Pattern Recognition最新文献

Spectral analysis on 3D orthogonal moments for effective terrain matching 有效地形匹配的三维正交矩谱分析

IF 7.6 1区计算机科学

Pattern Recognition Pub Date : 2025-07-26 DOI: 10.1016/j.patcog.2025.112212

Junjie Zhou , Quan Hu , Kedong Wang , Jinling Wang

引用次数: 0

One-step late fusion multiple kernel clustering via tensorized adaptive multi-scale partition fusion 基于张张化自适应多尺度分区融合的一步延迟融合多核聚类

IF 7.6 1区计算机科学

Pattern Recognition Pub Date : 2025-07-26 DOI: 10.1016/j.patcog.2025.112220

Xinrui Lu

{"title":"One-step late fusion multiple kernel clustering via tensorized adaptive multi-scale partition fusion","authors":"Xinrui Lu","doi":"10.1016/j.patcog.2025.112220","DOIUrl":"10.1016/j.patcog.2025.112220","url":null,"abstract":"<div><div>Recent years, multiple kernel clustering (MKC) has extensive applications in data analysis field. Due to its characteristic of low computational complexity, Late Fusion Multiple Kernel Clustering (LFMKC) stands out among existing MKC algorithm methods. However, LFMKC is still confronted with some problems. Firstly, LFMKC only uses single-scale partition to obtain the clustering results, failing to learn about the information in the kernel matrices thoroughly. Secondly, LFMKC cannot utilize the higher-order correlations since the base partitions are fixed. Besides, LFMKC need an extra <em>k</em>-means step to yield the result. To overcome these limitations, we design a new method named One-step Late fusion Multiple kernel clustering via Tensorized adaptive Multi-scale partition Fusion (OLMTMF), which integrates multi-scale fusion, high-order tensor information and spectral rotation (SR) into a unified framework. To be more precise, we first design an adaptive algorithm to integrate multi-scale partition instead of single partition, fully exploring the information in different kernel matrices. Next, OLMTMF builds a tensor with different fused partitions, constrained by Tensor Nuclear Norm (TNN), to discover the higher-order correlations among these fused partitions. Finally, OLMTMF utilizes SR to directly obtain clustering results, further improving the clustering performance. We also develop an alternative procedure with theoretical convergence guarantee to optimize the objective function of OLMTMF. Extensive experiments indicate that OLMTMF achieves excellent clustering performance on different datasets with low computational complexity. The source code can be downloaded from: <span><span>https://github.com/luxinrui018/OLMTMF</span><svg><path></path></svg></span> .</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"171 ","pages":"Article 112220"},"PeriodicalIF":7.6,"publicationDate":"2025-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144737953","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A general dual-view framework for instance weighted naive Bayes 一个通用的双视图框架，例如加权朴素贝叶斯

IF 7.6 1区计算机科学

Pattern Recognition Pub Date : 2025-07-25 DOI: 10.1016/j.patcog.2025.112181

Huan Zhang , Kexin Meng , Pei Lv , Shuo He , Mingliang Xu

{"title":"A general dual-view framework for instance weighted naive Bayes","authors":"Huan Zhang , Kexin Meng , Pei Lv , Shuo He , Mingliang Xu","doi":"10.1016/j.patcog.2025.112181","DOIUrl":"10.1016/j.patcog.2025.112181","url":null,"abstract":"<div><div>Instance weighting is an effective and flexible method to alleviate the attribute conditional independence assumption in naive Bayes (NB). However, existing instance weighting methods mainly focus on how to learn a specific weight for each instance, ignoring the limitation of the original view. In this study, we argue that real-world applications are rather complicated, and it is sub-optimal to learn instance weights only using the original view. Based on this premise, we propose a novel general framework called dual-view instance weighted naive Bayes (DIWNB). In DIWNB, we first construct multiple K-nearest neighbor (KNN) classifiers and select those with the lowest error rate to classify each training instance in turn to build the generated view. Next, we learn a specific weight for each training instance, and build an instance weighted NB model in each view. Finally, we weightedly fuse the class-membership probabilities of dual views to predict the class label for each test instance. To construct the generated view, we design a hard label approach and a soft label approach, and thus two different versions are created, which we denote as DIWNB<span><math><msup><mrow></mrow><mrow><mi>H</mi></mrow></msup></math></span> and DIWNB<span><math><msup><mrow></mrow><mrow><mi>S</mi></mrow></msup></math></span>, respectively. Experimental results on 60 benchmark and 2 real-world datasets demonstrate the effectiveness of DIWNB.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"171 ","pages":"Article 112181"},"PeriodicalIF":7.6,"publicationDate":"2025-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144723676","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multi-branch perturbation learning with constraint simulation for semi-supervised semantic segmentation 半监督语义分割的约束仿真多分支摄动学习

