Fourier Ptychographic Microscopy (FPM) is a kind of micro-computational imaging technique that simultaneously realizes high resolution and a large field of view. To address the problems of artifacts and incomplete information extraction in the reconstructed area features of FPM, we propose a reconstruction method based on the Window Self-attention Network (WSAN) for Fourier Ptychographic Microscopy. Window Self-attention Network (WSAN). The application of multi-head self-attention mechanism in WSAN enables the network to focus its attention on the critical regions in the frequency spectrum of the input image, effectively capturing the local details and structural information of automotive micro-sensors.
The effects of smoke spread, decrease in visibility and uneven temperature distribution on evacuation safety during a fire are thoroughly investigated with respect to the characteristics of public building fires and the theory of safe evacuation. The fire model is established through BIM technology and its application and optimization methods in building models. Pyrosim software was used and data simulation analysis was performed based on the BIM model. Emphasis was placed on the effects of factors such as smoke spread, visibility and temperature distribution on the fire spread and evacuation process. In addition, the evacuation process was simulated using Pathfinder software in conjunction with the BIM model. The parameters affecting the evacuation efficiency were investigated with the aim of finding the best evacuation strategy. Finally, a fire evacuation path planning optimization scheme based on improved ant colony algorithm was proposed. By combining the ant colony algorithm with evacuation path planning, the optimal evacuation path can be found in complex building environments and the evacuation efficiency can be improved.
KEYWORDS: Detection and tracking algorithms, Convolution, Feature extraction, Education and training, Video, Mobile devices, Deep learning, Neural networks
Aiming at the problem of slow running speed of pedestrian single-target tracking task using SiamBAN algorithm under limited hardware conditions, this paper proposes a lightweight pedestrian single-target tracking network GhostSiamBAN. Specifically, we use GhostModule to build a new lightweight backbone network to replace the original Resnet50. The evaluation on OTB2015 and VOT2018 shows that our method can well balance the tracking accuracy and speed. Compared with SiamBAN, the test results of pedestrian single target tracking sequence on Nvidia GTX 1050 show that the average running speed of our method is about 26fps with an accuracy loss of about 5.46%, and the speed is increased by nearly 271.4%
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.