A neural network-based electromyography motion classifier for upper limb activities

The objective of the work is to investigate the classifcation of different movements based on the surface electromyogram(SEMG)pattern recognition method.The testing was conducted for four arm movements using several experiments with artificial neural network class fication scheme.Six time domain features were extracted and consequently dlassification was implemented using back propagation neural dassifier(BPNC).Further,the realization of projected network was verified using cross validation(CV)process;hence ANOVA algorithm was carried out.Performance of the network is analyzed by considering mean square error(MSE)value.A comparison was performed between the extracted feat ures and back propagation network results reported in the literature.The concurrent result indicates the significance of proposed network with classification accuracy(CA)of 100%recorded from two channels,while analysis of variance technique helps in investigating the effectiveness of classified sigmal for recognition tasks.

Spectral and mathematical evaluation of electromyography signals for clinical use

The surface electromyography （SEMG） is a complicated biomedical signal, generated during voluntary or involuntary muscle activities and these muscle activities are always controlled by the nervous system. In this paper, the processing and analysis of SEMG signals at multiple muscle points for different operations were carried out. Myoelectric signals were detected using designed acquisition setup which consists of an instrumenta- tion amplifier, filter circuit, an amplifier with gain adjustment. Fhrther, Labview^-based data programming code was used to record SEMG signals for independent activities. The whole system consists of bipolar noninvasive electrodes, signal acquisition protocols and signal conditioning at different levels. This work uses recorded SEMG signals generated by biceps and triceps muscles for four different arm activities. Feature extraction was done on the recorded signal for investigating the voluntary muscular contraction relationship for exercising statistic measured index method to evaluate distance between two independent groups by directly addressing the quality of signal in separability class for different arm movements. Thereafter repeated factorial analysis of variance technique was implemented to evaluate the effectiveness of processed signal. From these results, it demonstrates that the proposed method can be used as SEMG feature evaluation index.

Intuitive approach towards detection of attention tunneling while using a head-up display

Head-up display （HUD）, a primary cockpit display, helps in optimizing a pilot＇s attention towards aircraft and outside events. Slight mismatch in the balance may cause missed events; this phenomenon is called attention tunneling and affects the situational awareness of the pilot. This work reports an intuitive approach to detect attention tunneling while use of HUD in aircrafts. Texture analysis of a composite HUD camera video provided three distinguishing parameters, viz., contrast, correlation, and homogeneity. These three texture parameters are used as inputs for a fuzzy inference-based assistive detection system which could be used for distinguishing tunneled and nontunneled HUD operation.