High-Speed Rail Tunnel Panoramic Inspection Image Recognition Technology Based on Improved YOLOv5

Duan, Yingying; Qiu, Su; Jin, Weiqi; Lu, Taoran; Li, Xingsheng

doi:10.3390/s23135986

Cited by 5 publications

(4 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…According to the binding information between the camera and the parking space recorded in the database, the point range of the hot spot area is obtained, and whether there is a vehicle parked in the hot spot area of the parking space is determined in the monitoring image of the camera. LPRnet is used to identify the license plate of parked vehicles and write the license plate information into the database for license plate parking space binding [37][38][39].…”

Section: Parking Vehicle Detection and Identification Modulementioning

confidence: 99%

Research on the Car Searching System in the Multi-Storey Garage with the RSSI Indoor Locating Based on Neural Network

Ma,

Wang,

Zhu

et al. 2024

Electronics

View full text Add to dashboard Cite

To solve the problem of reverse car searching in intelligent multi-story garages or parking lots, the reverse car searching method based on the intelligent garage of the PC client and mobile client APP was studied, and the interface design and function development of the system’s PC and mobile client APP were carried out. YOLOv5 network and LPRNet network were used for license plate location and recognition to realize parking and entry detection. The indoor pedestrian location method based on RSSI fingerprint signal fusion BPNet network and KNN algorithm was studied, and the location accuracy within 2.5 m was found to be 100%. The research on the A* algorithm based on spatial accessibility was conducted to realize the reverse car search function. The research results indicate that the guidance of the vehicle finding path can be completed while the number of invalid search nodes for the example maps was reduced by more than 55.0%, and the operating efficiency of the algorithm increased to 28.5%.

show abstract

Section: Parking Vehicle Detection and Identification Modulementioning

confidence: 99%

Research on the Car Searching System in the Multi-Storey Garage with the RSSI Indoor Locating Based on Neural Network

Ma,

Wang,

Zhu

et al. 2024

Electronics

View full text Add to dashboard Cite

show abstract

“…The YOLOv8 network [5,22], which debuted in 2023 and is the most recent iteration of the YOLO architecture series [4], does away with the bounding box operation that YOLOv5 [23][24][25] utilized. Its primary input is 640 × 640, and depending on the scaling factor, the network offers different size models with N, S, M, L, and X dimensions to accommodate the requirements of a wide variety of use cases.…”

Section: Yolov8s Network Architecturementioning

confidence: 99%

Small Target Detection in Refractive Panorama Surveillance Based on Improved YOLOv8

Zheng,

Zou,

et al. 2024

Sensors

View full text Add to dashboard Cite

Panoramic imaging is increasingly critical in UAVs and high-altitude surveillance applications. In addressing the challenges of detecting small targets within wide-area, high-resolution panoramic images, particularly issues concerning accuracy and real-time performance, we have proposed an improved lightweight network model based on YOLOv8. This model maintains the original detection speed, while enhancing precision, and reducing the model size and parameter count by 10.6% and 11.69%, respectively. It achieves a 2.9% increase in the overall mAP@0.5 and a 20% improvement in small target detection accuracy. Furthermore, to address the scarcity of reflective panoramic image training samples, we have introduced a panorama copy–paste data augmentation technique, significantly boosting the detection of small targets, with a 0.6% increase in the overall mAP@0.5 and a 21.3% rise in small target detection accuracy. By implementing an unfolding, cutting, and stitching process for panoramic images, we further enhanced the detection accuracy, evidenced by a 4.2% increase in the mAP@0.5 and a 12.3% decrease in the box loss value, validating the efficacy of our approach for detecting small targets in complex panoramic scenarios.

show abstract

“…The spatial points obtained from laser scanning are constructed into a spatial distribution image, and then data processing is carried out based on the point cloud. Image acquisition system and the detection algorithms of computer vision are mature, and they can save high-resolution tunnel infographics and quickly recognize tunnel diseases [8][9][10][11].…”

Section: Introductionmentioning

confidence: 99%

A subway tunnel image stitching method based on point cloud mapping relationships and high-resolution image

Xia,

Mao,

Wang

et al. 2024

Eng. Res. Express

View full text Add to dashboard Cite

The paper presents an intelligent tunnel 3D laser image acquisition system. The system can quickly acquire the 3D point cloud and tunnel surface image of the whole tunnel section. The paper proposes the subway tunnel image stitching method based on point cloud mapping relationships and high-resolution image. Firstly, the method extracts the homonymous points from the mapped image generated from the 3D point cloud and the camera image based on the maximum information coefficient method. Secondly, the 3D point cloud positions corresponding to the homonymous points are determined. The optimization of camera parameters is achieved based on the gradient descent method. Then, the tunnel point cloud is expanded to generate a gray scale image based on spherical projection. Based on the optimized parameters, the point cloud is projected to the camera image to determine the depth value of each camera image pixel point. Finally, the pixel values of the camera image are projected based on the spherical projection to obtain the stitching result images. This method stitches individual images together to form a more complete tunnel surface image. The larger tunnel surface image helps in identifying tunnel defects.

show abstract

High-Speed Rail Tunnel Panoramic Inspection Image Recognition Technology Based on Improved YOLOv5

Cited by 5 publications

References 30 publications

Research on the Car Searching System in the Multi-Storey Garage with the RSSI Indoor Locating Based on Neural Network

Research on the Car Searching System in the Multi-Storey Garage with the RSSI Indoor Locating Based on Neural Network

Small Target Detection in Refractive Panorama Surveillance Based on Improved YOLOv8

A subway tunnel image stitching method based on point cloud mapping relationships and high-resolution image

Contact Info

Product

Resources

About