NMNachuan Ma

Nachuan Ma

*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


Nachuan Ma has a total of 35 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Disparity Arbiter: Pseudo label mining with student deficiency awareness

    by Chuang-Wei Liu, Mengtan Zhang, Nachuan Ma, Jin Wu, Hanli Wang, Qijun Chen, Rui Fan

  • VLLM-LAD: Visual Large Language Model for Zero-shot Logical Anomaly Detection

    by Yun Peng, Xiao Lin, Nachuan Ma, Chengju Liu, Qijun Chen

    Logical anomaly detection aims to identify inconsistencies in object relationships and scene semantics, which is a challenging task requiring high-level reasoning capabilities that go beyond low-level structural cues. Existing methods rely heavily on large amounts of labeled data, struggle in zero-shot scenarios and lack deep semantic understanding in complex scenes with multiple objects. To address this, we propose VLLM-LAD, a visual large language model framework specifically designed for zero-shot logical anomaly detection. It incorporates a Local Detail Visual Perceiver (LDVP) module for fine-grained anomaly localization and a Global Semantic Feature Fusion (GSFF) module for enhanced multimodal reasoning. We also introduce a global-local-regional contrastive learning strategy to align visual and textual representations, thereby achieving robust zero-shot performance. Extensive experiments on MVTec LOCO demonstrate that VLLM-LAD outperforms existing methods in accuracy and interpretability. To further advance research in this area, we introduce the RLAD dataset. This large-scale, diverse dataset is designed for logical anomaly detection, encompassing complex scenarios that demand reasoning about object relationships and global semantics. Our method not only excels on the RLAD dataset but also provides detailed textual explanations and segmentation masks, enhancing its transparency and applicability to real-world tasks. This work establishes a new benchmark for zero-shot logical anomaly detection, offering a robust framework and a valuable dataset to propel future research.


  • Self-Derived Multimodal Knowledge Distillation for Real-Time Road Crack Detection

    by Nachuan Ma, Qiang Hu, Zhengfei Song, Sicen Guo, Rui Fan

    Road crack detection is vital for intelligent transportation and infrastructure maintenance. In recent years, deep learning-based methods have emerged for automated road crack detection with enhanced performance. Nevertheless, existing methods fail to balance detection performance, model complexity, and processing efficiency, hindering their deployment in autonomous road inspection systems. To fill this gap, we propose a novel lightweight network for real-time road crack detection, called LR-CrackNet. It employs a teacher-student network architecture, where the teacher network leverages both image data and adaptively self-derived boundary information from pixel-level annotations, while the student network relies solely on image input. The training process focuses on distilling the teacher network’s detection capabilities into the student network. Both networks share a U-shaped segmentation design, integrating novel residual dual-layer depthwise convolution blocks and a fast Transformer block for efficient integration of detailed hierarchical spatial information and global long-range contextual information. Furthermore, an attention-enhanced discriminator network is designed, which aims to improve detection performance by enforcing global consistency and providing pixel-level feedback. Comprehensive experimental results demonstrate that LR-CrackNet achieves state-of-the-art (SoTA) detection performance on the UDTIRI-Crack, DeepCrack, and CamCrack789 datasets and surpasses existing publicly available algorithms in processing efficiency, with 324.40 frames per second (FPS) on a single NVIDIA RTX3090. The source code package is available at https://mias.group/LR-CrackNet/.


  • Top Cited Papers

    SAM-LAD: Segment Anything Model meets zero-shot logic anomaly detection

    by Yun Peng, Xiao Lin, Nachuan Ma, Jiayuan Du, Chuangwei Liu, Chengju Liu, Qijun Chen


