Abstract

Abstract

Abstract
Facial expression analysis has a wide area of applications including health, psychology, sports etc. In this study, we explored different methods of automatic classification of exercise intensities using facial image processing of a subject performing exercise on a cycloergometer during an incremental standardized protocol. The method can be implemented in real time using facial video analysis. The experiments were done with images extracted from a 12 min HD video collected in laboratorial normalized settings (TechSport from the University of Trás-os-Montes e Alto Douro) with a static camera (90° angle with face and camera). The time slot for video to extract images for a particular class of exercise intensity is correspondence to the incremental heart rate. The facial expression recognition has been performed mainly in two steps: facial landmark detection and classification using the facial landmarks. Luxand application was used to detect 70 landmarks were detect using the adaptation of code available in Luxand application and we applied machine learning classification algorithms including discriminant analysis, KNN and SVM to classify the exercise intensities from the facial images. KNN algorithms presents up to 100% accuracy in classification into 2 and 3 classes. The distances between a lowermost landmark of the faces, which is indicated in landmark number 11 in the Luxand application, and the 26 landmarks around mouth were calculated and considered as features vector to train and test the classifier. Separate experiments were done for classification into two, three, and four classes and the accuracy of each algorithm was analyzed. From the overall results, classification into two and three classes was easy and resulted in very good classification performance whereas the classification with four classes had poor classification performance in each algorithm. Preliminary results suggest that distinguishing more levels of exertion, might require additional feature variables. © 2018 IEEE.

Abstract
Facial expression recognition (FER) is currently one of the most active research topics due to its wide range of applications in the human-computer interaction field. An important part of the recent success of automatic FER was achieved thanks to the emergence of deep learning approaches. However, training deep networks for FER is still a very challenging task, since most of the available FER data sets are relatively small. Although transfer learning can partially alleviate the issue, the performance of deep models is still below of its full potential as deep features may contain redundant information from the pre-trained domain. Instead, we propose a novel end-to-end neural network architecture along with a well-designed loss function based on the strong prior knowledge that facial expressions are the result of the motions of some facial muscles and components. The loss function is defined to regularize the entire learning process so that the proposed neural network is able to explicitly learn expression-specific features. Experimental results demonstrate the effectiveness of the proposed model in both lab-controlled and wild environments. In particular, the proposed neural network provides quite promising results, outperforming in most cases the current state-of-the-art methods.

Abstract
The use of Wireless Sensor Network (WSN) based technologies is an attractive solution for large-scale sensing applications (wide area deployment), such as environmental monitoring, precision agriculture and industrial automation. IEEE 802.15.4/ZigBee standards are the most used communication protocols for WSN technologies, where the cluster-tree topology is pointed out as a suitable topology to support the implementation of large-scale WSNs. These networks are usually scheduled to prioritise convergecast (upstream) traffic generated from sensor nodes toward the sink node. However, this scheduling pattern results in higher delays for control messages (downstream traffic). Within this context, this paper proposes a Hybrid Beacon Scheduling (Fast-HyBeS) scheme to enable the periodic reconfiguration of cluster-tree WSNs. The underlying idea is to periodically schedule a downstream opportunity window, to allow a faster dissemination of control messages. This opportunity window follows a top-down scheduling approach that prioritises the downstream traffic. Simulation results show that the use of Fast-HyBeS can significantly decrease the end-to-end communication delay for control messages, when compared to the use of static convergecast scheduling schemes. Moreover, the simulation results also highlight that the Fast-HyBeS has a negligible impact upon end-to-end communication delays of the monitoring traffic.

Abstract
Sensory tests are quality assurance tools commonly used to measure and/or detect the presence of abnormal characteristics perceived through the senses in lots of raw material and final products in many manufacturing and food industries. In this paper two acceptance sampling plans for sensory evaluation are designed, and an illustration of the performance of such plans applied to a real data set is presented.