Anchor Points Research Articles

Improving Safety in High-Altitude Work: Semantic Segmentation of Safety Harnesses with CEMFormer

The symmetry between production efficiency and safety is a crucial aspect of industrial operations. To enhance the identification of proper safety harness use by workers at height, this study introduces a machine vision approach as a substitute for manual supervision. By focusing on the safety rope that connects the worker to an anchor point, we propose a semantic segmentation mask annotation principle to evaluate proper harness use. We introduce CEMFormer, a novel semantic segmentation model utilizing ConvNeXt as the backbone, which surpasses the traditional ResNet in accuracy. Efficient Multi-Scale Attention (EMA) is incorporated to optimize channel weights and integrate spatial information. Mask2Former serves as the segmentation head, enhanced by Poly Loss for classification and Log-Cosh Dice Loss for mask loss, thereby improving training efficiency. Experimental results indicate that CEMFormer achieves a mean accuracy of 92.31%, surpassing the baseline and five state-of-the-art models. Ablation studies underscore the contribution of each component to the model’s accuracy, demonstrating the effectiveness of the proposed approach in ensuring worker safety.

Automatically Detecting Anchor Cells and Clustering for scRNA-Seq Data Using scTSNN.

Advancing in single-cell RNA sequencing techniques enhances the resolution of cell heterogeneity study. Density-based unsupervised clustering has the potential to detect the representative anchor points and the number of clusters automatically. Meanwhile, discovering the true cell type of scRNA-seq data in the unsupervised scenario is still challenging. To this end, we proposed a tensor shared nearest neighbor anchor clustering for scRNA-seq data, named scTSNN, which first makes use of the tensor affinity learning module to mine the local-global balanced topological structures among cells, next designs density-based shared nearest neighbor measurement method to automatically detect anchor cells, finally partitions the non-anchor cells to obtain the clustering results. Validated on synthetic datasets and scRNA-seq datasets, scTSNN not only exactly detects the complicated structures but also has better performance in accuracy and robustness compared with the state-of-the-art methods. Moreover, case studies on mammalian cells and cervical cancer tumor cells demonstrate the selected anchor cells of scTSNN benefit the cell pseudotime inference and rare cell identification, which show good application and research value of scTSNN.

Behaviour of paddled energy-absorbing rockbolts under complex loading laboratory conditions

Since their introduction in 2010, paddled energy-absorbing rockbolts have been widely used in seismically active hard rock mines. This paper provides new data and reviews previous work to quantify the performance of paddled energy-absorbing rockbolts under controlled laboratory conditions. Of significance is the realization that the typical split location in impact tests, at the centre between two paddle anchor points can at best provide an upper limit value. This inherent variability in performance under different testing configurations should be acknowledged and taken into consideration in the design of ground support in seismic conditions. This paper discusses the reduction in capacity of paddled rockbolts as a function of loading angle from a maximum value during axial tests (0° loading angle) to the lowest value during pure shear (loading angle of 90°). A significant reduction in displacement capacity is observed as the loading angle changes between axial (0°) and 40°. Beyond a 40° loading angle up to shear (90°) loading the reduction in displacement capacity is less significant. Due to the variances in the capacity observed through the variance of the testing configuration: loading mechanism, direction of loading, location / presence of a discontinuity, it must be recognized that results from laboratory-based testing are not independent of the testing configuration. This should be acknowledged when extrapolating anticipated performance in the field.

The osteohistology of Orthosuchus stormbergi using synchrotron radiation microcomputed tomography.

Orthosuchus stormbergi was a small-bodied crocodyliform, representative of a diverse assemblage of Early Jurassic, early branching crocodylomorph taxa from the upper Elliot Formation of South Africa. The life history of these early branching taxa remains poorly understood, with only sparse investigations into their osteohistology, yet species like Orthosuchus have potential to inform about the macroevolution of growth strategies on the stem leading to crown crocodilians. In order to elucidate the growth patterns of Orthosuchus, we used propagation phase contrast X-ray synchrotron micro-computed tomography to virtually image the osteohistology of the postcrania of two specimens, including multiple elements from the type (SAM-PK-K409), and the femur of a referred specimen (BP/1/4242). In total, we scanned nine mid-diaphyseal sections of the humerus, radius, ulna, radiale, femur, tibia, fibula, and a rib. We then compared our results to osteohistological sections of crocodylomorph taxa from the published literature. Our results show that the most predominant bone tissue type in Orthosuchus is lamellar, with a few patches of woven and parallel-fibred bone. The type specimen contains four to five lines of arrested growth and the hindlimb elements present outer circumferential lamellae, whereas the referred specimen contains six to seven. Both specimens grew at similar rates, reaching adult skeletal body size at year four or five. The sectioned bones, most notably the radius and ulna, are comparatively thick walled and compact. Our virtual osteohistological sections are one of the first for an early branching crocodyliform, and the broad sample of skeletal elements makes Orthosuchus a key anchor point for understanding the plesiomorphic life history traits of the clade.

