Songrong Qian

Dr. Songrong Qian

orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

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

Representative Publications

View All Publications
30 September 2022 in Complex & Intelligent Systems
Verified
Journal Article
Open access
PDF file
JZJian Zhang
Jian Zhang
SQSongrong Qian
Songrong Qian
CTCan Tan
Can Tan
Deep learning-based bridge crack detection methods have advantages over traditional methods. We proposed an automated bridge crack detection method using lightweight vision models. First, our study applied the You Only Look Once 4th version (YOLO v4) (Bochkovskiy et al. in Yolov4: Optimal speed and accuracy of object detection. arXiv:200410934, 2020) to bridge surface crack detection. Then, to achieve model acceleration, some lightweight networks were used to replace the feature extraction network in YOLO v4, which reduced the parameter numbers and the backbone layers. The lightweight design can reduce the computational overhead of the model, making it convenient to deploy on edge platforms with limited computational power. The experimental results showed that the lightweight network-based bridge crack detection model required significantly less storage space at the expense of a slight reduction in precision. Therefore, an improved YOLO v4 crack detection method was proposed to meet real-time running without sacrificing accuracy. The precision, recall, and F1 score of the proposed crack detection method are 93.96%, 90.12%, and 92%, respectively. And the model only required 23.4 MB of storage space, and its frames per second could reach 140.2 frames. Compared with existing bridge crack detection methods, the proposed method showed precision, speed, and model size advantages.

13 August 2023 in Applied Sciences
Verified
Journal Article
Open access
PDF file
XZXin Zheng
Xin Zheng
SQSongrong Qian
Songrong Qian
SWShaodong Wei
Shaodong Wei
SZShiyun Zhou
Shiyun Zhou
YHYi Hou
Yi Hou
The real-time detection of cracks is an important part of road maintenance and an important initiative to reduce traffic accidents caused by road cracks. In response to the lack of efficiency of current research results for the real-time detection of road cracks and the low storage and computational capacity of edge devices, a new automatic crack detection algorithm is proposed: BT–YOLO. We combined Bottleneck Transformer with You Only Look Once (YOLO), which is more conducive to extracting the features of small cracks than YOLOv5s. The introduction of DWConv to the feature extraction network reduced the number of parameters and improved the inference speed of the network. We embedded the SimAM (Simple, Parameter-Free Attention Module) non-parametric attention mechanism to make the crack features more prominent. The experimental results showed that the accuracy of BT–YOLO in crack detection was increased by 4.5%, the mapped value was increased by 8%, and the parameter amount was decreased by 24.9%. Eventually, we deployed edge devices for testing. The frame rate reached 89, which satisfied the requirements of real-time crack detection.

22 July 2024 in IET Image Processing
Verified
Journal Article
Open access
MYMing Yang
Ming Yang
SQSongrong Qian
Songrong Qian
XWXiaoqin Wu
Xiaoqin Wu
Recent years have seen increased interest in object detection‐based applications for fire detection in digital images and videos from edge devices. The environment's complexity and variability often lead to interference from factors such as fire and smoke characteristics, background noise, and camera settings like angle, sharpness, and exposure, which hampers the effectiveness of fire detection applications. Limited picture data for fire and smoke scenes further challenges model accuracy and robustness, resulting in high false detection and leakage rates. To address the need for efficient detection and adaptability to various environments, this paper focuses on (1) proposing a cloud‐edge collaborative architecture for real‐time fire and smoke detection, incorporating an iterative transfer learning strategy based on user feedback to enhance adaptability; (2) improving the detection capabilities of the base model YOLOv8 by enhancing the data augmentation method and introducing the coordinate attention mechanism to improve global feature extraction. The improved algorithm shows a 2‐point accuracy increase. After three iterations of transfer learning in the production environment, accuracy improves from 93.3% to 96.4%, and mAP0.5:0.95 increases by nearly 5 points. This program effectively addresses false detection issues in fire and smoke detection systems, demonstrating practical applicability.

