LJLin Jiang

Lin Jiang

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Lin Jiang has a total of 45 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • An Adaptive Sensor for Detecting Hybrid Defects on Steel Plate With Unknown Thickness

    by Hang Xu, Jinhai Liu, Lin Jiang, Jiayue Sun, Huaguang Zhang, Qiangxin Li

    Hybrid defects of the oil and gas pipelines with unknown thickness occur frequently influenced by the complex working environment. Due to the mutual interference between hybrid-surface and hybrid-back defects and the lack of thickness information, it is difficult to accurately detect hybrid defects with unknown wall thickness. The integrated method has been widely developed for detecting hybrid defects due to its high detection efficiency and integrity. However, existing methods are unable to accurately detect hybrid defects with unknown wall thickness due to the difficulty in detecting completely covered hybrid-back defect and the nonadaptation to unknown thickness perfectly. To the above issues, an adaptive sensor is proposed, which does not require hybrid-surface and hybrid-back defects to be staggered or the known thickness in advance. First, an integrated structure is proposed by sharing magnetic source to synchronously collect multiple signals, which extends the information dimension without changing the excitation. Second, a circuitous strategy is proposed by energy detour to overcome the difficulty in detecting completely covered hybrid-back defect. Third, an adaptive strategy is proposed by matching the best frequency for unknown thicknesses. Finally, the experimental results indicate that the proposed sensor can accurately detect hybrid defects with unknown wall thickness.

  • Open Access

    ERSGNet: An Edge-Guided Residual Spatial Gating Fusion Network for Real-Time Semantic Segmentation

    by Zheming Wu, Lin Jiang, Tete Wang

    With the development of autonomous driving and general real-time vision applications, semantic segmentation is increasingly expected to provide reliable pixel-level scene understanding under strict latency and computational constraints. Existing real-time semantic segmentation methods still struggle to effectively balance contextual semantics and local details during feature fusion. This difficulty may lead to boundary blur and the loss of fine-grained structures, and is mainly related to insufficient selectivity and boundary friendliness in the fusion process. To address this issue, the Edge-Guided Residual Spatial Gating Fusion Network (ERSGNet)is proposed as a three-branch decoupled framework for real-time semantic segmentation. In the ERSGNet, detail features Fd, contextual features Fc, and edge features Fe are separately extracted to reduce mutual interference among different types of information. Spatially Gated Interaction (SGI) is introduced to inject contextual semantics into the detail branch in a pixel-wise selective manner, the Parallel Context Pyramid (PCP) is used to enhance multi-scale contextual representation, and Edge-Guided Residual Spatial Gating Fusion (ERSG-Fuse) performs edge-guided adaptive fusion between Fd and Fc under the guidance of Fe. The ERSGNet is evaluated on the CamVid and Cityscapes datasets. On CamVid, the ERSGNet achieves 73.33% mean Intersection-over-Union (mIoU) at 166.27 frames per second (FPS). On Cityscapes, the ERSGNet achieves 69.21% mIoU at 53.93 FPS. These results suggest that the ERSGNet achieves a favorable accuracy–efficiency trade-off while supporting boundary-related and fine-grained structural representation.

  • Path Planning for Underground Pipeline Inspection Based on ABC-RRT* in Constrained Environments

    by Jiarui Liu, Jinhai Liu, Lin Jiang, Zhenning Wu, Yanhong Luo

    For ground-based autonomous mobile robots inspecting underground pipeline, path planning critically impacts task performance and operational safety. However, current planners struggle with the stringent demands for time efficiency, path quality, obstacle avoidance, and pipeline proximity in this context. To address these challenges, we propose a sampling-based planner termed adaptive backtracking contraction-RRT*. First, we propose a pipeline position-biased sampling strategy, which introduces Gaussian distribution based on the pipeline localization model to ensure stable sensor coverage of the pipeline. Second, we propose a bidirectional dynamic ring guide mechanism that adaptively boosts sample density near obstacle boundaries, enhancing the planner’s real-time performance and adaptability in constrained environments. Third, a comprehensive path optimization method is proposed to improve overall path quality. By deeply optimizing forward tree exploration, reverse tree connection, and the complete path, it effectively enhances the planner’s convergence. Finally, we introduce an obstacle-aware spline parameterization (OASP) method to reduce the time cost caused by failed path smoothing. By leveraging real-time collision checks, OASP dynamically adjusts control points along the path, avoiding path re-planning and ensuring curvature-continuous paths. Simulations and real-world experiments demonstrate that our planner improves task execution efficiency, inspection success rate, planning time cost, and path quality.