Challenges of Continuous Wave EPR of Broad Signals—The Ferritin Case

AbstractThe study of continuous wave (cw) electron paramagnetic resonance (EPR) spectra still poses a challenge for very broad signals, especially when the spectrum extends over a large part of the accessible field range. The difficulties derive from instrumental challenges, because of insufficient modulation depth and the need to apply measurement conditions that enhance cavity background. The biggest problem, however, is how to define a baseline such that spectral distortions are minimized. Conventional methods rely on a suitable choice of points outside the range of the signal of interest to perform a polynomial interpolation. These methods are effective in most cases where the signal of interest comprises only a narrow range of magnetic field (narrow features). In this study, a novel method of baseline correction for broad signals is proposed and compared to conventional methods. It takes into account that there are only few anchor points for the baseline. The method is applied to the signal of the iron-storage protein ferritin. The ferritin signal is a broad band that extends from zero to 0.8 T. An approach is developed by which this broad signal is analyzed reliably. The method is also extended to the case where the broad signal is superimposed on narrow signals and enables to extract the parameters of both types of signals in a fitting pipeline.

Analysis of Multi-Dimensional Layers in Historic Districts Based on Theory of the Historic Urban Landscape: Taking Shenyang Fangcheng as an Example

The accelerated process of urbanisation in China is resulting in a decline in and threat to the historic landscape of historic districts. This study is based on the theory of historic urban landscapes and employs a multi-dimensional layers research framework for historic districts. It adopts a single case study and a research method that combines quantitative and qualitative methods. The stratification elements of Fangcheng in Shenyang are identified and summarised, the process of stratification is analysed, and the stratification patterns and laws are summarised through the acquisition and collection of multivariate data. The findings of this study indicate that the stratification elements of the Fangcheng Historic District have undergone five distinct phases of stratification evolution. The resulting stratification pattern can be summarised as follows: newborn, preserve, override, juxtaposition and decession. The spatial elements are layered in the following pattern: The historic landscape can be conceptualised as comprising four layers: (1) the layering of the historic landscape with large public buildings as the anchor point; (2) the layering of the historic landscape with the spatial pattern as the skeleton; (3) the layering of the historic landscape with the iconic buildings as the nodes; and (4) the layering of the historic landscape with the correlative elements as the substrate. The law of value element layering primarily reflects the principles of concentration, diversity and adaptation. In light of the urban historic landscape theory, the reconstruction and restoration strategies, integration of old and new and adaptive conservation of historic landscapes are proposed to offer novel insights and guidance for the conservation of the historic landscape in the Fangcheng Historic District.

Enhanced Cross Layer Refinement Network for robust lane detection across diverse lighting and road conditions

With the rapid development of autonomous driving technology, lane detection, a key component of intelligent vehicle systems, is crucial for ensuring road safety and efficient vehicle navigation. In this paper, a new lane detection method is proposed to address the problem of degraded performance of existing lane detection methods when dealing with complex road environments. The proposed method evolves from the original Cross Layer Refinement Network (CLRNet) by incorporating two of our carefully designed core components: the Global Feature Optimizer (GFO) and the Adaptive Lane Geometry Aggregator (ALGA). The GFO is a multi-scale attention mechanism that mimics the human visual focusing ability, effectively filtering out unimportant information and focusing on the image regions most relevant to the task. The ALGA is a shape feature-aware aggregation module that utilizes the shape prior of lanes to enhance the correlation of anchor points in an image, better fusing global and local information. By integrating both components into CLRNet, an enhanced version called Enhanced CLRNet (E-CLRNet) is presented, which exhibits higher performance stability in complex roadway scenarios. Experiments on the CULane dataset reveal that E-CLRNet demonstrates superior performance stability over the original CLRNet in complex scenarios, including curves, shadows, missing lines, and dazzling light conditions. In particular, in the curves, the F1 score of E-CLRNet is improved by almost 3% over the original CLRNet. This study not only improves the accuracy and performance stability of lane detection but also provides a new solution for the application of autonomous driving technology in complex environments, which promotes the development of intelligent vehicle systems.

