Article
Chemistry, Analytical
Dingzhong Feng, Shanyu He, Zihao Zhou, Ye Zhang
Summary: This paper proposes a novel feature extraction method called principal component local preservation projections (PCLPP) for finger vein recognition. The method combines principal component analysis (PCA) and locality preserving projections (LPP) to construct a projection matrix that preserves both global and local features of the image, thereby improving the accuracy of image recognition.
Article
Computer Science, Artificial Intelligence
Jianyong Zhu, Jingwei Chen, Bin Xu, Hui Yang, Feiping Nie
Summary: This paper proposes a method called OLPPFS, which preserves the local geometric structure within the feature subspace by imposing the 2,0-norm pound constraint. The graph-embedding learning method is used to accelerate the construction of a sparsity affinity graph. Experimental results demonstrate that the proposed method is superior to others in preserving the local geometric structure of the dataset with less time consumption.
Article
Computer Science, Artificial Intelligence
Xiaohuan Lu, Jiang Long, Jie Wen, Lunke Fei, Bob Zhang, Yong Xu
Summary: In this paper, a novel method called LPP_SGE is proposed for unsupervised dimensionality reduction. LPP_SGE introduces a novel adaptive graph learning model and obtains the intrinsic graph and projection in a unified framework by fully exploring the representation information and distance information of the original data. It simultaneously captures the representation information and distance information in one term. Moreover, LPP_SGE enhances the robustness by introducing an 'l2,1' norm based projection constraint to select the most discriminative features from the complex data.
PATTERN RECOGNITION
(2022)
Article
Computer Science, Artificial Intelligence
Yifan Xia, Jiayi Ma
Summary: This paper proposes a novel framework called LOcality-guided Global-preserving Optimization (LOGO) for feature matching. The framework utilizes a graph-based optimization approach to identify inliers and remove mismatches, enhancing robustness to outliers. It also introduces a locality-guided matching strategy and local affine transformations for various scenarios.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2022)
Article
Automation & Control Systems
Jiayi Ma, Kaining Zhang, Junjun Jiang
Summary: A novel appearance-based loop closure detection system is proposed in this work, which selects candidate frames using Super-features and ASMK, and verifies loops using LPM-GC algorithm. Experimental results demonstrate that the proposed method achieves good performance in loop closure detection task.
IEEE-CAA JOURNAL OF AUTOMATICA SINICA
(2023)
Article
Engineering, Electrical & Electronic
Ning Liu, Zhihui Lai, Xuechen Li, Yudong Chen, Dongmei Mo, Heng Kong, Linlin Shen
Summary: In this paper, a novel regression method called Locality Preserving Robust Regression (LPRR) is proposed to address the issues encountered by conventional L-2, L-1 norm regression methods. Experimental results demonstrate that LPRR outperforms some famous subspace learning methods in classification tasks.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2021)
Article
Computer Science, Artificial Intelligence
Tingting Su, Dazheng Feng, Haoshuang Hu, Meng Wang, Mohan Chen
Summary: This study proposes a locality-preserving triplet discriminative projections algorithm to address the challenges of neighbor selection and intrinsic structure disruption in graph embedding. The algorithm constructs locality-preserving and discriminative graphs to enhance separability and preserve local structures. Experimental results demonstrate its superiority over other dimensionality reduction methods.
Article
Computer Science, Hardware & Architecture
Ning Zhang, Yuan Xu, Qun-Xiong Zhu, Yan-Lin He
Summary: This article proposes an improved locality preserving projections method based on the heat-kernel and cosine weight matrix, named HC-LPP, for fault diagnosis. By optimizing the weight matrix, HC-LPP considers both the distance and correlation among samples, and effectively reduces the dimensionality of data while preserving the spatial geometric structure.
IEEE TRANSACTIONS ON RELIABILITY
(2023)
Article
Multidisciplinary Sciences
Kangkana Bora, M. K. Bhuyan, Kunio Kasugai, Saurav Mallik, Zhongming Zhao
Summary: This paper presents a comprehensive analysis of critical features in assessing dysplasia in colonic polyps, using shape, texture, and color descriptors. The study includes feature extraction, statistical analysis, and classification with machine learning algorithms, demonstrating efficient designation and early detection. The proposed approach out-performs existing methods in colonic polyp identification, as demonstrated through comparison with deep learning models.
SCIENTIFIC REPORTS
(2021)
Article
Computer Science, Artificial Intelligence
Ruisheng Ran, Yinshan Ren, Shougui Zhang, Bin Fang
Summary: In this paper, a novel dimensionality reduction method named PDLPP is proposed, which addresses the small-sample-size problem of the DLPP method and achieves better pattern classification performance through nonlinear mapping. Experimental results demonstrate the superiority of PDLPP over state-of-the-art methods.
