Synthetic‐to‐realistic domain adaptation for cold‐start of rail inspection systems
نویسندگان
چکیده
Abstract Rail surface defects are potential danger factors for railway systems, and visual inspection of plays a vital role in rail maintenance. Recently, the methods based on deep learning have been widely used but such systems often face problem lack defect samples training models at start‐up, which is called cold‐start problem. It challenging to obtain sufficient since sparse even non‐existent running system. Therefore, synthetic‐to‐realistic domain adaptation (SRDA) method proposed real‐world inspection. SRDA adapts synthetic images look more realistic reducing gap between obtains translated that consist information background. After that, train detector defects. In order make robust complex backgrounds, generates with same defect‐level semantics different texture appearances makes align these learned feature space. addition, (SRRSD) dataset containing 20,662 built. The experimental results SRRSD show achieves higher detection performance than other established adaption 19.0% 26.7% average precision.
منابع مشابه
Sample-oriented Domain Adaptation for Image Classification
Image processing is a method to perform some operations on an image, in order to get an enhanced image or to extract some useful information from it. The conventional image processing algorithms cannot perform well in scenarios where the training images (source domain) that are used to learn the model have a different distribution with test images (target domain). Also, many real world applicat...
متن کاملRail Inspection of Rcf Defects
Z. Popović, L. Lazarević, University of Belgrade, Faculty of Civil Engineering, Belgrade, Serbia V. Radović, University of Montenegro, Faculty of Civil Engineering in Podgorica, Montenegro V. Vukadinović, Serbian Railways, Belgrade, Serbia G. Tepić, University of Novi Sad, Faculty of Technical Sciences, Novi Sad, Serbia Rail defects due to rolling contact fatigue (RCF) threaten the traffi c saf...
متن کاملAutomatic Laser Ultrasonics for Rail Inspection
This paper presents the state of the current activities to develop a high-speed inspection system for evaluating rail defects. Since conventional contact ultrasonic techniques are speed limited, the aim was to build an automatic inspection system, which applies non-contact ultrasonic methods. An automatic Laser Ultrasonic system for Rail Inspection (LURI) has been developed and tested on a rail...
متن کاملRail Inspection with Guided Waves
The critical subject of transverse crack detection in a rail head is treated in this paper. Conventional bulk wave ultrasonic techniques often fail because of shelling and other surface imperfections that shield the defects that lie below the shelling. A guided wave inspection technique is introduced here that can send ultrasonic energy along the rail under the shelling with a capability of fin...
متن کاملReliability-Based Optimization of Rail Inspection
This thesis proposes a quantitative method to optimize inspection/repair intervention in rail defect management. Rail defect management is important for track maintenance, since rails are the most significant and basic component of the track systems. Rail inspection is a fundamental intervention to prevent rail failure. Railroads have evolved the rail inspection interval based on their empirica...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Computer-aided Civil and Infrastructure Engineering
سال: 2023
ISSN: ['1093-9687', '1467-8667']
DOI: https://doi.org/10.1111/mice.13087