IF 7.6 1区计算机科学

Pattern Recognition Pub Date : 2025-07-25 DOI: 10.1016/j.patcog.2025.112200

Ruyu Liu , Feng Xiao , Jianhua Zhang , Xiufeng Liu , Xu Cheng , Shengyong Chen , Bo Sun , Houxiang Zhang

{"title":"Multi-branch perturbation learning with constraint simulation for semi-supervised semantic segmentation","authors":"Ruyu Liu , Feng Xiao , Jianhua Zhang , Xiufeng Liu , Xu Cheng , Shengyong Chen , Bo Sun , Houxiang Zhang","doi":"10.1016/j.patcog.2025.112200","DOIUrl":"10.1016/j.patcog.2025.112200","url":null,"abstract":"<div><div>Current semi-supervised semantic segmentation (SSS) methods improve generalization via weak-to-strong pseudo-supervision with image perturbations. However, many methods are limited by employing a single perturbation mode and a specific weak-to-strong learning strategy, restricting exploration of the perturbation space and hindering performance in fine-grained segmentation. While diverse perturbations are intuitively beneficial, simply combining them can lead to inefficient optimization and instability. In this paper, we propose a multi-branch strong perturbation constraint learning framework for SSS. Our framework introduces a novel multi-branch perturbation learning (MSPL) strategy, employing multiple parallel branches with diverse strong augmentations to expand the perturbation space and capture complex semantic variations. We further design a novel constraint simulation loss (CSSL), based on a hierarchical consistency learning structure (weak-to-strong and strong-to-strong), which enforces strong-to-strong consistency between different perturbation branches. CSSL mitigates instability and enhances robustness to perturbation-induced noise, enabling the network to better generalize and achieve more accurate segmentation, especially for fine object boundaries. Extensive evaluations on benchmark datasets (PASCAL VOC 2012, Cityscapes, COCO) demonstrate that our method achieves state-of-the-art performance. Ablation studies further validate the effectiveness of our proposed MSPL and CSSL components.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"171 ","pages":"Article 112200"},"PeriodicalIF":7.6,"publicationDate":"2025-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144737955","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Corrigendum to “MA-FSAR: Multimodal Adaptation of CLIP for few-shot action recognition” [Pattern Recognition 169 (2026) 111902] “MA-FSAR: CLIP的多模态适应用于少量动作识别”的勘误表[模式识别169 (2026)111902]

IF 7.5 1区计算机科学

Pattern Recognition Pub Date : 2025-07-24 DOI: 10.1016/j.patcog.2025.112160

Jiazheng Xing , Jian Zhao , Chao Xu , Mengmeng Wang , Guang Dai , Yong Liu , Jingdong Wang , Xuelong Li

引用次数: 0

Self-supervised learning video anomaly detection based on time interval prediction and noise classification 基于时间间隔预测和噪声分类的自监督学习视频异常检测

IF 7.6 1区计算机科学

Pattern Recognition Pub Date : 2025-07-24 DOI: 10.1016/j.patcog.2025.112198

Yishuo Liu , Chuanxu Wang , Qingyang Yang , Lanxiao Li , Binghui Wang

{"title":"Self-supervised learning video anomaly detection based on time interval prediction and noise classification","authors":"Yishuo Liu , Chuanxu Wang , Qingyang Yang , Lanxiao Li , Binghui Wang","doi":"10.1016/j.patcog.2025.112198","DOIUrl":"10.1016/j.patcog.2025.112198","url":null,"abstract":"<div><div>Video Anomaly Detection (VAD) aims to automatically identify anomalous events in videos that significantly deviate from normal behavioral patterns. Self-supervised learning motivates models to learn effective features from unlabeled data by designing proxy tasks. However, existing approaches often rely on coarse-grained modeling, focusing mainly on global sequence order or holistic scene structures, which may limit their ability to capture subtle motion changes or localized anomalies. Therefore, this paper proposes a self-supervised learning framework combined with fine-grained spatio-temporal proxy tasks to extract key features more accurately. For the temporal branch, we design a time interval prediction task: given a fixed middle frame and randomly sampled frames from both sides, the model predicts their temporal intervals relative to the center frame, thereby modeling the dynamic patterns of behavior. To enhance temporal modeling capabilities, we introduce a multi-head self-attention mechanism to capture inter-frame dependencies in the input sequence. The spatial branch employs a noise classification task inspired by diffusion models, where varying levels of noise are added to image patches, and the model predicts the corresponding noise levels. This encourages learning of local appearance features and patch-level sensitivity to perturbations. Our method is trained in an end-to-end manner and does not rely on pre-trained models. Experiments on three benchmark datasets demonstrate stable performance: the method achieves AUC scores of 98.6 % on UCSD Ped2, 91.7 % on CUHK Avenue, and 83.7 % on ShanghaiTech. These results suggest that the proposed approach can generalize well across different scenes, perspectives, and types of anomalous behavior.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"171 ","pages":"Article 112198"},"PeriodicalIF":7.6,"publicationDate":"2025-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144723677","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