Study on foaming behavior of block copolymer/inorganic particles co‐construction of toughened epoxy resin foams

AbstractLow toughness and brittleness limit its application range of thermosetting epoxy resin (EP) based foams. In current research, inorganic particles are commonly utilized as heterogeneous nucleating agents to enhance foaming quality and EP toughness. However, the increasing inorganic particle content usually leads to particle agglomeration, which negatively impacts foaming quality and reduces EP performance. Block copolymers, which appear as liquid to effectively mitigate agglomeration, are employed as toughening materials for EP. In this study, a small number of modified calcium silicate particles (MCS) were used as anchor points to induce the distribution of block copolymers (P‐F68) to improve EP toughness and effectively enhance cell nucleation, thus, EP/MCS/P‐F68 foams performance was improved. The research results showed that EP foams obtained excellent cell structure and very small foam density (0.256 g/cm3) at the P‐F68 content of 10 wt%. The continuous increase of P‐F68 content would lead to the thickening of cell wall and the increase of cell size. Simultaneously, the core‐shell structure formed by P‐F68 in EP effectively enhanced the toughness of EP foams and mitigated its brittleness, offering crucial theoretical guidance for toughening and regulating the cell structure of EP foams.

Sequential Linker Installation in Metal-Organic Frameworks.

ConspectusMetal-organic frameworks (MOFs) represent a sophisticated blend of inorganic and organic components, promoting the development of coordination chemistry greatly and offering a versatile platform for tailored functionalities. By combining various metal nodes, organic linkers, and functional guests, MOFs provide numerous pathways for their design, synthesis, and customization. Among these, sequential linker installation (SLI) stands out as a novel and crucial strategy, enabling the precise integration of desired properties and functions at the atomic scale. SLI enhances structural diversity and stability while facilitating the meticulous construction of robust frameworks by leveraging open metal sites and functional organic linkers at targeted locations. Compared to the direct synthesis of MOFs, postsynthetic modification methods allow for precise regulation of their structures and corresponding properties. While unlike conventional postsynthetic modification methods, SLI requires the careful selection of linkers and framework design to ensure precise positioning for installation, which gives rise to the well-designed and ordered positions for the installed linkers, confirmed directly by X-ray diffraction technology.Recent advancements in MOF synthesis have led to the creation of increasingly tailored flexible matrix structures, particularly due to the diverse connection modes of multicore metal clusters, especially for the Zr6 cluster. The spatial hindrance of certain ligands has resulted in the formation of unsaturated metal clusters and various missing linker pockets. Examples of these advanced MOFs include PCN-606, PCN-608, PCN-609, PCN-700, and PCN-808, which feature specific open metal sites and certain framework flexibility conducive to SLI. Strategically positioned open metal sites within these frameworks serve as predetermined anchor points for desired functional molecules, while the frameworks' flexibility can accommodate molecules of varying sizes to a certain extent, enlarging the scopes of application greatly. This precise positioning of functional groups enables the creation of tailored sites for enhanced applications, such as adsorption, catalysis, and recognition.In this Account, we delve into the intricate process of designing and synthesizing MOFs with appropriate missing-linker pockets for the aforementioned applications. We discuss the meticulous selection of functional linkers and the methods used to insert them into the corresponding missing-linker pockets within the MOFs. Additionally, we explore the diverse properties and functionalities of the resulting MOFs, focusing on their adsorptive, catalytic, and recognition performance. Furthermore, we provide insights into the future trajectory of SLI methods, complemented by our recent works. This Account not only reviews the evolution of the SLI method but also underscores its practical applications across various functional domains, paving a rational pathway for the future development of advanced multifunctional MOFs through this method.

