Linlin Wang

Dr. Linlin Wang

orcid

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

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

Representative Publications

View All Publications
8 December 2022 in Applied Sciences
Verified
Journal Article
Open access
PDF file
LWLinlin Wang
Linlin Wang
JLJunjie Li
Junjie Li
Fei KangFei Kang
Fei Kang
Damage detection and evaluation are concerns in structural health monitoring. Traditional damage detection techniques are inefficient because of the need for damage detection before evaluation. To address these problems, a novel crack location and degree detector based on YOLOX is proposed, which directly realizes damage detection and evaluation. Moreover, the detector presents a superior detection effect and speed to other advanced deep learning models. Additionally, rather than at the pixel level, the detection results are determined in actual scales according to resolution. The results demonstrate that the proposed model can detect and evaluate damage accurately and automatically.

1 November 2021 in Journal of Computing in Civil Engineering
Verified
Journal Article
Closed access
PDF file
LWLinlin Wang
Linlin Wang
JLJunjie Li
Junjie Li
Unmanned aerial vehicles (UAVs) have been widely used in the visual inspection of structural cracks. However, blurry images are inevitably generated during image collecting by UAVs, as they are caused by the motion of UAVs and other factors. This blur affects the retrieval of crack properties from images and degrades the accuracy and reliability of crack damage assessment. At present, blur detection and blurred image removal are mainly achieved manually, which is inefficient and fallible, especially for large image sets. To address this problem, a novel automatic blur detection method for UAV crack image data sets is proposed. This algorithm defines a blur detection metric named the edge average width difference (EAWD), which is based on the principle of a smaller difference between pixels of a more blurred image. Moreover, it is combined with the characteristics of the crack image itself. By calculating this metric and comparing it with other EAWD values from the same data set, the crack images are judged to be blurred or not. Furthermore, a support vector machine classifier is applied to the aforementioned metrics, serving as the image blur quality evaluator. For proper training and assessment of the proposed approach, an image data set consisting of 1,200 crack images is created, which also contains some thin crack images. Several experimental results are provided in this paper to demonstrate that the proposed method is fast, accurate, and reliable.

18 March 2020
Verified
Conference Paper
Open access
PDF file
LWLinlin Wang
Linlin Wang
JZJunyou Zhang
Junyou Zhang
YFYuan Feng
Yuan Feng
Intersections are the places where traffic accidents occur frequently. As the hub of urban transportation, ensuring the efficient and safe operation of vehicles has become a problem that needs to be solved. The research on the mechanism of the severity of traffic accident injury at intersections can provide an important basis for traffic managers or autonomous vehicles to make driving decisions under dangerous conditions. The K-means clustering is used to analyze the traffic accident data, and the vehicle-road-environment is taken as the internal cause variable to characterize the severity of the collision injury, and the clustering results are used to classify the severity of the injury. The severity of injury can be divided into mild injury, moderate injury and severe injury. The clustering results based on the severity of the injury show that the four most significant factors are the type of intersection, the type of collision vehicles, the relative speed and the signal control of the intersection. The results of the study put forward policy suggestions for the planning and management of urban intersections from many angles.

19 April 2024
Verified
Conference Paper
Open access
LWLinlin Wang
Linlin Wang
JLJunjie Li
Junjie Li
The complex apparent environment of concrete dams causes crack detection difficult. To address the problems of low precision of crack detection by traditional methods, this paper proposes a concrete dam apparent crack detection algorithm based on the improved DeepLabv3+ model, which provides more accurate and intuitive detection results for concrete dam health monitoring. First, a self-made concrete dam crack dataset is constructed for model training and testing. Then, the optimal initial learning rate is selected by comparing and analyzing in five different learning rates. Finally, on the test set, the model proposed in this paper performs well, with high IOU and PA. It can conclude that the proposed method efficiently and accurately detects the concrete dam cracks in complex backgrounds, providing effective technical support for the long-term monitoring of concrete dams.

29 August 2026 in Structures
Verified
Journal Article
Closed access
LWLinlin Wang
Linlin Wang
BBBoya Bu
Boya Bu
DZDezhi Zai
Dezhi Zai
LMLiang Meng
Liang Meng
JLJunjie Li
Junjie Li

1 November 2026 in Natural Hazards Review
Verified
Journal Article
Closed access
PDF file
GLGuoyang Liu
Guoyang Liu
JMJianbin Miao
Jianbin Miao
LWLinlin Wang
Linlin Wang
DLDefang Liu
Defang Liu
WBWu Bo
Wu Bo
SWShuonan Wang
Shuonan Wang
Landslides are common geological disasters that cause extensive damage and significant harm. Efficient recognition of landslide remote sensing images holds both theoretical and practical importance. However, practical landslide detection faces challenges including limited data, high computation, large memory use, and low accuracy. In this study, high-quality landslide remote sensing images were collected from southeastern Xizang to construct a landslide remote sensing data set. To address the issue of insufficient samples, an improved model, landslide-deep convolutional generative adversarial network (L-DCGAN), based on DCGAN, was proposed to enhance landslide remote sensing data. The enhanced data set was constructed by incorporating efficient channel attention (ECA) and squeeze-and-excitation (SE) modules into the generator and discriminator, respectively. Additionally, an improved YOLOv8 model, landslide-you only look once version 8(L-YOLOv8), was proposed. First, the cross-stage partial pyramid convolution (CSPPC) and grouped and shuffled convolution (GSConv) modules replaced the cross stage partial layer with 2 branches (C2f) module and ordinary convolution in the network to reduce the number of model parameters and computational complexity. Then, the spatial and channel synergistic attention (SCSA) mechanism was introduced to improve detection accuracy. Experimental results showed that compared with the original YOLOv8 algorithm, the number of model parameters in L-YOLOv8 was reduced by approximately 41.78%, and the average accuracy on the landslide sample data set was increased by 0.57%, and frames per second (FPS) was improved by 33.87%. The model achieved lightweight design while improving detection speed and accuracy, and demonstrated superior performance on the landslide remote sensing data set.