GrEp: Graph-based epithelial cell classification refinement in histopathology H&E images GrEp：组织病理学H&E图像中基于图的上皮细胞分类改进

IF 7.6 1区计算机科学

Pattern Recognition Pub Date : 2025-07-24 DOI: 10.1016/j.patcog.2025.112197

Ana Leni Frei , Javier Garcia-Baroja , Tilman Rau , Christina Neppl , Alessandro Lugli , Wiebke Solass , Martin Wartenberg , Andreas Fischer , Inti Zlobec

{"title":"GrEp: Graph-based epithelial cell classification refinement in histopathology H&E images","authors":"Ana Leni Frei , Javier Garcia-Baroja , Tilman Rau , Christina Neppl , Alessandro Lugli , Wiebke Solass , Martin Wartenberg , Andreas Fischer , Inti Zlobec","doi":"10.1016/j.patcog.2025.112197","DOIUrl":"10.1016/j.patcog.2025.112197","url":null,"abstract":"<div><div>The automatic cell segmentation and classification from whole slide images plays an important role in digital pathology, unlocking new opportunities for biomarker discovery. Despite extensive research, this task faces persistent challenges such as the differentiation of epithelial cells into normal and malignant. Many existing models lack reporting of epithelial subtyping, and when available, their performance is often suboptimal. This work benchmarks state-of-the-art methods to highlight this limitation and introduces GrEp, a geometric deep learning strategy that considers the broader epithelium tissue architecture to infer cell-level classification rather than relying exclusively on nuclei morphology. The proposed graph-based workflow significantly outperformed state-of-the-art nuclei classification models in colorectal cancer and generalized effectively to two unseen tissue types, endometrium and pancreas, proving the robustness of the geometry-based model. Given its speed and accuracy, we believe GrEp to be a valuable method to refine epithelial cell classification for downstream analyses in clinical and research settings.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"171 ","pages":"Article 112197"},"PeriodicalIF":7.6,"publicationDate":"2025-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144723678","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A coarse-fine matching method as the visual guidance of uniform allowances for robot grinding 以粗精匹配法作为机器人磨削均匀余量的视觉指导

IF 7.6 1区计算机科学

Pattern Recognition Pub Date : 2025-07-24 DOI: 10.1016/j.patcog.2025.112189

Jingyu Sun , Yadong Gong , Mingjun Liu , Xianli Zhao , Jibin Zhao

{"title":"A coarse-fine matching method as the visual guidance of uniform allowances for robot grinding","authors":"Jingyu Sun , Yadong Gong , Mingjun Liu , Xianli Zhao , Jibin Zhao","doi":"10.1016/j.patcog.2025.112189","DOIUrl":"10.1016/j.patcog.2025.112189","url":null,"abstract":"<div><div>Matching, as a critical component in the visual guidance of robotic abrasive belt grinding and polishing, enables the establishing a material allowance distribution model between the scanned data and design model of the aero blade workpiece. However, conventional matching methods yield models with poor consistency, which exacerbates the instability of the robotic grinding and polishing system. This paper constructs an objective function that minimizes the component distance under local Frenet frame. It introduces the constraint on uniform allowances about machining conditions. A space location for subsequent processing is found in solving the multi-objective optimization function. To speed up the iterative process, the method utilizes a cosine similarity vector and incorporates the double features of torsion and curvature to assist the similarity function in making judgments. The paper also examines the principle behind setting the similarity threshold. However, the initial position relationship must be sufficiently close for the target point cloud to reach the desired position during iterations. Otherwise, the matching calculation will converge to the local minimum. To address this issue, the paper proposes the objective constraint of minimizing the density entropy difference on the partitions after cylinder parametrization of the point cloud. It determines the initial spatial positional relationship within the controllable error. Through comparison with both state-of-the-art and classical matching algorithms, the proposed method demonstrates superior performance in efficiency, precision, convergence, and consistency of allowance distribution.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"171 ","pages":"Article 112189"},"PeriodicalIF":7.6,"publicationDate":"2025-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144723679","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Fast multi-view clustering via tensor hyperbolic tangent-p norm minimization 基于张量双曲切-p范数最小化的快速多视图聚类

