Article
Computer Science, Artificial Intelligence
Denis Coquenet, Clement Chatelain, Thierry Paquet
Summary: This paper proposes an end-to-end segmentation-free network model called Document Attention Network for handwritten document recognition, which labels text parts and sequentially outputs characters and logical layout tokens. The model achieves competitive results on the READ 2016 dataset and performs well on the RIMES 2009 dataset.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Computer Science, Information Systems
Dian Ding, Lanqing Yang, Yi-Chao Chen, Guangtao Xue
Summary: The small size of mobile touch screens, such as smartphones and watches, greatly hinders the efficiency of human-computer interaction. This has led to a growing interest in handwriting recognition systems, which can be categorized into active and passive systems. Active systems require additional hardware devices or have insufficient tracking accuracy for handwriting recognition. Passive methods use acoustic signals but are susceptible to environmental noise. This paper presents a novel handwriting recognition system based on vibration signals detected by the built-in accelerometer of smartphones. The system achieves high resistance to interferences and demonstrates promising accuracy in various writing positions and environmental conditions.
IEEE TRANSACTIONS ON MOBILE COMPUTING
(2023)
Article
Computer Science, Information Systems
Xiao-Long Yun, Yan-Ming Zhang, Fei Yin, Cheng-Lin Liu
Summary: This paper proposes an online handwritten diagram recognition method based on graph neural networks, which tackles symbol segmentation and symbol recognition problems simultaneously under a unified learning framework. The experimental results show that the proposed method consistently outperforms previous methods on multiple datasets, and a large-scale annotated online handwritten flowchart dataset is released.
IEEE TRANSACTIONS ON MULTIMEDIA
(2022)
Article
Computer Science, Information Systems
Xiangping Wu, Qingcai Chen, Yulun Xiao, Wei Li, Xin Liu, Baotian Hu
Summary: In this paper, an efficient semantic segmentation model LCSegNet based on label coding (LC) is proposed for recognizing large-scale Chinese characters, achieving state-of-the-art performances in both complex scene and handwritten character recognition tasks. The method utilizes a new label coding method called Wubi-CRF and a conditional random field (CRF) module to learn constraint rules, significantly improving the accuracy of Chinese character recognition in scene text recognition tasks.
IEEE TRANSACTIONS ON MULTIMEDIA
(2021)
Article
Computer Science, Artificial Intelligence
Denis Coquenet, Clement Chatelain, Thierry Paquet
Summary: This paper proposes a unified end-to-end model using hybrid attention for unconstrained handwritten text recognition. By iteratively processing a paragraph image line by line, the model can generate vertical weighted masks to implicitly segment lines and recognize the character sequence of the whole paragraph.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Computer Science, Information Systems
Haitham Q. Ghadhban, Muhaini Othman, Noor Samsudin, Shahreen Kasim, Aisyah Mohamed, Yazan Aljeroudi
Summary: Deep learning has made progress in handwriting recognition, but still requires large amounts of data and computation. Handcrafted features are crucial for specific language types, with Segments Interpolation (SI) proposed for handwriting feature extraction in this study.
Article
Computer Science, Information Systems
Marcin Adamski, Kacper Sarnacki, Khalid Saeed
Summary: This paper proposes a technique for processing handwriting images with an improved algorithm for binarisation process, achieving significant accuracy improvement and reducing unwanted artefacts compared to standard approaches.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Artificial Intelligence
Sukhdeep Singh, Anuj Sharma, Vinod Kumar Chauhan
Summary: Handwriting recognition has been actively researched in the field of pattern recognition and machine learning. It has various applications and has gained momentum for Indic scripts recognition due to planned funding by the Indian government. This study provides an inclusive survey report of recent advances in Indic script offline handwriting recognition, focusing on character, digit, and word recognition, and suggests a great opportunity for further research in this emerging area.
ARTIFICIAL INTELLIGENCE REVIEW
(2023)
Article
Agronomy
Chao Chen, Shanlin Yi, Jinyi Mao, Feng Wang, Baofeng Zhang, Fuxin Du
Summary: This paper proposes a watershed-based segmentation recognition algorithm for accurate recognition of Agaricus bisporus. Tests show that the average correct recognition rate of the proposed algorithm is 95.7%, with low measurement errors and processing time, outperforming the current Circle Hough Transform implementation. It provides a sound basis for subsequent research on mechanized harvesting equipment for A. bisporus.
Article
Computer Science, Information Systems
C. Vinotheni, S. Lakshmana Pandian
Summary: This paper proposes an end-to-end deep learning model for Tamil handwritten document recognition. The model utilizes segmentation at the word and line levels, deep convolutional neural network, and water strider optimization algorithm to achieve real-time recognition with high accuracy and precision.
