Kaili Li

Kaili Li

orcid

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 January 2022 in Open Geosciences
Verified
Journal Article
Open access
PDF file
YSYangen Shen
Yangen Shen
ALAnbo Li
Anbo Li
JHJian-Chu Huang
Jian-Chu Huang
GLGuo-Nian Lü
Guo-Nian Lü
KLKaili Li
Kaili Li
Loose layers are the locus of human activities. The high-quality 3D modeling of loose layers has essential research significance and applicability in engineering geology, hydraulic and hydroelectric engineering, and urban underground space design. To address the shortcomings of traditional 3D loose-layer modeling based on borehole data, such as the lack of bedrock surface constraints, simple strata pinch-out processing, and the higher fitting error of the strata surface, a 3D loose-layer modeling method based on the stratum development law is proposed. The method mainly uses three different virtual boreholes, bedrock-boundary virtual boreholes, pinch-out virtual boreholes, and densified virtual boreholes, to control the stratigraphic distribution. Case studies demonstrate the effectiveness of this 3D loose-layer modeling method in the Qinhuai District of Nanjing and Hangkonggang District of Zhengzhou. Compared to the previous methods that interpolated stratigraphic surfaces with elevation information, the method proposed in this article interpolates the stratum thickness based on stacking, which could improve the interpolation accuracy. In the area where the loose layers and exposed bedrock are alternately distributed, stratigraphic thickness errors’ mean and standard deviation decreased by 2.11 and 2.13 m. In the pure loose-layer area, they dropped by 0.96 and 0.33 m. In addition, the proposed approach allows us to infer the different stratigraphic distribution patterns accurately and complete 3D loose-layer model construction with higher accuracy and a good visualization effect.

29 October 2023 in Electronics
Verified
Journal Article
Open access
PDF file
WHWei Huang
Wei Huang
KLKaili Li
Kaili Li
MXMengfan Xu
Mengfan Xu
RHRui Huang
Rui Huang
Low-light image enhancement is a challenging task in non-uniform low-light conditions, often resulting in local overexposure, noise amplification, and color distortion. To obtain satisfactory enhancement results, most models must resort to carefully selected paired or multi-exposure data sets. In this paper, we propose a self-supervised framework for non-uniform low-light image enhancement to address these issues, only requiring low-light images on their own for training. We first design a robust Retinex model-based image exposure enhancement network (EENet) to obtain global brightness enhancement and noise removal of images by carefully designing the loss function of each decomposition map. Then, to correct overexposed areas in the enhanced image, we incorporate the inverse image of the low-light image for enhancement using EENet. Furthermore, a three-branch asymmetric exposure fusion network (TAFNet) is designed. The two enhanced images and the original image are used as the TAFNet inputs to obtain a globally well-exposed and detail-rich image. Experimental results demonstrate that our framework outperforms some state-of-the-art methods in visual and quantitative comparisons.