IF 7.6 1区计算机科学

Pattern Recognition Pub Date : 2025-07-24 DOI: 10.1016/j.patcog.2025.112195

Yongbo Yu , Zhoumin Lu , Jingjing Xue , Rong Wang , Zongcheng Miao , Feiping Nie

{"title":"Fast multi-view clustering via tensor hyperbolic tangent-p norm minimization","authors":"Yongbo Yu , Zhoumin Lu , Jingjing Xue , Rong Wang , Zongcheng Miao , Feiping Nie","doi":"10.1016/j.patcog.2025.112195","DOIUrl":"10.1016/j.patcog.2025.112195","url":null,"abstract":"<div><div>Tensor-based multi-view clustering methods have gained significant attention due to their ability to directly capture high-order information, often outperforming matrix-based approaches. However, these methods face challenges in efficiently processing large-scale datasets due to their high computational complexity. Moreover, most existing tensor-based approaches rely on the tensor nuclear norm (TNN) to approximate the tensor rank function. However, TNN penalizes larger singular values, which are essential for preserving critical structural information, thus constraining the extraction of multi-view information. To address these challenges, we propose a novel fast multi-view clustering method via tensor hyperbolic tangent-<span><math><mi>p</mi></math></span> norm minimization. First, we incorporate an efficient anchor selection strategy and construct tensors from anchor-based representations, significantly reducing the computational burden of tensor-based approaches for large-scale datasets. Second, we introduce the tensor hyperbolic tangent-<span><math><mi>p</mi></math></span> norm (THT<span><math><msub><mrow></mrow><mi>p</mi></msub></math></span>N), a more robust and accurate approximation of the tensor rank function, enabling improved extraction of multi-view consistency and complementarity. Extensive experiments on eight real-world datasets show that our proposed model not only surpasses tensor-based methods in clustering performance but also outperforms matrix-based methods in computational efficiency, establishing a new benchmark for fast multi-view clustering. Code is available at <span><span>https://github.com/usualheart/FTHMC</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"171 ","pages":"Article 112195"},"PeriodicalIF":7.6,"publicationDate":"2025-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144723673","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Experimental evaluation of Szemerédi’s regularity lemma in graph-based clustering 基于图的聚类中szemersamedi正则引理的实验评价

IF 7.6 1区计算机科学

Pattern Recognition Pub Date : 2025-07-24 DOI: 10.1016/j.patcog.2025.112205

Jian Hou , Juntao Ge , Huaqiang Yuan , Marcello Pelillo

{"title":"Experimental evaluation of Szemerédi’s regularity lemma in graph-based clustering","authors":"Jian Hou , Juntao Ge , Huaqiang Yuan , Marcello Pelillo","doi":"10.1016/j.patcog.2025.112205","DOIUrl":"10.1016/j.patcog.2025.112205","url":null,"abstract":"<div><div>One major problem of graph-based clustering lies in the large computation load resulted by the similarity graph. Several previous works have shown that Szemerédi’s regularity lemma can be useful in relieving this problem. Based on this lemma, we partition the original graph to obtain a reduced graph, which inherits the major structure of the original graph with a much smaller cardinality. By performing clustering on the reduced graph and mapping data labels back to the original graph, the computation load can be reduced significantly. In further works we found that the parameters of this method have significant influences on the clustering results, and this issue hasn’t been dealt with in previous works. In this paper we present a thorough investigation of the influences of the parameters on clustering results, in experiments with four representative algorithms and a large number of real datasets. As a result, we find out the appropriate ranges of parameters to improve both clustering accuracy and computation efficiency significantly. We also show that regularity partitioning outperforms ordinary k-means-based partitioning, demonstrating the advantage of the regularity lemma in building the reduced graph. Furthermore, experimental results show that relatively old algorithms can be enhanced based on this lemma to outperform recent state-of-the-art ones. This work goes a step further in extending the application of the regularity lemma from pure theoretical to practical realms.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"171 ","pages":"Article 112205"},"PeriodicalIF":7.6,"publicationDate":"2025-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144723672","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0