  • Stray current inversion method based on sparrow search algorithm with kernel extreme learning machine

    by Lin Jiang, Zhishun Zhao, Jianhua Tang, Jinkun Zhao, HongYang Ding, Jinhai Liu, Hang Xu

  • A Robust Defect Detection Method for Vision-Based Measurement System via Wavelet-Guided Spatial–Frequency Prior Network

    by Zhitao Wen, Jinhai Liu, Jianhua Tang, He Zhao, Lin Jiang, Qiannan Wang

    Robust defect detection is a crucial link in vision-based measurement systems. However, noise and lighting in complex industrial environments can weaken the structural and edge information of defects, leading to unsatisfactory measurement accuracy. Recent studies have employed discrete wavelet transform (DWT) to enhance defect features using multi-resolution frequency subbands. However, they exhibit a bias toward frequency branches that compromise spatial cues, resulting in inadequate feature enhancement. To address these challenges, this study considers enhancing measurement features from a joint spatial–frequency perspective and proposes a wavelet-guided spatial-frequency prior network (WSP-Net). At the structural level, a dynamic prior excavator and a dual-domain detail aggregator are constructed to adaptively mine and aggregate discriminative spatial-frequency defect priors. At the decision level, a frequency-assisted task-aware head (FT-Head) is customized to further enhance task-specific semantics and edges through decoupling DWT. In addition, joint supervision is introduced to constrain fine regression for complex defects with ambiguous boundaries. Extensive experiments demonstrate that the proposed method achieves state-of-the-art (SOTA) performance across various vision-based measurement systems (e.g., PCB: 95.7% mAP; steel plates: 82.1% mAP; pipelines: 98.7% mAP), while maintaining low computational cost (Parameters: 9.1M, FLOPs: 12.8G FPS: 235-284 imgs/s). It also achieves remarkable performance under strong noise and low-light conditions. Furthermore, generalization analysis using real-world pipeline data from MFL-based measurements indicates that the method holds strong potential for engineering applications in real-world measurement systems.


  • An Integrated Method for Fully Covered Back Crack Defect Detection on Steel Plate Under Variable Lift Off

    by Hang Xu, Jinhai Liu, Jiayue Sun, Huaguang Zhang, Lin Jiang, Qiangxin Li

    Crack and corrosion defects commonly occur on pipeline walls under stress and erosion, with back cracks often fully covered by surface corrosion due to size differences. The detection of the covered back crack defects is challenging under variable lift off because of signal shielding and interference, increasing pipeline fracture risks. To address this, integrated methods with lift off compensation are widely used to back crack defect detection under variable lift off. However, current methods fail to accurately detect fully covered back crack defects under variable lift off, as they cannot break through surface corrosion coverage or meet the timeliness of lift off compensation. To solve the above problems, an integrated method is proposed to detect the fully covered back crack defects under variable lift off. First, a mutually supervised bypass strategy is proposed to detect back crack defects by bypassing the coverage of surface corrosion defects and to prevent missed or false detections. Then, an integrated structure of two EMATs and one ECT is proposed to synchronously acquire defect and lift off information by sharing excitation, which ensures the timeliness of lift off compensation. Next, based on the ECT, EMAT1, and EMAT2 signals, a quantification method using the lift off compensation element is proposed to quantify the depth of back crack defects under variable lift off. Finally, the experimental results indicate that the proposed integrated method can accurately detect the fully covered back crack defects under variable lift off.


  • Open Access

    Multiscale Interaction Purification-Based Global Context Network for Industrial Process Fault Diagnosis

    by Yukun Huang, Jianchang Liu, Peng Xu, Lin Jiang, Xiaoyu Sun, Haotian Tang

    The application of deep convolutional neural networks (CNNs) has gained popularity in the field of industrial process fault diagnosis. However, conventional CNNs primarily extract local features through convolution operations and have limited receptive fields. This leads to insufficient feature expression, as CNNs neglect the temporal correlations in industrial process data, ultimately resulting in lower diagnostic performance. To address this issue, a multiscale interaction purification-based global context network (MIPGC-Net) is proposed. First, we propose a multiscale feature interaction refinement (MFIR) module. The module aims to extract multiscale features enriched with combined information through feature interaction while refining feature representations by employing the efficient channel attention mechanism. Next, we develop a wide temporal dependency feature extraction sub-network (WTD) by integrating the MFIR module with the global context network. This sub-network can capture the temporal correlation information from the input, enhancing the comprehensive perception of global information. Finally, MIPGC-Net is constructed by stacking multiple WTD sub-networks to perform fault diagnosis in industrial processes, effectively capturing both local and global information. The proposed method is validated on both the Tennessee Eastman and the Continuous Stirred-Tank Reactor processes, confirming its effectiveness.

