Skeleton Driven Action Recognition Using an Image-Based Spatial-Temporal Representation and Convolution Neural Network.

Individuals with Autism Spectrum Disorder (ASD) typically present difficulties in engaging and interacting with their peers. Thus, researchers have been developing different technological solutions as support tools for children with ASD. Social robots, one example of these technological solutions, are often unaware of their game partners, preventing the automatic adaptation of their behavior to the user. Information that can be used to enrich this interaction and, consequently, adapt the system behavior is the recognition of different actions of the user by using RGB cameras or/and depth sensors. The present work proposes a method to automatically detect in real-time typical and stereotypical actions of children with ASD by using the Intel RealSense and the Nuitrack SDK to detect and extract the user joint coordinates. The pipeline starts by mapping the temporal and spatial joints dynamics onto a color image-based representation. Usually, the position of the joints in the final image is clustered into groups. In order to verify if the sequence of the joints in the final image representation can influence the model's performance, two main experiments were conducted where in the first, the order of the grouped joints in the sequence was changed, and in the second, the joints were randomly ordered. In each experiment, statistical methods were used in the analysis. Based on the experiments conducted, it was found statistically significant differences concerning the joints sequence in the image, indicating that the order of the joints might impact the model's performance. The final model, a Convolutional Neural Network (CNN), trained on the different actions (typical and stereotypical), was used to classify the different patterns of behavior, achieving a mean accuracy of 92.4% ± 0.0% on the test data. The entire pipeline ran on average at 31 FPS.

Multilevel model of safety climate for furniture industries.

BACKGROUND: Furniture companies can analyze their safety status using quantitative measures. However, the data needed are not always available and the number of accidents is under-reported. Safety climate scales may be an alternative. However, there are no validated Portuguese scales that account for the specific attributes of the furniture sector. OBJECTIVE: The current study aims to develop and validate an instrument that uses a multilevel structure to measure the safety climate of the Portuguese furniture industry. METHODS: The Safety Climate in Wood Industries (SCWI) model was developed and applied to the safety climate analysis using three different scales: organizational, group and individual. A multilevel exploratory factor analysis was performed to analyze the factorial structure. The studied companies' safety conditions were also analyzed. RESULTS: Different factorial structures were found between and within levels. In general, the results show the presence of a group-level safety climate. The scores of safety climates are directly and positively related to companies' safety conditions; the organizational scale is the one that best reflects the actual safety conditions. CONCLUSIONS: The SCWI instrument allows for the identification of different safety climates in groups that comprise the same furniture company and it seems to reflect those groups' safety conditions. The study also demonstrates the need for a multilevel analysis of the studied instrument.

A digital calliper for training and study purposes.

A quick and valid method for evaluating percentage body fat is based on the use of skinfold callipers. However, limitations associated to their use and characteristics led the authors to improve a traditional calliper (Harpenden) and to integrate it with a software application. Such a measuring system, LipoTool, is meant to have better accuracy and reliability, including data processing and digital recording at a very low cost. At first, a sample of 49 older adults was used to evaluate the performance of LipoTool by comparing its results to those obtained with the traditional Harpenden calliper. A strong positive association in %BF was achieved. This digital sensing system was later improved by incorporating wireless communication between the calliper and the software application, adding other functionalities. The software application works in any computer and is flexible to incorporate new coming models, linear regressions or new algorithms. This new system was validated against the standard Dual-Energy X-Ray Absorptiometry system, using a sample of 40 adults with positive results. This solution is a valid and reliable alternative to traditional reference callipers, simplifying the percentage of body fat evaluation and providing a more effective use in daily practice with less expenditure of time and resources. Its implemented guided procedure turns it into a precious training tool based on a non-invasive, portable device, and not requiring special individual preparation. Ongoing activities are focused on the design of a new mechanical structure, with novel functionalities and for exploring other studies.