addition, we reduce the number of parameters in the IBPNN for structure optimization and adopted the artificial bee colony (ABC) algorithm to perform global search for initial weight and threshold value to eliminate system uncertainty because randomly generated initial values tend to result in a failure to converge or falling into local optima. Experimental results demonstrate that recognition accuracy of the IBPNN-DTS with ABC optimization (ABC-IBPNN-DTS) was up to 96.71% (97.29% for the IBPNN-DTS). Compared to IBPNN-DTS without optimization, the number of parameters in ABC-IBPNN-DTS shrank by 66% with only a 0.58% reduction in accuracy while the classification model kept high robustness.

scholarly journals Computer Vision and Deep Learning for Environment-Adaptive Control of Robotic Lower-Limb Exoskeletons

2021 ◽  
Author(s):  
Brokoslaw Laschowski ◽  
William McNally ◽  
Alexander Wong ◽  
John McPhee
Keyword(s):  
Computer Vision ◽  
Deep Learning ◽  
Image Classification ◽  
Recognition System ◽  
Camera System ◽  
Human Control ◽  
Mode Recognition ◽  
Environment Recognition ◽  
Locomotion Mode ◽  
High Level

Robotic exoskeletons require human control and decision making to switch between different locomotion modes, which can be inconvenient and cognitively demanding. To support the development of automated locomotion mode recognition systems (i.e., high-level controllers), we designed an environment recognition system using computer vision and deep learning. We collected over 5.6 million images of indoor and outdoor real-world walking environments using a wearable camera system, of which ~923,000 images were annotated using a 12-class hierarchical labelling architecture (called the ExoNet database). We then trained and tested the EfficientNetB0 convolutional neural network, designed for efficiency using neural architecture search, to predict the different walking environments. Our environment recognition system achieved ~73% image classification accuracy. While these preliminary results benchmark EfficientNetB0 on the ExoNet database, further research is needed to compare different image classification algorithms to develop an accurate and real-time environment-adaptive locomotion mode recognition system for robotic exoskeleton control.

A Beam Search Based Decision Tree Induction Algorithm

2012 ◽  
pp. 357-370 ◽  
Author(s):  
Márcio Porto Basgalupp ◽  
Rodrigo Coelho Barros ◽  
André C. P. L. F. de Carvalho ◽  
Alex A. Freitas
Keyword(s):  
Evolutionary Algorithms ◽  
Decision Tree ◽  
Training Data ◽  
Classification Model ◽  
Optimal Solutions ◽  
Beam Search ◽  
Ensemble Of Classifiers ◽  
Local Optima ◽  
Computation Cost ◽  
Decision Tree Induction

Decision tree induction algorithms are highly used in a variety of domains for knowledge discovery and pattern recognition. They have the advantage of producing a comprehensible classification model and satisfactory accuracy levels in several application domains. Most well-known decision tree induction algorithms perform a greedy top-down strategy for node partitioning that may lead to sub-optimal solutions that overfit the training data. Some alternatives for the greedy strategy are the use of ensemble of classifiers or, more recently, the employment of the evolutionary algorithms (EA) paradigm to evolve decision trees by performing a global search in the space of candidate trees. Both strategies have their own disadvantages, like the lack of comprehensible solutions (in the case of ensembles) or the high computation cost of EAs. Hence, the authors of this chapter present a new algorithm that seeks to avoid being trapped in local-optima by doing a beam search during the decision tree growth. In addition, their strategy keeps the comprehensibility of the traditional methods and is much less time-consuming than evolutionary algorithms.

Flexible and Wearable GRF and EMG Sensors Enabled Locomotion Mode Recognition for IoHT Based In-home Rehabilitation

2020 ◽  
Author(s):  
Chaoming Fang ◽  
Yixuan Wang ◽  
Shuo Gao
Keyword(s):  
Sensitive Element ◽  
Small Volume ◽  
Peak Stress ◽  
Higher Learning ◽  
Force Sensing ◽  
Sensing System ◽  
Mode Recognition ◽  
Locomotion Mode ◽  
Stress Sensing ◽  
Manipulation Process

In order to quantify the manipulation process of acupuncture, in this article, a piezoelectric glove based wearable stress sensing system is presented. Served as the sensitive element with small volume and high tensile resistance, PVDF greatly meet the need of quantitative analysis. Through piezoelectric force sensing glove, the system is capable of detecting both perpendicular stress as well as shear stress. Besides, key parameters including peak stress at needle are detected and extracted, potentially allowing for a higher learning efficiency hence advancing the development of acupuncture.

scholarly journals A novel algorithm to detect non-wear time from raw accelerometer data using deep convolutional neural networks

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Shaheen Syed ◽  
Bente Morseth ◽  
Laila A. Hopstock ◽  
Alexander Horsch
Keyword(s):  
Threshold Value ◽  
Detection Algorithm ◽  
Classification Model ◽  
Type I ◽  
Accelerometer Data ◽  
Deep Convolutional Neural Networks ◽  
Major Drawback ◽  
Wear Time ◽  
Detection Algorithms ◽  
Type Ii Errors