Enhanced Biomechanical Properties of the Pectineal Ligament Support Its Reliability for Apical Pelvic Organ Prolapse Repair

Pelvic organ prolapse impacts an increasing number of women in the United States. The standard approach to correcting apical pelvic organ prolapse uses the sacral anterior longitudinal ligament (SALL) to lift the vaginal apex; however, this approach may result in recurrent prolapse. A newer procedure utilizes the pectineal ligament (PL), which may be a more reliable anchor point. This study compares the biomechanical properties of these two ligaments to elucidate which can withstand more stress to provide long-term stability following prolapse. Seventeen formalin-embalmed donors were used (PL: 17 right, 16 left; SALL, 15). The PL was evaluated to better characterize the ligament’s properties within the pelvis using digital calipers and descriptive statistics. Mean values were statistically evaluated using an independent t test (p = 0.05) but no differences in laterality were appreciable. The PL and SALL samples were harvested and evaluated using a mechanical tester to determine their force at failure (N), toughness (Jm−2), and elastic modulus (MPa). The PL had increased values in the mean force at failure and toughness than the SALL when evaluated by each side as well as a combined mean value. These differences were statistically significant (p = 0.05) for toughness as evaluated using an independent t-test (right, p = 0.004; left, p = 0.005; combined, p = 0.002) and force at failure [right, p = 0.001 (independent t-test); left, p = 0.004 and combined, p = 0.005 (Mann–Whitney U test)], indicating that the PL may permit more deformation, but greater resistance to catastrophic failure as compared to the SALL. When evaluating any statistical differences in modulus, the individual and combined values were increased for the PL as compared to the SALL but were not significant (right, p = 0.290; left, p = 0.143; combined, p = 0.110) suggesting a stiffer material that may be more prone to catastrophic failure once a tear has begun. Collectively, these inherent biomechanical properties of the pectineal ligament indicate the ligament may be a more reliable anchor point for pelvic organ prolapse repair than the SALL.

UWB/IMU integrated indoor positioning algorithm based on robust extended Kalman filter

Abstract As a fundamental method for integrated indoor positioning, Ultra wideband (UWB)/ Inertial Measurement Unit (IMU) offers more stable performance compared to separate UWB sensor and IMU sensor positioning and uses the least squares (LS) and extended Kalman filter (EKF) for positioning processing, however, the performance of conventional LS and EKF will be seriously affected when the contamination rate of measurement noise is high. To address this problem, this paper proposes a robust extended Kalman filter (REKF) based algorithm for robot-integrated indoor positioning, which combines UWB and IMU positioning methods to obtain a more robust system performance and optimal positioning accuracy. In this paper, a vehicle equipped with a UWB/IMU sensor and four anchor points is set up to perform linear, circular, and trajectory motions in an experimental environment with visible obstacles. The experimental results show that compared with EKF, REKF improves the localization precision and accuracy in the three motions by 20.98%, 35.51%, and 47.06%, respectively. REKF effectively improves the accuracy and robustness of the robot’s indoor positioning, and provides an improved and practical method for integrated indoor positioning algorithms to make indoor positioning more accurate and reliable.

Large-scale multi-view spectral clustering based on two-stage well-distributed anchor selection

Spectral clustering has attracted much attention because of its good clustering effect, but its high computational cost makes it difficult to apply to large-scale multi-view clustering. In response to this issue, a simple and efficient large-scale multi-view spectral clustering algorithm is proposed, which is based on a Two-stage Well-distributed Anchor Selection strategy (TWAS). Firstly, the data set is divided into several disjoint sample blocks to get the global well-distributed anchor candidate. Then, the algorithm proceeds to select anchor points within each local candidate anchor set. This two-stage anchor selection strategy facilitates the identification of anchors with significant representativeness at a reduced computational expense, thereby adeptly capturing the intrinsic data structure. Secondly, the present study devises an adaptive near-neighbor graph learning approach to construct an anchor-based intra-view similarity matrix. Finally, the multiple views are fused to obtain a consistent inter-view similarity matrix, and the clustering results are obtained. Extensive experiments demonstrate the effectiveness, efficiency, and stability of the TWAS algorithm.

Enhanced tensor based embedding anchor learning for multi-view clustering

Existing anchor graph based multi-view clustering methods can overcome the problem of high computational cost in traditional multi-view clustering methods. However, the anchor points selected from high-dimensional data often contain irrelevant noise and outliers, which would affect the clustering performance. To address this issue, we propose an embedding anchor based multi-view clustering method, called enhanced tensor based embedding anchor learning (ETEAL). Specifically, we unify the learning process of latent embedding space, anchor points, and anchor graphs into a common framework, which eliminates noise and redundant data in the original space and enhances the synergistic optimization between the components. Meanwhile, we employ an enhanced tensor strategy to constrain the embedding anchor graphs, which exploits the higher-order relationships between views and recovers the global low-rank property of the embedding anchor graphs. Finally, we develop an anchor graph fusion strategy, which significantly reduces the huge overhead of traditional graph fusion that requires the construction of complete graphs. Experimental results on eight benchmark datasets show that the proposed method significantly outperforms other state-of-the-art methods in terms of scalability and clustering accuracy.