Article
Computer Science, Information Systems
Husam Ahmed Al Hamad, Laith Abualigah, Mohammad Shehab, Khalil H. A. Al-Shqeerat, Mohammad Otair
Summary: This paper proposes a new segmentation technique called ILDT for Arabic handwritten scripts, which uses vertical linear density to determine character boundaries and districting. Experimental results show that the proposed method outperforms other comparative methods in terms of segmentation and recognition accuracy.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Palaiahnakote Shivakumara, Tanmay Jain, Umapada Pal, Nitish Surana, Apostolos Antonacopoulos, Tong Lu
Summary: This paper proposes a text line segmentation approach that can be applied to both cleanly-written and struck-out text. The approach consists of three steps to progressively form text lines, and experiments show its superior performance compared to existing methods.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Omar Krichen, Simon Corbille, Eric Anquetil, Nathalie Girard, Elisa Fromont, Pauline Nerdeux
Summary: This study aims to improve the performance of children handwriting analysis by combining a handwriting analysis engine with deep learning word recognition methods. The analysis process is guided by prior knowledge and predictions from deep networks.
INTERNATIONAL JOURNAL ON DOCUMENT ANALYSIS AND RECOGNITION
(2022)
Article
Computer Science, Artificial Intelligence
Nishatul Majid, Elisa H. Barney Smith
Summary: This paper presents a framework for offline handwriting recognition using character spotting and autonomous tagging. The proposed approach provides a simple and powerful workflow that can be adjusted for different alphabetic scripts. Experimental results show high accuracy in recognizing handwritten characters in Bangla and Hangul/Korean.
INTERNATIONAL JOURNAL ON DOCUMENT ANALYSIS AND RECOGNITION
(2022)
Article
Environmental Sciences
Fayez Tarsha Kurdi, Zahra Gharineiat, Glenn Campbell, Mohammad Awrangjeb, Emon Kumar Dey
Summary: This paper presents a new algorithm for automatic building point cloud filtering based on the Z coordinate histogram, dividing the building point cloud into three zones and recognizing high tree crown points by analyzing normal vectors and the curvature factor. The suggested approach was tested on five datasets with different point densities and urban typology, showing high efficacy with accuracy values of 97.9%, 97.6%, and 95.6%.
Article
Computer Science, Artificial Intelligence
Mohammad Esmaeilpour, Patrick Cardinal, Alessandro Lameiras Koerich
APPLIED SOFT COMPUTING
(2020)
Review
Computer Science, Information Systems
Jonathan de Matos, Steve Tsham Mpinda Ataky, Alceu de Souza Britto, Luiz Eduardo Soares de Oliveira, Alessandro Lameiras Koerich
Summary: This paper reviews machine learning methods for histopathological image analysis, including shallow and deep learning methods, covering common tasks and datasets used in HI research.
Article
Computer Science, Artificial Intelligence
Bernardo B. Gatto, Eulanda M. dos Santos, Alessandro L. Koerich, Kazuhiro Fukui, Waldir S. S. Junior
Summary: This paper introduces a new method for multi-dimensional data classification, utilizing tensor representation and subspace concept to enhance classification accuracy. The use of generalized difference subspace (GDS) and n-mode GDS for data dimensionality reduction and discriminative feature extraction, along with the introduction of n-mode Fisher score and an improved metric based on geodesic distance for better tensor data classification performance.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Voncarlos M. Araujo, Alceu S. Britto Jr, Luiz S. Oliveira, Alessandro L. Koerich
Summary: This study proposed a novel method based on a two-view leaf image representation and a hierarchical classification strategy for fine-grained plant species recognition, achieving effective results in identifying plant genus and species by using botanical taxonomy as a basis.
Article
Chemistry, Analytical
Lucas O. Teixeira, Rodolfo M. Pereira, Diego Bertolini, Luiz S. Oliveira, Loris Nanni, George D. C. Cavalcanti, Yandre M. G. Costa
Summary: The study demonstrated the impact of lung segmentation in COVID-19 identification using CXR images, achieving good Jaccard distance and Dice coefficient for segmentation. It investigated the generalization of COVID-19 from images created from different sources, finding a strong bias introduced by underlying factors from different sources even after segmentation.
Article
Chemistry, Multidisciplinary
Thomas Teixeira, Eric Granger, Alessandro Lameiras Koerich
Summary: This paper investigates the use of deep learning architectures for continuous emotion recognition, extending 2D CNN models to learn spatiotemporal information from videos. Experimental results on the SEWA-DB dataset show that these architectures can effectively encode spatiotemporal information and achieve state-of-the-art results.
APPLIED SCIENCES-BASEL
(2021)
Article
Computer Science, Artificial Intelligence
Steve Tsham Mpinda Ataky, Alessandro Lameiras Koerich
Summary: This paper proposes a novel approach to quantifying complex systems of diverse patterns in texture, using species diversity, richness, and taxonomic distinctiveness. The method takes advantage of ecological patterns' invariance to build a permutation, rotation, and translation invariant descriptor. Experimental results show the advantages of this method.
