Publications by authors named "Zhizhong Kang"
Article Synopsis
- A retrofit involving an oxygen-rich burner with liquefied natural gas (LNG) was introduced to improve unstable combustion and reduce NOx emissions from a 330 MW subcritical boiler operating under low load conditions in China.
- The effectiveness of this retrofit was analyzed using Chemkin and Fluent software to create a NOx generation model and simulate combustion reactions.
- Results demonstrated that after the retrofit, combustion stability improved at loads above 50%, and NOx emissions decreased from 380 mg/m to 316 mg/m, especially notable at low load conditions due to enhanced combustion from LNG.
CMANet: Cross-Modality Attention Network for Indoor-Scene Semantic Segmentation.
Sensors (Basel)
November 2022
Indoor-scene semantic segmentation is of great significance to indoor navigation, high-precision map creation, route planning, etc. However, incorporating RGB and HHA images for indoor-scene semantic segmentation is a promising yet challenging task, due to the diversity of textures and structures and the disparity of multi-modality in physical significance. In this paper, we propose a Cross-Modality Attention Network (CMANet) that facilitates the extraction of both RGB and HHA features and enhances the cross-modality feature integration.
View Article and Find Full Text PDFBecause tunnels generally have tubular shapes, the distribution of tie points between adjacent scans is usually limited to a narrow region, which makes the problem of registration error accumulation inevitable. In this paper, a global registration method is proposed based on an augmented extended Kalman filter and a central-axis constraint. The point cloud registration is regarded as a stochastic system, and the global registration is considered to be a process that recursively estimates the rigid transformation parameters between each pair of adjacent scans.
View Article and Find Full Text PDFAlthough RANSAC is proven to be robust, the original RANSAC algorithm selects hypothesis sets at random, generating numerous iterations and high computational costs because many hypothesis sets are contaminated with outliers. This paper presents a conditional sampling method, multiBaySAC (Bayes SAmple Consensus), that fuses the BaySAC algorithm with candidate model parameters statistical testing for unorganized 3D point clouds to fit multiple primitives. This paper first presents a statistical testing algorithm for a candidate model parameter histogram to detect potential primitives.
View Article and Find Full Text PDFEfficient simplification of large vector maps rendered onto 3D landscapes.
IEEE Comput Graph Appl
October 2014
Article Synopsis
- The text discusses the challenge of rendering large-scale vector maps over terrain, which involves handling many polylines and polygons.
- It introduces a proposed method that simplifies these maps to improve rendering efficiency.
- This simplification helps reduce latency, making the display and interaction within a virtual environment smoother.
Article Synopsis
- This paper introduces a method for automatically registering terrestrial laser scanning (TLS) point clouds by using panoramic reflectance images.
- The process involves two main steps: pair-wise registration, which matches images and point clouds, and global registration to align all point clouds using a bundle adjustment technique.
- The results demonstrate that the method achieves high accuracy, with pair-wise registration accurate to within millimeters and global registration within centimeters, and it operates quickly and fully automatically.