Comparison and Evaluation of Machine Learning-Based Classification of Hand Gestures Captured by Inertial Sensors
Gesture recognition is a topic in computer science and language technology that aims to interpret human gestures with computer programs and many different algorithms. It can be seen as the way computers can understand human body language. Today, the main interaction tools between computers and human...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-09-01
|
| Series: | Computation |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-3197/10/9/159 |
| _version_ | 1797489719238459392 |
|---|---|
| author | Ivo Stančić Josip Musić Tamara Grujić Mirela Kundid Vasić Mirjana Bonković |
| author_facet | Ivo Stančić Josip Musić Tamara Grujić Mirela Kundid Vasić Mirjana Bonković |
| author_sort | Ivo Stančić |
| collection | DOAJ |
| description | Gesture recognition is a topic in computer science and language technology that aims to interpret human gestures with computer programs and many different algorithms. It can be seen as the way computers can understand human body language. Today, the main interaction tools between computers and humans are still the keyboard and mouse. Gesture recognition can be used as a tool for communication with the machine and interaction without any mechanical device such as a keyboard or mouse. In this paper, we present the results of a comparison of eight different machine learning (ML) classifiers in the task of human hand gesture recognition and classification to explore how to efficiently implement one or more tested ML algorithms on an 8-bit AVR microcontroller for on-line human gesture recognition with the intention to gesturally control the mobile robot. The 8-bit AVR microcontrollers are still widely used in the industry, but due to their lack of computational power and limited memory, it is a challenging task to efficiently implement ML algorithms on them for on-line classification. Gestures were recorded by using inertial sensors, gyroscopes, and accelerometers placed at the wrist and index finger. One thousand and eight hundred (1800) hand gestures were recorded and labelled. Six important features were defined for the identification of nine different hand gestures using eight different machine learning classifiers: Decision Tree (DT), Random Forests (RF), Logistic Regression (LR), Linear Discriminant Analysis (LDA), Support Vector Machine (SVM) with linear kernel, Naïve Bayes classifier (NB), K-Nearest Neighbours (KNN), and Stochastic Gradient Descent (SGD). All tested algorithms were ranged according to Precision, Recall, and F1-score (abb.: P-R-F1). The best algorithms were SVM (P-R-F1: 0.9865, 0.9861, and 0.0863), and RF (P-R-F1: 0.9863, 0.9861, and 0.0862), but their main disadvantage is their unusability for on-line implementations in 8-bit AVR microcontrollers, as proven in the paper. The next best algorithms have had only slightly poorer performance than SVM and RF: KNN (P-R-F1: 0.9835, 0.9833, and 0.9834) and LR (P-R-F1: 0.9810, 0.9810, and 0.9810). Regarding the implementation on 8-bit microcontrollers, KNN has proven to be inadequate, like SVM and RF. However, the analysis for LR has proved that this classifier could be efficiently implemented on targeted microcontrollers. Having in mind its high F1-score (comparable to SVM, RF, and KNN), this leads to the conclusion that the LR is the most suitable classifier among tested for on-line applications in resource-constrained environments, such as embedded devices based on 8-bit AVR microcontrollers, due to its lower computational complexity in comparison with other tested algorithms. |
| first_indexed | 2024-03-10T00:21:39Z |
| format | Article |
| id | doaj.art-3904cebd5b284b288e635c4e0c15afa7 |
| institution | Directory Open Access Journal |
| issn | 2079-3197 |
| language | English |
| last_indexed | 2024-03-10T00:21:39Z |
| publishDate | 2022-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Computation |
| spelling | doaj.art-3904cebd5b284b288e635c4e0c15afa72023-11-23T15:41:39ZengMDPI AGComputation2079-31972022-09-0110915910.3390/computation10090159Comparison and Evaluation of Machine Learning-Based Classification of Hand Gestures Captured by Inertial SensorsIvo Stančić0Josip Musić1Tamara Grujić2Mirela Kundid Vasić3Mirjana Bonković4Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture—FESB, University of Split, 21000 Split, CroatiaFaculty of Electrical Engineering, Mechanical Engineering and Naval Architecture—FESB, University of Split, 21000 Split, CroatiaFaculty of Electrical Engineering, Mechanical Engineering and Naval Architecture—FESB, University of Split, 21000 Split, CroatiaFaculty of Mechanical Engineering, Computing and Electrical Engineering, University of Mostar, 88000 Mostar, Bosnia and HerzegovinaFaculty of Electrical Engineering, Mechanical Engineering and Naval