30 November 2024 in Applied Sciences
Verified
Journal Article
Open access
PDF file
CCCheng Chen
Cheng Chen
JYJiantao Yu
Jiantao Yu
SQSongrong Qian
Songrong Qian
Path planning is a key task in mobile robots, and the application of Deep Q Network (DQN) algorithm for mobile robot path planning has become a hotspot and challenge in current research. In order to solve the obstacle avoidance limitations faced by the DQN algorithm in indoor robot path planning, this paper proposes a solution based on an improved DQN algorithm. In view of the low learning efficiency of the DQN algorithm, the Duel DQN structure is introduced to enhance the performance and combined with a Prioritized Experience Replay (PER) mechanism to ensure the stability of the robot during the learning process. In addition, the idea of Munchausen Deep Q Network (M-DQN) is incorporated to guide the robot to learn the optimal policy more effectively. Based on the above improvements, the PER-D2MQN algorithm is proposed in this paper. In order to validate the effectiveness of the proposed algorithm, we conducted multidimensional simulation comparison experiments of the PER-D2MQN algorithm with DQN, Duel DQN, and the existing methodology PMR-DQN in the Gazebo simulation environment and examined the cumulative and average rewards for reaching the goal point, the number of convergent execution steps, and the time consumed by the robot in reaching the goal point. The simulation results show that the PER-D2MQN algorithm obtains the highest reward in both static and complex environments, exhibits the best convergence, and finds the goal point with the lowest average number of steps and the shortest elapsed time.

1 January 2023 in Computers, Materials & Continua
Verified
Journal Article
Open access
PDF file
XWXiaoqin Wu
Xiaoqin Wu
SQSongrong Qian
Songrong Qian
MYMing Yang
Ming Yang
Safety helmets can reduce head injuries from object impacts and lower the probability of safety accidents, as well as being of great significance to construction safety. However, for a variety of reasons, construction workers nowadays may not strictly enforce the rules of wearing safety helmets. In order to strengthen the safety of construction site, the traditional practice is to manage it through methods such as regular inspections by safety officers, but the cost is high and the effect is poor. With the popularization and application of construction site video monitoring, manual video monitoring has been realized for management, but the monitors need to be on duty at all times, and thus are prone to negligence. Therefore, this study establishes a lightweight model YOLO_CA based on YOLOv5 for the automatic detection of construction workers’ helmet wearing, which overcomes the shortcomings of the current manual monitoring methods that are inefficient and expensive. The coordinate attention (CA) addition to the YOLOv5 backbone strengthens detection accuracy in complex scenes by extracting critical information and suppressing non-critical information. Further parameter compression with deeply separable convolution (DWConv). In addition, to improve the feature representation speed, we swap out C3 with a Ghost module, which decreases the floating-point operations needed for feature channel fusion, and CIOU_Loss was substituted with EIOU_Loss to enhance the algorithm’s localization accuracy. Therefore, the original model needs to be improved so as to enhance the detection of safety helmets. The experimental results show that the YOLO_CA model achieves good results in all indicators compared with the mainstream model. Compared with the original model, the mAP value of the optimized model increased by 1.13%, GFLOPs cut down by 17.5%, and there is a 6.84% decrease in the total model parameters, furthermore, the weight size cuts down by 4.26%, FPS increased by 39.58%, and the detection effect and model size of this model can meet the requirements of lightweight embedding.

8 December 2021 in Applied Artificial Intelligence
Verified
Journal Article
Closed access
ZXZhengyun Xu
Zhengyun Xu
SQSongrong Qian
Songrong Qian
XRXiu Ran
Xiu Ran
JZJi Zhou
Ji Zhou
The surface crack of structure is an important sign to evaluate the safety of structure. In order to ensure the safety and reliability of the building structure, it is necessary to detect and monitor the surface cracks of the structure. Traditional artificial surface inspections are time-consuming because inspectors have different experience and knowledge, which can lead to misjudgments. Based on the basic framework of four deep convolution neural networks, their classifiers are reconstructed. To fully train these networks and simulate crack images taken in various situations in life, image enhancement techniques are used to extend the dataset. After training, compared with the established shallow network structure, they can learn the feature information in the image more fully, and finally improve the accuracy. After further verification, it is found that one of the models can achieve an accuracy of 96.5%. To verify the universality and validity of the model, two cross-datasets experiments were performed. The experimental results show the validity of the model, and the diagnostic precision is 98.23% and 99.04%, respectively.