A stable posture requires the coordination of multiple joints of the body. This coordination of the multiple joints of the human body to maintain a stable posture is a subject of research. The number of degrees of freedom (DOFs) of the human motor system is considerably larger than the DOFs required for posture balance. The manner of managing this redundancy by the central nervous system remains unclear. To understand this phenomenon, in this study, three local inter-joint coordination pattern (IJCP) features were introduced to characterize the strength, changing velocity, and complexity of the inter-joint couplings by computing the correlation coefficients between joint velocity signal pairs. In addition, for quantifying the complexity of IJCPs from a global perspective, another set of IJCP features was introduced by performing principal component analysis on all joint velocity signals. A Microsoft Kinect depth sensor was used to acquire the motion of 15 joints of the body. The efficacy of the proposed features was tested using the captured motions of two age groups (18–24 and 65–73 years) when standing still. With regard to the redundant DOFs of the joints of the body, the experimental results suggested that an inter-joint coordination strategy intermediate to that of the two extreme coordination modes of total joint dependence and independence is used by the body. In addition, comparative statistical results of the proposed features proved that aging increases the coupling strength, decreases the changing velocity, and reduces the complexity of the IJCPs. These results also suggested that with aging, the balance strategy tends to be more joint dependent. Because of the simplicity of the proposed features and the affordability of the easy-to-use Kinect depth sensor, such an assembly can be used to collect large amounts of data to explore the potential of the proposed features in assessing the performance of the human balance control system.

Dynamic voltage and frequency scaling (DVFS) is a well-known method for saving energy consumption. Several DVFS studies have applied learning-based methods to implement the DVFS prediction model instead of complicated mathematical models. This paper proposes a lightweight learning-directed DVFS method that involves using counter propagation networks to sense and classify the task behavior and predict the best voltage/frequency setting for the system. An intelligent adjustment mechanism for performance is also provided to users under various performance requirements. The comparative experimental results of the proposed algorithms and other competitive techniques are evaluated on the NVIDIA JETSON Tegra K1 multicore platform and Intel PXA270 embedded platforms. The results demonstrate that the learning-directed DVFS method can accurately predict the suitable central processing unit (CPU) frequency, given the runtime statistical information of a running program, and achieve an energy savings rate up to 42%. Through this method, users can easily achieve effective energy consumption and performance by specifying the factors of performance loss.

Most object detection models cannot achieve satisfactory performance under nighttime and other insufficient illumination conditions, which may be due to the collection of data sets and typical labeling conventions. Public data sets collected for object detection are usually photographed with sufficient ambient lighting. However, their labeling conventions typically focus on clear objects and ignore blurry and occluded objects. Consequently, the detection performance levels of traditional vehicle detection techniques are limited in nighttime environments without sufficient illumination. When objects occupy a small number of pixels and the existence of crucial features is infrequent, traditional convolutional neural networks (CNNs) may suffer from serious information loss due to the fixed number of convolutional operations. This study presents solutions for data collection and the labeling convention of nighttime data to handle various types of situations, including in-vehicle detection. Moreover, the study proposes a specifically optimized system based on the Faster region-based CNN model. The system has a processing speed of 16 frames per second for 500 × 375-pixel images, and it achieved a mean average precision (mAP) of 0.8497 in our validation segment involving urban nighttime and extremely inadequate lighting conditions. The experimental results demonstrated that our proposed methods can achieve high detection performance in various nighttime environments, such as urban nighttime conditions with insufficient illumination, and extremely dark conditions with nearly no lighting. The proposed system outperforms original methods that have an mAP value of approximately 0.2.

In this study, a head-mounted device was developed to track the gaze of the eyes and estimate the gaze point on the user’s visual plane. To provide a cost-effective vision tracking solution, this head-mounted device is combined with a sized endoscope camera, infrared light, and mobile phone; the devices are also implemented via 3D printing to reduce costs. Based on the proposed image pre-processing techniques, the system can efficiently extract and estimate the pupil ellipse from the camera module. A 3D eye model was also developed to effectively locate eye gaze points from extracted eye images. In the experimental results, average accuracy, precision, and recall rates of the proposed system can achieve an average of over 97%, which can demonstrate the efficiency of the proposed system. This study can be widely used in the Internet of Things, virtual reality, assistive devices, and human-computer interaction applications.

Numerous vehicle detection methods have been proposed to obtain trustworthy traffic data for the development of intelligent traffic systems. Most of these methods perform sufficiently well under common scenarios, such as sunny or cloudy days; however, the detection accuracy drastically decreases under various bad weather conditions, such as rainy days or days with glare, which normally happens during sunset. This study proposes a vehicle detection system with a visibility complementation module that improves detection accuracy under various bad weather conditions. Furthermore, the proposed system can be implemented without retraining the deep learning models for object detection under different weather conditions. The complementation of the visibility was obtained through the use of a dark channel prior and a convolutional encoder–decoder deep learning network with dual residual blocks to resolve different effects from different bad weather conditions. We validated our system on multiple surveillance videos by detecting vehicles with the You Only Look Once (YOLOv3) deep learning model and demonstrated that the computational time of our system could reach 30 fps on average; moreover, the accuracy increased not only by nearly 5% under low-contrast scene conditions but also 50% under rainy scene conditions. The results of our demonstrations indicate that our approach is able to detect vehicles under various bad weather conditions without the need to retrain a new model.