Architecture—FESB, University of Split, 21000 Split, CroatiaGesture recognition is a topic in computer science and language technology that aims to interpret human gestures with computer programs and many different algorithms. It can be seen as the way computers can understand human body language. Today, the main interaction tools between computers and humans are still the keyboard and mouse. Gesture recognition can be used as a tool for communication with the machine and interaction without any mechanical device such as a keyboard or mouse. In this paper, we present the results of a comparison of eight different machine learning (ML) classifiers in the task of human hand gesture recognition and classification to explore how to efficiently implement one or more tested ML algorithms on an 8-bit AVR microcontroller for on-line human gesture recognition with the intention to gesturally control the mobile robot. The 8-bit AVR microcontrollers are still widely used in the industry, but due to their lack of computational power and limited memory, it is a challenging task to efficiently implement ML algorithms on them for on-line classification. Gestures were recorded by using inertial sensors, gyroscopes, and accelerometers placed at the wrist and index finger. One thousand and eight hundred (1800) hand gestures were recorded and labelled. Six important features were defined for the identification of nine different hand gestures using eight different machine learning classifiers: Decision Tree (DT), Random Forests (RF), Logistic Regression (LR), Linear Discriminant Analysis (LDA), Support Vector Machine (SVM) with linear kernel, Naïve Bayes classifier (NB), K-Nearest Neighbours (KNN), and Stochastic Gradient Descent (SGD). All tested algorithms were ranged according to Precision, Recall, and F1-score (abb.: P-R-F1). The best algorithms were SVM (P-R-F1: 0.9865, 0.9861, and 0.0863), and RF (P-R-F1: 0.9863, 0.9861, and 0.0862), but their main disadvantage is their unusability for on-line implementations in 8-bit AVR microcontrollers, as proven in the paper. The next best algorithms have had only slightly poorer performance than SVM and RF: KNN (P-R-F1: 0.9835, 0.9833, and 0.9834) and LR (P-R-F1: 0.9810, 0.9810, and 0.9810). Regarding the implementation on 8-bit microcontrollers, KNN has proven to be inadequate, like SVM and RF. However, the analysis for LR has proved that this classifier could be efficiently implemented on targeted microcontrollers. Having in mind its high F1-score (comparable to SVM, RF, and KNN), this leads to the conclusion that the LR is the most suitable classifier among tested for on-line applications in resource-constrained environments, such as embedded devices based on 8-bit AVR microcontrollers, due to its lower computational complexity in comparison with other tested algorithms.https://www.mdpi.com/2079-3197/10/9/159hand gesturesinertial sensorsmachine learning algorithmsmodel optimisationoff-line classificationevaluation |
| spellingShingle | Ivo Stančić Josip Musić Tamara Grujić Mirela Kundid Vasić Mirjana Bonković Comparison and Evaluation of Machine Learning-Based Classification of Hand Gestures Captured by Inertial Sensors Computation hand gestures inertial sensors machine learning algorithms model optimisation off-line classification evaluation |
| title | Comparison and Evaluation of Machine Learning-Based Classification of Hand Gestures Captured by Inertial Sensors |
| title_full | Comparison and Evaluation of Machine Learning-Based Classification of Hand Gestures Captured by Inertial Sensors |
| title_fullStr | Comparison and Evaluation of Machine Learning-Based Classification of Hand Gestures Captured by Inertial Sensors |
| title_full_unstemmed | Comparison and Evaluation of Machine Learning-Based Classification of Hand Gestures Captured by Inertial Sensors |
| title_short | Comparison and Evaluation of Machine Learning-Based Classification of Hand Gestures Captured by Inertial Sensors |
| title_sort | comparison and evaluation of machine learning based classification of hand gestures captured by inertial sensors |
| topic | hand gestures inertial sensors machine learning algorithms model optimisation off-line classification evaluation |
| url | https://www.mdpi.com/2079-3197/10/9/159 |
| work_keys_str_mv | AT ivostancic comparisonandevaluationofmachinelearningbasedclassificationofhandgesturescapturedbyinertialsensors AT josipmusic comparisonandevaluationofmachinelearningbasedclassificationofhandgesturescapturedbyinertialsensors AT tamaragrujic comparisonandevaluationofmachinelearningbasedclassificationofhandgesturescapturedbyinertialsensors AT mirelakundidvasic comparisonandevaluationofmachinelearningbasedclassificationofhandgesturescapturedbyinertialsensors AT mirjanabonkovic comparisonandevaluationofmachinelearningbasedclassificationofhandgesturescapturedbyinertialsensors |