6 November 2024 in Communications in Transportation Research
22 June 2022 in IEEE Transactions on Instrumentation and Measurement
Studying automated rail surface defect detection methods instead of manual inspections can not only save time and cost, but also increase the safety of railway transport. Consequently, many methods have been developed, which utilize convolutional neural networks (CNNs) in order to automatically detect rail surface defects. In these works, however, little attention has been paid to the detection of rail surface crack defects. The rail surface crack defect is actually an early defect that, if not detected and maintained in time, can result in worse defects. In order to detect crack defects on the rail surface rapidly and accurately, we propose an effective deep multitask learning architecture. Our architecture contains not only a segmentation decoder in order to segment the crack defect region on the rail, but also an object detection decoder to detect rail objects. During the training stage, as an auxiliary task, the rail object detection can transfer its learned knowledge to the rail crack region segmentation task, so as to improve the accuracy of the model. During the testing phase, the rail object detection task does not require inference, which allows the model to have a fast inference speed. Our model has been validated using the challenging Rail-5k dataset, achieving a balance between accuracy and speed.
10 February 2022 in IEEE Transactions on Circuits and Systems for Video Technology
Co-salient object detection (CoSOD) aims to detect common salient objects sharing the same attributes in an image group. The key issue of CoSOD is how to model the inter-saliency relations within an image group. The major limitation of previous methods is that they pre-define the group-to-one relations within an image group. In this paper, we propose a new concept of structural inter-saliency relations and solve the CoSOD with deep reinforcement learning framework. Firstly, we design a semantic relation graph (SRG) to model the structural inter-saliency relations. Then the feature selecting agent (FS-agent) aims to select the informative features, which can help the SRG effectively model structural inter-saliency relations. Finally, relation updating agent (RU-agent) progressively updates the SRG to focus on the co-salient relations like human decision-making process. Extensive experiments on co-saliency datasets show that because of well modeling inter-saliency relations in image group, our proposed method achieves superior performance compared to the state-of-the-art methods. We hope that this paper can motivate future research for visual co-analysis tasks.
13 October 2022 in IEEE Transactions on Image Processing
In this paper, we present a novel model for simultaneous stable co-saliency detection (CoSOD) and object co-segmentation (CoSEG). To detect co-saliency (segmentation) accurately, the core problem is to well model inter-image relations between an image group. Some methods design sophisticated modules, such as recurrent neural network (RNN), to address this problem. However, order-sensitive problem is the major drawback of RNN, which heavily affects the stability of proposed CoSOD (CoSEG) model. In this paper, inspired by RNN-based model, we first propose a multi-path stable recurrent unit (MSRU), containing dummy orders mechanisms (DOM) and recurrent unit (RU). Our proposed MSRU not only helps CoSOD (CoSEG) model captures robust inter-image relations, but also reduces order-sensitivity, resulting in a more stable inference and training process. Moreover, we design a cross-order contrastive loss (COCL) that can further address order-sensitive problem by pulling close the feature embedding generated from different input orders. We validate our model on five widely used CoSOD datasets (CoCA, CoSOD3k, Cosal2015, iCoseg and MSRC), and three widely used datasets (Internet, iCoseg and PASCAL-VOC) for object co-segmentation, the performance demonstrates the superiority of the proposed approach as compared to the state-of-the-art (SOTA) methods.
8 September 2025 in Transportation Research Part C: Emerging Technologies
3 June 2024 in IEEE Transactions on Intelligent Vehicles
Crack detection is essential for ensuring road safety and preserving the integrity of infrastructure. As intelligent transportation systems mature, vehicles equipped with crack detection capabilities have become more sophisticated. Traditional methods for detecting cracks have relied on designing complex networks, which can be time-consuming and may obscure the task's underlying simplicity, posing challenges for researchers and hindering integration into road-perceptive vehicles. This paper introduces an innovative, universal framework for crack detection that circumvents these issues by utilizing existing pre-trained segmentation methods with visual prompts. We introduce the visual crack prompt (VCP) mechanism, which refines the focus of pre-trained models on high-frequency features, significantly improving their ability to identify and segment specific crack features. Additionally, we present the diverse crack detection 1 K dataset (DCD1K), comprising 1000 images of 16 different crack types, to validate the VCP mechanism's effectiveness. Our experimental results showcase the framework's outstanding performance across six distinct datasets, highlighting its potential to revolutionize crack detection methods.