  • UP-CrackNet: Unsupervised Pixel-Wise Road Crack Detection via Adversarial Image Restoration

    by Nachuan Ma, Rui Fan, Lihua Xie

    Over the past decade, automated methods have been developed to detect cracks more efficiently, accurately, and objectively, with the ultimate goal of replacing conventional manual visual inspection techniques. Among these methods, semantic segmentation algorithms have demonstrated promising results in pixel-wise crack detection tasks. However, training such networks requires a large amount of human-annotated datasets with pixel-level annotations, which is a highly labor-intensive and time-consuming process. Moreover, supervised learning-based methods often struggle with poor generalizability in unseen datasets. Therefore, we propose an unsupervised pixel-wise road crack detection network, known as UP-CrackNet. Our approach first generates multi-scale square masks and randomly selects them to corrupt undamaged road images by removing certain regions. Subsequently, a generative adversarial network is trained to restore the corrupted regions by leveraging the semantic context learned from surrounding uncorrupted regions. During the testing phase, an error map is generated by calculating the difference between the input and restored images, which allows for pixel-wise crack detection. Our comprehensive experimental results demonstrate that UP-CrackNet outperforms other general-purpose unsupervised anomaly detection algorithms, and exhibits satisfactory performance and superior generalizability when compared with state-of-the-art supervised crack segmentation algorithms. Our source code is publicly available at mias.group/UP-CrackNet.


  • Semi-Supervised Bolt Anomaly Detection Based on Local Feature Reconstruction

    by Yun Peng, Chuangwei Liu, Yi Yan, Nachuan Ma, Deming Wang, Chengju Liu, Qijun Chen

    During the process of high-speed railway online maintenance, anomaly detection of targets in grease injection facility is extremely challenging. Environmental uncertainty (noise disturbance, various illumination conditions, changeable camera perspective, etc.) limits the performance of existing unsupervised anomaly detection methods, which tend to struggle in such complex backgrounds without prior knowledge of anomalies. Therefore, we propose a novel semi-supervised anomaly detection algorithm that only learns from anomaly-free samples and annotations while can recognize both anomaly-free and anomaly targets (key-areas) and detecting anomalous targets simultaneously. We firstly propose a key-area recognition module (KRM) to segment anomaly-free targets and predict coordinates of both anomaly-free and anomalous key-areas’ center point (key-points). By using the predicted coordinates, we generated a circular mask to erase features within the key-areas. Based on the erased features, a local feature reconstruction and segmentation module (LFRSM) is proposed to restore the latent embeddings within the key-areas and segment both anomaly-free and anomalous key-areas. The final anomaly results are obtained by analyzing the original segmentation results of KRM and the reconstructed segmentation results of LFRSM. To validate the effectiveness of the proposed algorithm, we develop a dataset, called BOLT-DET, for bolt anomaly detection in complex scenes. The proposed algorithm outperforms other existing state-of-the-art (SOTA) anomaly detection algorithms (SPADE, DFR, CFLOW, Student-Teacher, PatchCore, and DRAEM) with mIoU of 0.66, AUROC of 0.92 and F1-Measure of 0.69 on BOLT-DET. It can accurately detect anomalies and meet the requirement of practical industrial applications.


  • Open Access

    Computer vision for road imaging and pothole detection: a state-of-the-art review of systems and algorithms

    by Nachuan Ma, Jiahe Fan, Wenshuo Wang, Jin Wu, Yu Jiang, Lihua Xie, Rui Fan

    Computer vision algorithms have been utilized for 3-D road imaging and pothole detection for over two decades. Nonetheless, there is a lack of systematic survey articles on state-of-the-art (SoTA) computer vision techniques, especially deep learning models, developed to tackle these problems. This article first introduces the sensing systems employed for 2-D and 3-D road data acquisition, including camera(s), laser scanners and Microsoft Kinect. It then comprehensively reviews the SoTA computer vision algorithms, including (1) classical 2-D image processing, (2) 3-D point cloud modelling and segmentation and (3) machine/deep learning, developed for road pothole detection. The article also discusses the existing challenges and future development trends of computer vision-based road pothole detection approaches: classical 2-D image processing-based and 3-D point cloud modelling and segmentation-based approaches have already become history; and convolutional neural networks (CNNs) have demonstrated compelling road pothole detection results and are promising to break the bottleneck with future advances in self/un-supervised learning for multi-modal semantic segmentation. We believe that this survey can serve as practical guidance for developing the next-generation road condition assessment systems.