JOURNAL OF MATHEMATICAL IMAGING AND VISION
(2021)
Article
Mathematics
Minghua Wan, Yuxi Zhang, Guowei Yang, Hongjian Guo
Summary: The 2DESDLPP algorithm addresses the small sample size problem and reduces redundant information by combining matrix exponential function and elastic net regression with the 2DDLPP algorithm. It effectively preserves feature information and demonstrates higher accuracy rates compared to other mainstream feature extraction algorithms.
Article
Engineering, Environmental
Ping Wu, Xujie Zhang, Jiajun He, Siwei Lou, Jinfeng Gao
Summary: The paper presents a novel locality preserving randomized canonical correlation analysis (LPRCCA) method for real-time nonlinear process monitoring which maps original data to a randomized low-dimensional feature space and integrates local geometric structure information to improve data mining performance, reducing computational cost and showing significant advantages over kernel-based methods.
PROCESS SAFETY AND ENVIRONMENTAL PROTECTION
(2021)
Article
Automation & Control Systems
Ning Zhang, Yuan Xu, Qun-Xiong Zhu, Yan-Lin He
Summary: This article presents a novel dimensionality reduction algorithm named DPNLP for fault diagnosis. To solve the singular matrix problem, a regularization-based version of DPNLP called RDPNLP is also introduced. Simulation results show that RDPNLP outperforms other related methods in fault diagnosis.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Mathematical & Computational Biology
Hongming Liu, Yunyuan Gao, Jianhai Zhang, Juanjuan Zhang
Summary: Existing epileptic seizure automatic detection systems often face difficulties caused by high-dimensional electroencephalogram (EEG) features. In this paper, a method based on supervised locality preserving canonical correlation analysis (SLPCCA) is proposed to effectively use both the sample category information and nonlinear relationships between features. The experimental results show that the proposed method achieves excellent classification accuracy compared with several state-of-the-art methods.
MATHEMATICAL BIOSCIENCES AND ENGINEERING
(2022)
Article
Engineering, Electrical & Electronic
Saibal Ghosh, Pritam Paral, Amitava Chatterjee, Sugata Munshi
Summary: This study proposes a novel rough entropy-based granular fusion scheme to capture intensity variation in data for feature extraction in combination with two-dimensional locality preserving projection (2DLPP). The fusion technique incorporates the advantages of crisp granulation (CG) and quad-tree decomposition (QTD), while considering the uncertainties caused by these homogenous and non-homogeneous granulation techniques. It also works in the RGB color space to address information loss encountered by conventional granulation techniques in the gray space. Extensive experimental studies demonstrate the effectiveness of the proposed granular computing-based technique, especially in rugged environments, within a real-world human-robot interaction (HRI) framework.
IEEE SENSORS JOURNAL
(2023)
Article
Computer Science, Artificial Intelligence
Yipeng Chen, Ke Xu, Di He, Xiaojuan Ban
Summary: This study introduces a new method for modeling bounding box parameters with a Gaussian distribution to quantify the reliability of neural network predictions, and redesigns the loss function by adding virtual adversarial training to improve model prediction performance. Lightweight models were chosen as the backbone of the detector in experiments, demonstrating the effectiveness of the proposed approach.
INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS
(2022)
Article
Metallurgy & Metallurgical Engineering
Yujie Zhou, Ke Xu, Fei He, Zhiyan Zhang
Summary: This paper proposes an automatic anomaly detection algorithm to mine the relationship between abnormal production processes and product quality in the continuous casting process using data-driven methods. Convolutional neural networks and autoencoder models are used to detect various types of anomalies in time-dependent process parameters, and abnormal intervals are detected from time series. The proposed scheme progresses from univariate detection to multi-variable process monitoring, considering the nonlinear coupling of the process. Finally, various anomaly detection results are fused to analyze the presence of inclusions in the cast slab.
ISIJ INTERNATIONAL
(2022)
Article
Energy & Fuels
Dongdong Zhou, Ke Xu, Jinyong Bai, Di He
Summary: Developing new injection energy in blast furnace is an effective measure to reduce carbon emission and achieve carbon neutrality. This study used machine vision and deep learning methods to simultaneously detect coke particle size and temperature distribution in the tuyere zone, providing insights into the combustion process and temperature variation in the all-coke smelting model.
Article
Computer Science, Artificial Intelligence
Yipeng Chen, Ke Xu, Peng Zhou, Xiaojuan Ban, Di He
Summary: The predictive performance of supervised learning algorithms depends on the quality of labels. However, in the label collection process, multiple annotators often provide subjective and noisy labels influenced by their varying skill levels and biases. Blindly treating these noisy labels as the ground truth limits the accuracy of learning algorithms, especially when there is strong disagreement. This problem is particularly critical in the field of agricultural imaging in leaf disease classification.
IET IMAGE PROCESSING
(2022)
Article
Chemistry, Analytical
Xianyou Li, Ke Xu, Shun Wang
Summary: This paper presents a large-scale shaft diameter precision measurement method based on a dual camera measurement system and verifies its effectiveness. The experimental results show that the measurement accuracy of the proposed method is comparable to that of CMM.