Enhancing ophthalmology medical record management with multi-modal knowledge graphs

The electronic medical record management system plays a crucial role in clinical practice, optimizing the recording and management of healthcare data. To enhance the functionality of the medical record management system, this paper develops a customized schema designed for ophthalmic diseases. A multi-modal knowledge graph is constructed, which is built upon expert-reviewed and de-identified real-world ophthalmology medical data. Based on this data, we propose an auxiliary diagnostic model based on a contrastive graph attention network (CGAT-ADM), which uses the patient’s diagnostic results as anchor points and achieves auxiliary medical record diagnosis services through graph clustering. By implementing contrastive methods and feature fusion of node types, text, and numerical information in medical records, the CGAT-ADM model achieved an average precision of 0.8563 for the top 20 similar case retrievals, indicating high performance in identifying analogous diagnoses. Our research findings suggest that medical record management systems underpinned by multimodal knowledge graphs significantly enhance the development of AI services. These systems offer a range of benefits, from facilitating assisted diagnosis and addressing similar patient inquiries to delving into potential case connections and disease patterns. This comprehensive approach empowers healthcare professionals to garner deeper insights and make well-informed decisions.

Anchor Space Optimal Transport as a Fast Solution to Multiple Optimal Transport Problems.

In machine learning, optimal transport (OT) theory is extensively utilized to compare probability distributions across various applications, such as graph data represented by node distributions and image data represented by pixel distributions. In practical scenarios, it is often necessary to solve multiple OT problems. Traditionally, these problems are treated independently, with each OT problem being solved sequentially. However, the computational complexity required to solve a single OT problem is already substantial, making the resolution of multiple OT problems even more challenging. Although many applications of fast solutions to OT are based on the premise of a single OT problem with arbitrary distributions, few efforts handle such multiple OT problems with multiple distributions. Therefore, we propose the anchor space OT (ASOT) problem: an approximate OT problem designed for multiple OT problems. This proposal stems from our finding that in many tasks the mass transport tends to be concentrated in a reduced space from the original feature space. By restricting the mass transport to a learned anchor point space, ASOT avoids pairwise instantiations of cost matrices for multiple OT problems and simplifies the problems by canceling insignificant transports. This simplification greatly reduces its computational costs. We then prove the upper bounds of its 1 -Wasserstein distance error between the proposed ASOT and the original OT problem under different conditions. Building upon this accomplishment, we propose three methods to learn anchor spaces for reducing the approximation error. Furthermore, our proposed methods present great advantages for handling distributions of different sizes with GPU parallelization.

Generalised eigenfunction expansion and singularity expansion methods for canonical time-domain wave scattering problems

The generalised eigenfunction expansion method (GEM) and the singularity expansion method (SEM) are applied to solve the canonical problem of wave scattering on an infinite stretched string in the time domain. The GEM, which is shown to be equivalent to d’Alembert’s formula when no scatterer is present, is also derived in the case of a point-mass scatterer coupled to a spring. The discrete GEM, which generalises the discrete Fourier transform, is shown to reduce to matrix multiplication. The SEM, which is derived from the Fourier transform and the residue theorem, is also applied to solve the problem of scattering by the mass–spring system. The GEM and SEM are also used to solve the problem of wave scattering by a mass positioned a fixed distance from an anchor point, which supports more complicated resonant behavior.

Floating body motion coupling study and economic evaluation of different floating photovoltaic arrangement schemes under hybrid solar-wind power generation

Abstract Reducing the number of anchor points and moorings for floating platforms is one of the ways to reduce the integration and maintenance costs of floating offshore wind and solar farms. This paper focuses on the technical feasibility and economic viability of reducing the number of anchor points and moorings for floating offshore PV platforms by using a monopile of a stationary wind turbine as a mooring anchor point and introducing a shared line system. In this paper, three different floating solar layouts and their corresponding mooring systems are numerically evaluated and optimized under different environmental loading conditions using the commercial software OrcaFlex. The results of the study show that the implementation of a shared mooring solution maximizes the cost-effectiveness economically while technically ensuring stable operation.