AbstractTo date, non-wear detection algorithms commonly employ a 30, 60, or even 90 mins interval or window in which acceleration values need to be below a threshold value. A major drawback of such intervals is that they need to be long enough to prevent false positives (type I errors), while short enough to prevent false negatives (type II errors), which limits detecting both short and longer episodes of non-wear time. In this paper, we propose a novel non-wear detection algorithm that eliminates the need for an interval. Rather than inspecting acceleration within intervals, we explore acceleration right before and right after an episode of non-wear time. We trained a deep convolutional neural network that was able to infer non-wear time by detecting when the accelerometer was removed and when it was placed back on again. We evaluate our algorithm against several baseline and existing non-wear algorithms, and our algorithm achieves a perfect precision, a recall of 0.9962, and an F1 score of 0.9981, outperforming all evaluated algorithms. Although our algorithm was developed using patterns learned from a hip-worn accelerometer, we propose algorithmic steps that can easily be applied to a wrist-worn accelerometer and a retrained classification model.

scholarly journals Development of an Automatic Functional Movement Screening System with Inertial Measurement Unit Sensors

2020 ◽  
Vol 11 (1) ◽  
pp. 96
Author(s):  
Wen-Lan Wu ◽  
Meng-Hua Lee ◽  
Hsiu-Tao Hsu ◽  
Wen-Hsien Ho ◽  
Jing-Min Liang
Keyword(s):  
Feature Selection ◽  
Inertial Measurement Unit ◽  
Learning Algorithm ◽  
Kappa Statistic ◽  
Weighted Kappa ◽  
Classification Model ◽  
Measurement Unit ◽  
Functional Movement ◽  
Inertial Measurement ◽  
Movement Screening

Background: In this study, an automatic scoring system for the functional movement screen (FMS) was developed. Methods: Thirty healthy adults fitted with full-body inertial measurement unit sensors completed six FMS exercises. The system recorded kinematics data, and a professional athletic trainer graded each participant. To reduce the number of input variables for the predictive model, ordinal logistic regression was used for subset feature selection. The ensemble learning algorithm AdaBoost.M1 was used to construct classifiers. Accuracy and F score were used for classification model evaluation. The consistency between automatic and manual scoring was assessed using a weighted kappa statistic. Results: When all the features were used, the predict model presented moderate to high accuracy, with kappa values between fair to very good agreement. After feature selection, model accuracy decreased about 10%, with kappa values between poor to moderate agreement. Conclusions: The results indicate that higher prediction accuracy was achieved using the full feature set compared with using the reduced feature set.

A DECISION TREE-BASED CLASSIFICATION APPROACH TO RULE EXTRACTION FOR SECURITY ANALYSIS

2006 ◽  
Vol 05 (01) ◽  
pp. 227-240 ◽  
Author(s):  
N. REN ◽  
M. ZARGHAM ◽  
S. RAHIMI
Keyword(s):  
Decision Tree ◽  
High Performance ◽  
Security Analysis ◽  
Predictive Performance ◽  
Classification Model ◽  
Stock Selection ◽  
Selection Rules ◽  
Stock Portfolios ◽  
Decision Tree Classification ◽  
C4.5 Decision Tree

Stock selection rules are extensively utilized as the guideline to construct high performance stock portfolios. However, the predictive performance of the rules developed by some economic experts in the past has decreased dramatically for the current stock market. In this paper, C4.5 decision tree classification method was adopted to construct a model for stock prediction based on the fundamental stock data, from which a set of stock selection rules was derived. The experimental results showed that the generated rules have exceptional predictive performance. Moreover, it also demonstrated that the C4.5 decision tree classification model can work efficiently on the high noise stock data domain.

scholarly journals Facial Recognition System for People with and without Face Mask in Times of the COVID-19 Pandemic

2021 ◽  
Vol 13 (12) ◽  
pp. 6900
Author(s):  
Jonathan S. Talahua ◽  
Jorge Buele ◽  
P. Calvopiña ◽  
José Varela-Aldás
Keyword(s):  
Facial Recognition ◽  
Face Mask ◽  
Recognition System ◽  
Classification Model ◽  
World Health ◽  
The Face ◽  
Recognition Systems ◽  
Facial Recognition System ◽  
Health Organization ◽  
Face Detector

In the face of the COVID-19 pandemic, the World Health Organization (WHO) declared the use of a face mask as a mandatory biosafety measure. This has caused problems in current facial recognition systems, motivating the development of this research. This manuscript describes the development of a system for recognizing people, even when they are using a face mask, from photographs. A classification model based on the MobileNetV2 architecture and the OpenCv’s face detector is used. Thus, using these stages, it can be identified where the face is and it can be determined whether or not it is wearing a face mask. The FaceNet model is used as a feature extractor and a feedforward multilayer perceptron to perform facial recognition. For training the facial recognition models, a set of observations made up of 13,359 images is generated; 52.9% images with a face mask and 47.1% images without a face mask. The experimental results show that there is an accuracy of 99.65% in determining whether a person is wearing a mask or not. An accuracy of 99.52% is achieved in the facial recognition of 10 people with masks, while for facial recognition without masks, an accuracy of 99.96% is obtained.

Sign in / Sign up

Export Citation Format

Share Document