Current Issue : April - June Volume : 2021 Issue Number : 2 Articles : 5 Articles
The Arabic language has many complex grammar rules that may seem complicated to the average user or learner. Automatic grammar checking systems can improve the quality of the text, reduce the costs of the proofreading process, and play a role in grammar teaching. This paper presents an initiative toward developing a novel and comprehensive Arabic auditor that can address vowelized texts. We called the “Arabic Grammar Detector” (AGD- ). AGD was successfully implemented based on a dependency grammar and decision tree classifier model. Its purpose is to extract patterns of grammatical rules from a projective dependency graph in order to designate the appropriate syntax dependencies of a sentence. The current implementation covers almost all regular Arabic grammar rules for nonvowelized texts as well as partially or fully vowelized texts. AGD was evaluated using the Tashkeela corpus. It can detect more than 94% of grammatical errors and hint at their causes and possible corrections....
Brain–computer interface (BCI)-guided robot-assisted training strategy has been increasingly applied to stroke rehabilitation, while few studies have investigated the neuroplasticity change and functional reorganization after intervention from multimodality neuroimaging perspective. The present study aims to investigate the hemodynamic and electrophysical changes induced by BCI training using functional magnetic resonance imaging (fMRI) and electroencephalography (EEG) respectively, as well as the relationship between the neurological changes and motor function improvement. Fourteen chronic stroke subjects received 20 sessions of BCI-guided robot hand training. Simultaneous EEG and fMRI data were acquired before and immediately after the intervention. Seed-based functional connectivity for resting-state fMRI data and effective connectivity analysis for EEG were processed to reveal the neuroplasticity changes and interaction between different brain regions. Moreover, the relationship among motor function improvement, hemodynamic changes, and electrophysical changes derived from the two neuroimaging modalities was also investigated. This work suggested that (a) significant motor function improvement could be obtained after BCI training therapy, (b) training effect significantly correlated with functional connectivity change between ipsilesional M1 (iM1) and contralesional Brodmann area 6 (including premotor area (cPMA) and supplementary motor area (SMA)) derived from fMRI, (c) training effect significantly correlated with information flow change from cPMA to iM1 and strongly correlated with information flow change from SMA to iM1 derived from EEG, and (d) consistency of fMRI and EEG results illustrated by the correlation between functional connectivity change and information flow change. Our study showed changes in the brain after the BCI training therapy from chronic stroke survivors and provided a better understanding of neural mechanisms, especially the interaction among motor-related brain regions during stroke recovery. Besides, our finding demonstrated the feasibility and consistency of combining multiple neuroimaging modalities to investigate the neuroplasticity change....
In recent years, with the continuous development of artificial intelligence and brain-computer interface technology, emotion recognition based on physiological signals, especially, electroencephalogram (EEG) signals, has become a popular research topic and attracted wide attention. However, how to extract effective features from EEG signals and accurately recognize them by classifiers have also become an increasingly important task. Therefore, in this paper, we propose an emotion recognition method of EEG signals based on the ensemble learning method, AdaBoost. First, we consider the time domain, time-frequency domain, and nonlinear features related to emotion, extract them from the preprocessed EEG signals, and fuse the features into an eigenvector matrix. Then, the linear discriminant analysis feature selection method is used to reduce the dimensionality of the features.Next, we use the optimized feature sets and train a classifier based on the ensemble learningmethod, AdaBoost, for binary classification. Finally, the proposed method has been tested in the DEAP data set on four emotional dimensions: valence, arousal, dominance, and liking. The proposed method is proved to be effective in emotion recognition, and the best average accuracy rate can reach up to 88.70% on the dominance dimension. Compared with other existing methods, the performance of the proposed method is significantly improved....
Several yield monitors are available for use on cotton harvesters, but none are able to maintain yield measurement accuracy across cultivars and field conditions that vary spatially and/or temporally. Thus, the utility of yield monitors as tools for on-farm research is limited unless steps are taken to calibrate the systems as cultivars and conditions change. This technical note details the man-machine-interface software system design portion of a harvester-based yield monitor calibration system for basket-type cotton strippers. The system was based upon the use of pressure sensors to measure the weight of the basket by monitoring the static pressure in the hydraulic lift cylinder circuit. To ensure accurate weighing, the system automatically lifted the basket to a target lift height, allowed basket time to settle, then weighed the contents of the basket. The software running the system was split into two parts that were run on an embedded low-level micro-controller, and a mobile computer located in the harvester cab. The system was field tested under commercial conditions and found to measure basket load weights within 2.5% of the reference scale. As such, the system was proven to be capable of providing an on-board auto-correction to a yield monitor for use in multi-variety field trials....
We described a real-time hair segmentation method based on a fully convolutional network with the basic structure of an encoder–decoder. In one of the traditional computer vision techniques for hair segmentation, the mean shift and watershed methodologies suffer from inaccuracy and slow execution due to multi-step, complex image processing. It is also difficult to execute the process in real-time unless an optimization technique is applied to the partition. To solve this problem, we exploited Mobile-Unet using the U-Net segmentation model, which incorporates the optimization techniques of MobileNetV2. In experiments, hair segmentation accuracy was evaluated by different genders and races, and the average accuracy was 89.9%. By comparing the accuracy and execution speed of our model with those of other models in related studies, we confirmed that the proposed model achieved the same or better performance. As such, the results of hair segmentation can obtain hair information (style, color, length), which has a significant impact on human-robot interaction with people....
Loading....