Graph-Based Semi-Supervised Learning with Bipartite Graph for Large-Scale Data and Prediction of Unseen Data

Recently, considerable attention has been directed toward graph-based semi-supervised learning (GSSL) as an effective approach for data labeling. Despite the progress achieved by current methodologies, several limitations persist. Firstly, many studies treat all samples equally in terms of weight and influence, disregarding the potential increased importance of samples near decision boundaries. Secondly, the detection of outlier-labeled data is crucial, as it can significantly impact model performance. Thirdly, existing models often struggle with predicting labels for unseen test data, restricting their utility in practical applications. Lastly, most graph-based algorithms rely on affinity matrices that capture pairwise similarities across all data points, thus limiting their scalability to large-scale databases. In this paper, we propose a novel GSSL algorithm tailored for large-scale databases, leveraging anchor points to mitigate the challenges posed by large affinity matrices. Additionally, our method enhances the influence of nodes near decision boundaries by assigning different weights based on their importance and using a mapping function from feature space to label space. Leveraging this mapping function enables direct label prediction for test samples without requiring iterative learning processes. Experimental evaluations on two extensive datasets (Norb and Covtype) demonstrate that our approach is scalable and outperforms existing GSSL methods in terms of performance metrics.

Peculiar k-mer Spectra Are Correlated with 3D Contact Frequencies and Breakpoint Regions in the Human Genome.

It is widely accepted that the 3D chromatin organization in human cell nuclei is not random and recent investigations point towards an interactive relation of epigenetic functioning and chromatin (re-)organization. Although chromatin organization seems to be the result of self-organization of the entirety of all molecules available in the cell nucleus, a general question remains open as to what extent chromatin organization might additionally be predetermined by the DNA sequence and, if so, if there are characteristic differences that distinguish typical regions involved in dysfunction-related aberrations from normal ones, since typical DNA breakpoint regions involved in disease-related chromosome aberrations are not randomly distributed along the DNA sequence. Highly conserved k-mer patterns in intronic and intergenic regions have been reported in eukaryotic genomes. In this article, we search and analyze regions deviating from average spectra (ReDFAS) of k-mer word frequencies in the human genome. This includes all assembled regions, e.g., telomeric, centromeric, genic as well as intergenic regions. A positive correlation between k-mer spectra and 3D contact frequencies, obtained exemplarily from given Hi-C datasets, has been found indicating a relation of ReDFAS to chromatin organization and interactions. We also searched and found correlations of known functional annotations, e.g., genes correlating with ReDFAS. Selected regions known to contain typical breakpoints on chromosomes 9 and 5 that are involved in cancer-related chromosomal aberrations appear to be enriched in ReDFAS. Since transposable elements like ALUs are often assigned as major players in 3D genome organization, we also studied their impact on our examples but could not find a correlation between ALU regions and breakpoints comparable to ReDFAS. Our findings might show that ReDFAS are associated with instable regions of the genome and regions with many chromatin contacts which is in line with current research indicating that chromatin loop anchor points lead to genomic instability.

Multi-view clustering with adaptive anchor and bipartite graph learning

Multi-view subspace clustering optimizes and integrates the graph structure information of each view. At present, the subspace clustering methods based on the abstract graph have better performance and improve the clustering results. Although the existing clustering algorithms have achieved excellent results, there are still four deficiencies: 1) Expensive time overhead, with most algorithms having a high time complexity; 2) Using fixed anchor points and the separation of anchor graph learning from subsequent graph construction, resulting in inadequate use of underlying graph structure between views; 3) Multi-view data have inevitable noise and in the original high-dimensional space data fusion may cause the loss of important information; 4) Most algorithms require additional post-processing to obtain clustering labels. To solve the above problems, we innovate a novel anchor-adaptive multi-view clustering algorithm based on a bipartite graph . Specifically, anchor graph learning and subspace graph construction are adaptively combined into a unified optimization framework. The projection matrix, consensus anchor matrix, and similarity matrix optimize each other to promote the clustering effect and structure a constrained bipartite graph to describe the relationship between representative points and sample points. At the same time, connectivity constraints are used to ensure that the connected components represent clusters directly. Our method has linear time complexity about sample size. Comparison experiments with the state-of-the-art algorithms demonstrate the effectiveness of the proposed method on various benchmark datasets. Moreover, our method is robust to noisy data and suitable for large-scale datasets.