Multi-Stroke handwriting character recognition based on sEMG using convolutional-recurrent neural networks

Beltrán-Hernández, José Guadalupe; Ruiz-Pinales, José; López-Rodríguez, Pedro; López-Ramírez, José Luis; Aviña-Cervantes, Juan Gabriel

doi:10.3934/mbe.2020293

Cited by 18 publications

(17 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[38] proposed a template matching framework for recognition of handwriting from sEMG signals. In [39], the authors proposed a methodology based on sEMG signals to recognize multi-user free-style handwriting characters. A CNN-LSTM based framework for classification of 36 handwritten characters (A-Z, 0-9) using filtered EMG signals was proposed in their study.…”

Section: B Related Workmentioning

confidence: 99%

SurfMyoAiR: A surface Electromyography based framework for Airwriting Recognition

Tripathi¹,

Kumar²,

Prathosh³

et al. 2022

Preprint

View full text Add to dashboard Cite

Airwriting Recognition is the task of identifying letters written in free space with finger movement. It is a dynamic gesture recognition with the vocabulary of gestures corresponding to letters in a given language. Electromyography (EMG) is a technique used to record electrical activity during muscle contraction and relaxation as a result of movement and is widely used for gesture recognition. Most of the current research in gesture recognition is focused on identifying static gestures. However, dynamic gestures are natural and userfriendly for being used as alternate input methods in Human-Computer Interaction applications. Airwriting recognition using EMG signals recorded from forearm muscles is therefore a viable solution. Since the user does not need to learn any new gestures and a large range of words can be formed by concatenating these letters, it is generalizable to a wider population. There has been limited work in recognition of airwriting using EMG signals and forms the core idea of the current work. The SurfMyoAiR dataset comprising of EMG signals recorded during writing English uppercase alphabets is constructed. Several different time-domain features to construct EMG envelope and two different timefrequency image representations: Short-Time Fourier Transform and Continuous Wavelet Transform were explored to form the input to a deep learning model for airwriting recognition. Several different deep learning architectures were exploited for this task. Additionally, the effect of various parameters such as signal length, window length and interpolation techniques on the recognition performance is comprehensively explored. The best-achieved accuracy was 78.50% and 62.19% in userdependent and independent scenarios respectively by using Short-Time Fourier Transform in conjunction with a 2D Convolutional Neural Network based classifier. Airwriting has great potential as a user-friendly modality to be used as an alternate input method in Human-Computer Interaction applications.

show abstract

Section: B Related Workmentioning

confidence: 99%

SurfMyoAiR: A surface Electromyography based framework for Airwriting Recognition

Tripathi¹,

Kumar²,

Prathosh³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The artificial neural network is combined with the convolutional structure to form a CNN, in which the feature extraction process of the input data is completed in the convolutional layer, which does not require too much preprocessing, and the data dimensionality reduction is in the pooling layer. Such extraction and dimensionality reduction reduces the computational load of the network model [17][18].…”

Section: Basic Structure Of Cnnmentioning

confidence: 99%

Print Character Recognition Method Based on Convolutional Neural Network

2022

View full text Add to dashboard Cite

With the rapid development of the economy, digital transformation has become a future trend. Character recognition (CR) technology, as a key technology to realize digitalization, has always been a hot research topic that people pay attention to. CR can be widely used in the fields of transportation, finance, industry, economy and artificial intelligence. In-depth study of CR technology is of great practical significance for promoting the development of science and technology and economy. In this context, this paper studies the printed CR method based on CNN. Combining wavelet theory and CNN, this paper studies the extraction and recognition methods and implementation ways of printed characters in detail. The algorithm applied in this paper is recognized on the DSP (DM642) chip, the recognition accuracy is over 99.93%, and the recognition time is controlled within 5ms, which meets the actual requirements. Comprehensive analysis verifies the feasibility and superiority of the recognition algorithm applied in this paper.

show abstract

“…Such interfaces find application in prosthetic control of upper and lower bionic limbs [3,4], but also for the realization of intelligent human-computer interactions in virtual and augmented reality, and for biometric identification [5][6][7][8]. Myoelectric interfaces can also find application in modern scenarios since they can be used to decode handwritten characters or digits [9,10], supporting the development of immersive rehabilitation protocols with a consistent involvement of the cognitive centres of the brain [11].…”

Section: Introductionmentioning

confidence: 99%

“…On the other hand, EMG can directly mirror the motor control volition and data and can be used in a sequential manner to create architectures that recognize the handwriting online., i.e., with an update frequency that can be used to making the interaction more fluent [10]. Hence, although EMG has been considered as a potential option for handwriting recognition, only a few studies have explored its use in classifying digits or letters [9,10]. The recognition of handwritten characters from EMG data has been tackled using template matching, dynamic time warping, and deep-learning methods [9,23,24].…”

Section: Introductionmentioning

confidence: 99%

“…Hence, although EMG has been considered as a potential option for handwriting recognition, only a few studies have explored its use in classifying digits or letters [9,10]. The recognition of handwritten characters from EMG data has been tackled using template matching, dynamic time warping, and deep-learning methods [9,23,24]. However, these approaches require large databases and impose a high computational burden during the training phase, making the practical applicability of such architectures limited.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Intelligent Human–Computer Interaction: Combined Wrist and Forearm Myoelectric Signals for Handwriting Recognition

Tigrini,

Ranaldi,

Verdini

et al. 2024

Bioengineering

View full text Add to dashboard Cite

Recent studies have highlighted the possibility of using surface electromyographic (EMG) signals to develop human–computer interfaces that are also able to recognize complex motor tasks involving the hand as the handwriting of digits. However, the automatic recognition of words from EMG information has not yet been studied. The aim of this study is to investigate the feasibility of using combined forearm and wrist EMG probes for solving the handwriting recognition problem of 30 words with consolidated machine-learning techniques and aggregating state-of-the-art features extracted in the time and frequency domains. Six healthy subjects, three females and three males aged between 25 and 40 years, were recruited for the study. Two tests in pattern recognition were conducted to assess the possibility of classifying fine hand movements through EMG signals. The first test was designed to assess the feasibility of using consolidated myoelectric control technology with shallow machine-learning methods in the field of handwriting detection. The second test was implemented to assess if specific feature extraction schemes can guarantee high performances with limited complexity of the processing pipeline. Among support vector machine, linear discriminant analysis, and K-nearest neighbours (KNN), the last one showed the best classification performances in the 30-word classification problem, with a mean accuracy of 95% and 85% when using all the features and a specific feature set known as TDAR, respectively. The obtained results confirmed the validity of using combined wrist and forearm EMG data for intelligent handwriting recognition through pattern recognition approaches in real scenarios.

show abstract

Multi-Stroke handwriting character recognition based on sEMG using convolutional-recurrent neural networks

Cited by 18 publications

References 1 publication

SurfMyoAiR: A surface Electromyography based framework for Airwriting Recognition

SurfMyoAiR: A surface Electromyography based framework for Airwriting Recognition

Print Character Recognition Method Based on Convolutional Neural Network

Intelligent Human–Computer Interaction: Combined Wrist and Forearm Myoelectric Signals for Handwriting Recognition

Contact Info

Product

Resources

About