With the improvement of people’s cultural level, more and more museums are being built or renovated. The design of lighting products for museums is a specialized field that requires designers to take into account a variety of factors, such as safety, presentation, and maintainability. As museum lighting systems meet the needs of conservation, visitor experience, and maintenance, the traditional design process is limited by the experience of the designer and the actual situation of the museum, and the actual light conservation effect of the exhibits is difficult to quantify. We have designed a digital-twin-based intelligent lighting system for museums, which can facilitate museum managers to quantify and manage the light life of exhibits while providing a more immersive viewing experience and recognition effect for visitors.

Action recognition is an important research direction in computer vision, which has worldwide applications, such as video surveillance, human-robot interaction and so on. Due to the influence of complex background and multi-angle changes, accurate recognition and analysis of human motion in real-life scenarios is still a challenging problem. In order to improve the accuracy of pedestrian detection and motion recognition, this paper proposes a novel edge-aware end-to-end deep network method, which uses the edge-aware pooling module to improve pedestrian contour accuracy and captures video sequences using multi-scale pyramid pooling layer spatial-time context feature. The complementary features of the edge-related features can effectively preserve the clear boundary, and the combination of the auxiliary side output and the pyramid pooling layer output can extract rich global context information. A large number of qualitative and quantitative experimental results show that the proposed model can effectively improve the performance of existing pedestrian detection and motion recognition networks on the UCF-101, HMDB-51, and KTH dataset. INDEX TERMS Motion recognition, edge perception, deep learning, pyramid pooling, spatial-temporal context.

The paper presents a hybrid strategy in a soft computing paradigm for the optimization of the low-pressure die casting process. Casting process parameters, such as various parts temperatures of die, pouring temperature are considered. The hybrid strategy combines numerical simulation software, a genetic algorithm and a multilayer neural network to optimize the process parameters. An approximate analysis model is developed using a BP neural network in order to avoid the expensive computation resulting from the numerical simulation software. According to the characteristic of the optimization problem, a real-code genetic algorithm is applied to solve the optimization model. The effectiveness of the improved strategy is shown by an A356 automotive wheel.

No abstract

No abstract