Article
Materials Science, Multidisciplinary
Dongdong Zhou, Feng Gao, Junjian Wang, Ke Xu
Summary: An online surface temperature detection system was built for high-temperature rail steel plates using multi-spectral photography, and the accuracy of temperature detection results was improved through investigations on emissivity model, colorimetric thermometry, and noise filtering methods.
Article
Computer Science, Artificial Intelligence
Ying Liang, Ke Xu, Peng Zhou, Dongdong Zhou
Summary: In this article, a simple yet powerful image transformation network is proposed to remove textures and highlight defects at full resolution. The network utilizes a polynomial loss function combining perceptual loss, structural similarity loss, and image gradient loss to effectively suppress texture and emphasize defects. The method demonstrates superior performance in experiments and has been successfully applied to the surface defect online detection system of an aluminum ingot milling production line.
ADVANCED ENGINEERING INFORMATICS
(2022)
Review
Chemistry, Analytical
Dongdong Zhou, Ke Xu, Zhimin Lv, Jianhong Yang, Min Li, Fei He, Gang Xu
Summary: This article provides an overview of intelligent manufacturing in China's steel industry, including construction goals, overall framework, typical models, and major technologies. The research not only helps to comprehend the development of intelligent manufacturing in China's steel industry, but also provides vital inspiration for the digital and intelligence updates and quality improvement of the manufacturing industry.
Article
Chemistry, Analytical
Xianyou Li, Shun Wang, Ke Xu
Summary: This study focuses on online size measurement of threads at the end of sucker rods based on machine vision, achieving accurate and efficient thread parameter measurements through optimized edge detection and interference elimination techniques.
Article
Optics
S. A. N. A. O. HUANG, Y. I. N. G. J. I. E. SHI, M. I. N. G. LI, J. I. N. G. W. E. I. QIAN, K. E. XU
Summary: This study analyzes the convergence of a near-field point light source in water using a light propagation model. The photometric stereo formula is determined based on accurate estimation of illuminance entering the camera. An underwater photometric stereo system is designed to validate the proposed method's feasibility. Experimental results demonstrate improved accuracy in normal calculation, enabling accurate 3D reconstruction for underwater surface microdefect detection.
Article
Optics
Shun Wang, Ke Xu
Summary: Surface roughness evaluation is crucial for enhancing surface properties. Current machine vision-based roughness measurement methods neglect the benefits of 3D morphology, resulting in complex and inaccurate algorithms. To address this, we propose an efficient surface roughness evaluation method using Near Point Lighting Photometric Stereo (NPL-PS).
OPTICS AND LASERS IN ENGINEERING
(2023)
Article
Optics
Qian Sun, Ke Xu, Huajie Liu, Jianer Wang
Summary: In the field of quality control for aluminum sheets, the proposed unsupervised defect detection method utilizes combined bright-field and dark-field illumination to enhance the visibility of defects. A patch-aware feature matching network is introduced, which only requires defect-free samples for training and can extract high-resolution features from bright-field and dark-field images simultaneously. The method achieves superior performance in defect identification and segmentation, outperforming other neural network-based methods, and has been implemented in a real-time machine vision system to improve detection efficiency and product quality.
OPTICS AND LASERS IN ENGINEERING
(2023)
Article
Chemistry, Analytical
Hanxin Zhang, Qian Sun, Ke Xu
Summary: This study presents a self-supervised binary classification algorithm for defect image classification in DCIP images. By leveraging data augmentation and deep convolutional neural networks, defect and defect-free data are combined for classification, and anomaly scores are used for evaluation. The results demonstrate the robust performance and high accuracy of this method in practical applications.
Article
Engineering, Multidisciplinary
Shun Wang, Xianyou Li, Yufei Zhang, Ke Xu
Summary: This paper systematically studies the measurement error of the calibration coefficient method when the camera's optical axis is not perpendicular to the measurement plane. It proposes a method to derive the distribution relationship of physical points corresponding to evenly distributed pixels in the pixel coordinate system. Theoretical analysis and experimental results validate the accuracy of the theoretical analysis and provide theoretical guidance for practical applications.
ENGINEERING RESEARCH EXPRESS
(2023)
Article
Construction & Building Technology
Shun Wang, Ke Xu, Baohua Li, Xiangyu Cao
Summary: This study proposes a symmetrical twin-light photometric stereo (TLPS) detection algorithm based on photometric stereo theory, along with an image correction algorithm, for detecting defects on the surface of ductile cast iron pipes (DCIPs). Experimental results demonstrate that the TLPS algorithm can effectively detect and locate micro defects such as pores, pinholes, and scratches on the cast pipe surfaces.
CASE STUDIES IN CONSTRUCTION MATERIALS
(2023)