PATTERN RECOGNITION
(2022)
Review
Chemistry, Analytical
Yandre M. G. Costa, Sergio A. Silva, Lucas O. Teixeira, Rodolfo M. Pereira, Diego Bertolini, Alceu S. Britto Jr, Luiz S. Oliveira, George D. C. Cavalcanti
Summary: This article reviews the top-100 most cited papers in the field of COVID-19 detection from thoracic medical imaging, discussing important aspects such as the type of medical imaging explored, learning settings, segmentation strategy, explainable artificial intelligence (XAI), and the availability of datasets and codes.
Article
Computer Science, Artificial Intelligence
Thiago M. Paixao, Rodrigo F. Berriel, Maria C. S. Boeres, Alessandro L. Koerich, Claudine Badue, Alberto F. De Souza, Thiago Oliveira-Santos
Summary: Advances in machine learning, especially deep learning, have improved the accuracy of automatically reconstructing shredded documents. However, there is still room for improvement in fully automatic reconstruction. To address this issue, we propose a human-in-the-loop reconstruction framework that allows users to verify the adjacency of adjacent shreds in the solution. Introducing human involvement can reduce errors by over 40%.
PATTERN RECOGNITION LETTERS
(2022)
Article
Computer Science, Theory & Methods
Mohammad Esmaeilpour, Patrick Cardinal, Alessandro Lameiras Koerich
Summary: This paper introduces a defense approach against adversarial attacks on speech-to-text systems. The proposed algorithm utilizes short-time Fourier transform, spectrogram subspace projection, and a novel GAN architecture trained with Sobolev integral probability metric. Experimental results demonstrate that it outperforms other defense algorithms in terms of accuracy and signal quality.
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
(2022)
Proceedings Paper
Acoustics
Mohammad Esmaeilpour, Patrick Cardinal, Alessandro Lameiras Koerich
Summary: This paper introduces a novel defense approach against end-to-end adversarial attacks by finding the optimal input vector through minimizing the relative chordal distance adjustment and reconstructing the signal. Experimental results show that this approach significantly outperforms conventional defense algorithms.
2021 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP 2021)
(2021)
Article
Engineering, Electrical & Electronic
Mohammad Esmaeilpour, Patrick Cardinal, Alessandro Lameiras Koerich
Summary: This letter introduces a new defense approach utilizing a cyclic generative adversarial network to reconstruct signals for countering state-of-the-art white and black-box adversarial attack algorithms. Experimental results show the effectiveness of this defense method in various adversarial attack scenarios.
IEEE SIGNAL PROCESSING LETTERS
(2021)
Article
Computer Science, Artificial Intelligence
Thiago M. Paixao, Rodrigo F. Berriel, Maria C. S. Boeres, Alessandro L. Koerich, Claudine Badue, Alberto F. De Souza, Thiago Oliveira-Santos
PATTERN RECOGNITION
(2020)
Article
Computer Science, Theory & Methods
Mohammad Esmaeilpour, Patrick Cardinal, Alessandro Koerich
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
(2020)
Article
Computer Science, Artificial Intelligence
Guang-Hai Liu, Zuo-Yong Li, Jing-Yu Yang, David Zhang
Summary: This article introduces a novel image retrieval method that improves retrieval performance by using sublimated deep features. The method incorporates orientation-selective features and color perceptual features, effectively mimicking these mechanisms to provide a more discriminating representation.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Fengguang Peng, Zihan Ding, Ziming Chen, Gang Wang, Tianrui Hui, Si Liu, Hang Shi
Summary: RGB-Thermal (RGB-T) semantic segmentation is an emerging task that aims to improve the robustness of segmentation methods under extreme imaging conditions by using thermal infrared modality. The challenges of foreground-background distinguishment and complementary information mining are addressed by proposing a cross modulation process with two collaborative components. Experimental results show that the proposed method achieves state-of-the-art performances on current RGB-T segmentation benchmarks.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Baihong Han, Xiaoyan Jiang, Zhijun Fang, Hamido Fujita, Yongbin Gao
Summary: This paper proposes a novel automatic prompt generation method called F-SCP, which focuses on generating accurate prompts for low-accuracy classes and similar classes. Experimental results show that our approach outperforms state-of-the-art methods on six multi-domain datasets.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Huikai Liu, Ao Zhang, Wenqian Zhu, Bin Fu, Bingjian Ding, Shengwu Xiong
Summary: Adverse weather conditions present challenges for computer vision tasks, and image de-weathering is an important component of image restoration. This paper proposes a multi-patch skip-forward structure and a Residual Deformable Convolutional module to improve feature extraction and pixel-wise reconstruction.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Oliver M. Crook, Mihai Cucuringu, Tim Hurst, Carola-Bibiane Schonlieb, Matthew Thorpe, Konstantinos C. Zygalakis
