Development of a human-friendly visual inspection method for painted vehicle bodies.

Visual inspections performed in the final stage of the vehicle manufacturing process are crucial for assuring the quality of painted vehicle bodies. However, lengthy visual inspections can cause fatigue and discomfort of the eyes, which may adversely affect inspection accuracy and efficiency. This study developed a new human-friendly visual inspection method for the detection of defects (e.g., scar and dent) on vehicle bodies, and compared its performance to a conventional inspection method in terms of critical fusion frequency (Hz, indirect measure of eye fatigue), defect detectability (%), and subjective satisfaction score (7-point Likert scale). The new method was devised to project bright-dark linear stripes onto the surface of the vehicle body and created emergent features (distorted stripes) where a defect existed. The critical fusion frequency of the new method decreased slightly (3.7%) after a 30-minute visual inspection task, whereas that of the conventional method dropped substantially (11.0%), which implied more severe eye fatigue. Additionally, the new method had significantly higher defect detectability (92.1%) and satisfaction score (5.8 points) than those (73.4% and 3.5 points) of the conventional method.

Analysis of natural finger-press motions for design of trackball buttons.

This study analysed natural press motions of the index, middle and ring fingers for ergonomic design of the positions and surface angles of the left, middle and right trackball buttons. Finger motions of 26 male participants for naturally pressing the trackball buttons were recorded after the participants adjusted the trackball buttons to their preferred locations for comfortable pressing. The natural positions of the finger pulps formed a symmetrically rainbow-shaped reach zone for the fingers. The natural press angles of the fingers' motion trajectories to the vertical reference line ranged from 14.2° to 20.5°, suggesting an 18-degree surface from the horizontal line for the trackball buttons. Regression formulas (adjusted R2 = 0.90 ± 0.07 and mean squared error = 8.55 ± 7.52 mm) were established to estimate the natural positions of finger pulps from hand segment lengths and joint angles for a population having different hand sizes from this study. Relevance to industry.

Effects of button position on a soft keyboard: Muscle activity, touch time, and discomfort in two-thumb text entry.

Intensive use of the thumbs for text entry on smartphones may contribute to discomfort, pain, or musculoskeletal disorders. This study investigated the effect of twenty-five button positions (5 rows × 5 columns) on a soft keyboard for two-thumb entry. Two experiments measured muscle activity, touch time, and discomfort as a function of the button positions. In Phase I, the muscle activities of two intrinsic (abductor pollicis brevis and first dorsal interossei) and two extrinsic (abductor pollicis longus and extensor digitorum communis) muscles associated with thumb motions were observed for ten college students (age: 24.2). In Phase II, touch time and discomfort were measured for 40 college students (age: 23.6). The results demonstrated that the %MVCs of the intrinsic muscles significantly increased when the thumbs flexed and abducted. Also, the button positions near the rest positions of the thumbs resulted in significantly shorter touch times (0.66 s) and lower discomfort ratings (0.70 pt) than their peripheral buttons (0.76 s; 2.29 pt).

Classification of a Driver's cognitive workload levels using artificial neural network on ECG signals.

An artificial neural network (ANN) model was developed in the present study to classify the level of a driver's cognitive workload based on electrocardiography (ECG). ECG signals were measured on 15 male participants while they performed a simulated driving task as a primary task with/without an N-back task as a secondary task. Three time-domain ECG measures (mean inter-beat interval (IBI), standard deviation of IBIs, and root mean squared difference of adjacent IBIs) and three frequencydomain ECG measures (power in low frequency, power in high frequency, and ratio of power in low and high frequencies) were calculated. To compensate for individual differences in heart response during the driving tasks, a three-step data processing procedure was performed to ECG signals of each participant: (1) selection of two most sensitive ECG measures, (2) definition of three (low, medium, and high) cognitive workload levels, and (3) normalization of the selected ECG measures. An ANN model was constructed using a feed-forward network and scaled conjugate gradient as a back-propagation learning rule. The accuracy of the ANN classification model was found satisfactory for learning data (95%) and testing data (82%).