  • Deep Neural Network Enhanced Sampling-Based Path Planning in 3D Space

    by Jiankun Wang, Xiao Jia, Tianyi Zhang, Nachuan Ma, Max Q.-H. Meng

    Robot path planning in 3D space is a challenging problem for its complex configuration. Sampling-based algorithms have gained great success in solving path planning problems in 3D space, but the quality of the initial path is not guaranteed and the convergence to the optimal solution is slow. To address these problems, in this article, we present a novel sampling-based path planning framework enhanced by the deep neural network (DNN) with applications to 3D space. In the proposed framework, we first train the DNN with a number of successful path planning cases in 3D space. Then the DNN is utilized to predict the promising region where the feasible path probably exists for a given path planning problem. This predicted promising region serves as a nonuniform sampling heuristic to bias the sampling process of the path planner. In this way, the path planner can focus on the promising region in the exploration and exploitation process so that the path planning speed gets accelerated. We conduct numerical simulations to evaluate the performance of the proposed algorithm and the results show that it can perform much better than conventional path planning algorithms. Furthermore, we also investigate the performance of different DNN architectures for path planning in 3D space. Note to Practitioners—In this work, we aim to provide an efficient learning-based method to accelerate the robot path planning process in 3D space. Conventional path planning algorithms need to perceive the environment first, and then implement a series of calculations such as collision checking and data storing to generate a feasible path. When facing complex and high-dimensional environments, they do not perform well. But the proposed neural network method in this article can predict the promising region where the feasible path exists for any given environment. This prediction result is used to guide the path planning process so that the algorithm performance can get significantly improved. Apart from sampling-based algorithms, the proposed neural network model can also be extended to other types of path planning algorithms.


  • Open Access

    Deep Learning-based Biological Anatomical Landmark Detection in Colonoscopy Videos

    by Kaiwei Che, Chengwei Ye, Yibing Yao, Nachuan Ma, Ruo Zhang, Jiankun Wang, Max Q. -H. Meng

    Colonoscopy is a standard imaging tool for visualizing the entire gastrointestinal (GI) tract of patients to capture lesion areas. However, it takes the clinicians excessive time to review a large number of images extracted from colonoscopy videos. Thus, automatic detection of biological anatomical landmarks within the colon is highly demanded, which can help reduce the burden of clinicians by providing guidance information for the locations of lesion areas. In this article, we propose a novel deep learning-based approach to detect biological anatomical landmarks in colonoscopy videos. First, raw colonoscopy video sequences are pre-processed to reject interference frames. Second, a ResNet-101 based network is used to detect three biological anatomical landmarks separately to obtain the intermediate detection results. Third, to achieve more reliable localization of the landmark periods within the whole video period, we propose to post-process the intermediate detection results by identifying the incorrectly predicted frames based on their temporal distribution and reassigning them back to the correct class. Finally, the average detection accuracy reaches 99.75\%. Meanwhile, the average IoU of 0.91 shows a high degree of similarity between our predicted landmark periods and ground truth. The experimental results demonstrate that our proposed model is capable of accurately detecting and localizing biological anatomical landmarks from colonoscopy videos.


  • Open Access

    A survey of learning‐based robot motion planning

    by Jiankun Wang, Tianyi Zhang, Nachuan Ma, Zhaoting Li, Han Ma, Fei Meng, Max Q.‐H. Meng

    A fundamental task in robotics is to plan collision‐free motions among a set of obstacles. Recently, learning‐based motion‐planning methods have shown significant advantages in solving different planning problems in high‐dimensional spaces and complex environments. This article serves as a survey of various different learning‐based methods that have been applied to robot motion‐planning problems, including supervised, unsupervised learning, and reinforcement learning. These learning‐based methods either rely on a human‐crafted reward function for specific tasks or learn from successful planning experiences. The classical definition and learning‐related definition of motion‐planning problem are provided in this article. Different learning‐based motion‐planning algorithms are introduced, and the combination of classical motion‐planning and learning techniques is discussed in detail.