Summary: The transportation LP distance (TLP) is a generalization of the Wasserstein WP distance that can be applied directly to color or multi-channelled images, as well as multivariate time-series. TLP interprets signals as functions, while WP interprets signals as measures. Although both distances are powerful tools in modeling data with spatial or temporal perturbations, their computational cost can be prohibitively high for moderate pattern recognition tasks. The linear Wasserstein distance offers a method for projecting signals into a Euclidean space, and in this study, we propose linear versions of the TLP distance (LTLP) that show significant improvement over the linear WP distance in signal processing tasks while being several orders of magnitude faster to compute than the TLP distance.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Haitao Tian, Shiru Qu, Pierre Payeur
Summary: This paper proposes a method of target-dependent classifier, which optimizes the joint hypothesis of domain adaptation into a target-dependent hypothesis that better fits with the target domain clusters through an unsupervised fine-tuning strategy and the concept of meta-learning. Experimental results demonstrate that this method outperforms existing techniques in synthetic-to-real adaptation and cross-city adaptation benchmarks.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Qingsen Yan, Axi Niu, Chaoqun Wang, Wei Dong, Marcin Wozniak, Yanning Zhang
Summary: Deep learning-based methods have achieved remarkable results in the field of super-resolution. However, the limitation of paired training image sets has led researchers to explore self-supervised learning. However, the assumption of inaccurate downscaling kernel functions often leads to degraded results. To address this issue, this paper introduces KGSR, a kernel-guided network that trains both upscaling and downscaling networks to generate high-quality high-resolution images even without knowing the actual downscaling process.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Yifan Chen, Xuelong Li
Summary: Gait recognition is a popular technology for identification due to its ability to capture gait features over long distances without cooperation. However, current methods face challenges as they use a single network to extract both temporal and spatial features. To solve this problem, we propose a two-branch network that focuses on spatial and temporal feature extraction separately. By combining these features, we can effectively learn the spatio-temporal information of gait sequences.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Wei Shi, Wentao Zhang, Wei-shi Zheng, Ruixuan Wang
Summary: This article proposes a simple yet effective visualization framework called PAMI, which does not require detailed model structure and parameters to obtain visualization results. It can be applied to various prediction tasks with different model backbones and input formats.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Xiaobo Hu, Jianbo Su, Jun Zhang
Summary: This paper reviews the latest technologies in pattern recognition, highlighting their instabilities and failures in practical applications. From a control perspective, the significance of disturbance rejection in pattern recognition is discussed, and the existing problems are summarized. Finally, potential solutions related to the application of compensation on features are discussed to emphasize future research directions.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Andres Felipe Posada-Moreno, Nikita Surya, Sebastian Trimpe
Summary: Convolutional neural networks are widely used in critical systems, and explainable artificial intelligence has proposed methods for generating high-level explanations. However, these methods lack the ability to determine the location of concepts. To address this, we propose a novel method for automatic concept extraction and localization based on pixel-wise aggregations, and validate it using synthetic datasets.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Peng Bao, Jianian Li, Rong Yan, Zhongyi Liu
Summary: In this paper, a novel Dynamic Graph Contrastive Learning framework, DyGCL, is proposed to capture the temporal consistency in dynamic graphs and achieve good performance in node representation learning.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Kristian Schultz, Saptarshi Bej, Waldemar Hahn, Markus Wolfien, Prashant Srivastava, Olaf Wolkenhauer
Summary: Research indicates that deep generative models perform poorly compared to linear interpolation-based methods for synthetic data generation on small, imbalanced tabular datasets. To address this, a new approach called ConvGeN, combining convex space learning with deep generative models, has been proposed. ConvGeN improves imbalanced classification on small datasets while remaining competitive with existing linear interpolation methods.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Khondaker Tasrif Noor, Antonio Robles-Kelly
Summary: In this paper, the authors propose H-CapsNet, a capsule network designed for hierarchical image classification. The network effectively captures hierarchical relationships using dedicated capsules for each class hierarchy. A modified hinge loss is utilized to enforce consistency among the involved hierarchies. Additionally, a strategy for dynamically adjusting training parameters is presented to achieve better balance between the class hierarchies. Experimental results demonstrate that H-CapsNet outperforms competing hierarchical classification networks.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Lei Liu, Guorun Li, Yuefeng Du, Xiaoyu Li, Xiuheng Wu, Zhi Qiao, Tianyi Wang
Summary: This study proposes a new agricultural image segmentation model called CS-Net, which uses Simple-Attention Block and Simpleformer to improve accuracy and inference speed, and addresses the issue of performance collapse of Transformers in agricultural image processing.
PATTERN RECOGNITION
(2024)