  • Double Coil Compensation Method for Reducing Distortion of High-Speed MFL Signals

    by Jinhai Liu, Hang Xu, Lin Jiang, Huaguang Zhang, Jiarui Liu

    Magnetic flux leakage (MFL) testing is widely used in the health evaluation of oil pipelines due to its non-destructive and efficient characteristics. However, motion-induced eddy currents (MIECs), which will cause distortion of the high-speed MFL signals, are generated in pipelines due to the relative motion with magnets. The distortion signals will reduce the accuracy of defect evaluation. In terms of the above issues, a double coil compensation method is proposed. Firstly, the causes, distribution positions, and diffusion characteristics of MIECs are determined through theoretical analysis to clarify the reasons for the distortion of high-speed MFL signals. Secondly, given the asymmetric distribution characteristics of MIECs, an asymmetric structure is proposed to capture MIECs at the defects and the magnetic poles, which cause distortion of high-speed MFL signals. Thirdly, based on the double coil groups, an active-passive strategy is proposed to distinguish pipeline surface and back defects, which ensures the integrity of defect evaluation and enhances the reliability of pipeline repair. Finally, the compensation performance and distinguishing capacity of the proposed method are verified by simulated and experimental analysis. The results show that compensated signals approximate the static MFL signals and have satisfactory performances of defect size quantification.


  • An Integrated Sensor of EMAT and MFL Using the Shared Magnets for Detecting Mixed Defects on the Steel Plate

    by Hang Xu, Jinhai Liu, Jiayue Sun, Lin Jiang, Qiangxin Li

    Mixed defects of oil and gas pipelines are both surface and back defects present at the same location (mixed-surface and mixed-back defects), which occur frequently under corrosion, stress concentration, external forces, etc. Due to the mutual occlusion and concealment of mixed-surface and mixed-back defects, it is difficult to detect mixed defects accurately, which poses a serious risk of pipeline leakage. Integrated detection methods are widely used to detect mixed defects due to their high detection accuracy and completeness compared to single detection methods. However, the methods require the mixed-surface and mixed-back defects to be staggered or not completely overlapped, because fixed energy propagation paths cannot break through the complete occlusion of mixed-surface defects to detect hidden mixed-back defects. In response to this issue, a novel integrated sensor of electromagnetic acoustic transducer (EMAT) and magnetic flux leakage using shared magnets is proposed. First, the detection principle, structural advantages, and the mutual influence between the two methods are reported through theoretical analysis. Second, a detour strategy is proposed to precisely detect hidden mixed-back defects by bypassing completely occluded mixed-surface defects. Third, a fully matched strategy is proposed to obtain enhanced and pure EMAT signals, which improves signal intensity and quality. Finally, the actual performance of the proposed sensor is verified through experiments. The experimental results show that the proposed sensor can accurately detect mixed defects with a width of 20 mm: 1 mm for mixed-surface and mixed-back defects, and the amplitude and signal-to-noise ratio are improved by 19.23% and 28.44% compared to a typical EMAT.


  • A Novel Incremental Defect Detection Method via Elastic Heterogeneous Distillation Network

    by Xiangkai Shen, Jinhai Liu, Huaguang Zhang, Lin Jiang, He Zhao, Hongyu Yang

    In industrial processes, the defect data is continuously accumulated over time, and new classes of defects can arise at any time. When a well-trained model is adapted to new classes, its performance in old classes will sharply decline. To solve the above issue, a novel incremental defect detection method named elastic heterogeneous distillation network (EHD-Net) is proposed. First, an elastic knowledge transfer method is proposed to selectively transfer the core features of the old classes, so that the old knowledge is retained and new knowledge is effectively learned at the feature level. Second, warping heterogeneous distillation learning is proposed for the first time. In warping heterogeneous distillation learning, a gradient warping layer is proposed to balance the optimization direction of gradients for both old and new classes, and the proposed heterogeneous distillation learning strategy can clarify the association and difference between new and old classes, so that the rapid learning of new knowledge and comprehensive transfer of old knowledge are guaranteed at the decision level. Finally, a well-trained evolutionary model is employed to achieve the detection of both old and new classes. The proposed method can effectively overcome the catastrophic forgetting of old classes and guarantee the independent learning of new classes. Experimental results on two industrial datasets show that EHD-Net outperforms existing advanced methods. Note to Practitioners —The motivation for this paper is an important and highly practical issue called class-incremental defect detection. Specifically, new classes of defects always appear in industrial scenarios over time and when a well-trained detection model adapts to the new classes, its performance in the old classes decreases dramatically. To solve the above issues, an elastic heterogeneous distillation network (EHD-Net) via elastic knowledge transfer and warping heterogeneous distillation learning is proposed. The proposed method not only effectively suppresses the catastrophic forgetting of old classes while realizing the fast learning of new classes, so that the model can simultaneously achieve the high-accuracy detection of both the new and old classes without the participation of old class samples. The validity of the proposed method is verified under two industrial datasets, which fully guarantees its practical application value.