Publications by authors named "Dequan Zhao"

Article Synopsis
  • An improved A* algorithm is developed to enhance off-road emergency rescue path planning by considering unpaved roads and road factors for better efficiency.
  • The algorithm classifies nodes based on their angles to optimize search directions, leading to quicker identification of the shortest travel time for emergency rescues in wilderness settings.
  • Experimental results show this improved A* algorithm decreases travel time for off-road vehicles by over 21% and boosts search efficiency by nearly 94% compared to the traditional A* algorithm.
View Article and Find Full Text PDF

Current road extraction models from remote sensing images based on deep learning are computationally demanding and memory-intensive because of their high model complexity, making them impractical for mobile devices. This study aimed to develop a lightweight and accurate road extraction model, called Road-MobileSeg, to address the problem of automatically extracting roads from remote sensing images on mobile devices. The Road-MobileFormer was designed as the backbone structure of Road-MobileSeg.

View Article and Find Full Text PDF
Article Synopsis
  • The wear rates of the top compression piston-ring and cylinder-bore in a ZS1105 diesel engine were measured using a technique called thin layer activation (TLA).
  • The study details the setup and calibration methods for the testing specimens used.
  • The minimum detectable wear depths measured were 9 nm for the cylinder-bore and 75 nm for the piston-ring, indicating that in situ techniques can effectively assist in engine development despite potential localized wear variations.
View Article and Find Full Text PDF