Article
Computer Science, Artificial Intelligence
Adam Hammoumi, Maxime Moreaud, Christophe Ducottet, Sylvain Desroziers
Summary: Recent methods of image augmentation and prediction are based on deep learning paradigm, involving careful preparation of image dataset and selection of suitable network architecture. The proposed techniques, including adding structural information and using a patch-based procedure with stratified sampling, have been validated on two image datasets. Evaluation of the results using appropriate metrics demonstrates the reliability of the established framework for a wide range of applications.
Article
Mathematics
Ganbayar Batchuluun, Se Hyun Nam, Kang Ryoung Park
Summary: Previous research on plant image classification used various plant datasets, but faced difficulties due to small dataset sizes and limitations in constructing large-scale datasets. This study improved plant image classification performance by reducing training image numbers and then increasing them through augmentation methods.
Article
Computer Science, Artificial Intelligence
Shuanlong Niu, Yaru Peng, Bin Li, Yuanhong Qiu, Tongzhi Niu, Weifeng Li
Summary: Deep learning methods, like CNNs, are commonly used for surface defect segmentation in industries. However, overfitting and lack of generalization affect the performance of CNN-based models. To address this issue, we propose a plug-and-play data augmentation method that is specifically designed for CNN defect segmentation tasks. By occluding high-confidence regions and periodically updating confidence maps, our method effectively reduces overfitting and improves segmentation accuracy on different datasets.
JOURNAL OF INTELLIGENT MANUFACTURING
(2023)
Article
Environmental Sciences
Junlin Han, Mehrdad Shoeiby, Tim Malthus, Elizabeth Botha, Janet Anstee, Saeed Anwar, Ran Wei, Mohammad Ali Armin, Hongdong Li, Lars Petersson
Summary: Underwater image restoration is important for unveiling the underwater world. To support the development of new deep-learning based methods, we constructed a large-scale real underwater image dataset and proposed a method based on an unsupervised image-to-image translation framework.
Article
Engineering, Civil
Shasha Ren, Qiong Liu
Summary: This paper introduces a fast urban remote sensing (URS) image segmentation method based on a multi-layer pixel attention mechanism (MPAM). By designing feature enhancement modules and a data enhancement method, the accuracy and efficiency of URS image segmentation are improved.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Jiaqi Mi, Congcong Ma, Lihua Zheng, Man Zhang, Minzan Li, Minjuan Wang
Summary: The small-sample task in deep learning is challenging due to costly annotations and limitations in acquiring rare plant and animal images. Data augmentation is an effective method but needs improvement. This study proposes WGAN-CL, a Wasserstein GAN with confidence loss, to expand small-sample plant datasets. The experiments demonstrate the model's effectiveness and performance improvements over state-of-the-art technologies.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Paul Bergmann, Kilian Batzner, Michael Fauser, David Sattlegger, Carsten Steger
Summary: The detection of anomalous structures in natural image data is crucial for various tasks in the field of computer vision. Unsupervised anomaly detection methods require data for training and evaluating new approaches. The MVTec anomaly detection dataset includes high-resolution color images of different objects and textures, with normal and abnormal images for training and testing purposes.
INTERNATIONAL JOURNAL OF COMPUTER VISION
(2021)
Article
Plant Sciences
Lili Li, Bin Wang, Yanwen Li, Hua Yang
Summary: In this study, an accurate segmentation method for apple leaf disease spots based on DeepLabV3+ semantic segmentation network model was proposed. By extracting the features of apple leaf lesions, the recognition of apple leaf diseases and diagnosis of disease severity were effectively improved. The experimental results showed that the average pixel accuracy and average intersection over union of the model reached 97.26% and 83.85% respectively. After deploying the model on a smartphone platform, the detection time of the portable and intelligent apple leaf disease diagnostic system was 9 seconds per image. These research results can provide precise guidance for the prevention and precise control of apple diseases in fields.
Article
Multidisciplinary Sciences
Chin-Fu Liu, Richard Leigh, Brenda Johnson, Victor Urrutia, Johnny Hsu, Xin Xu, Xin Li, Susumu Mori, Argye E. Hillis, Andreia V. Faria
Summary: Extracting meaningful and reproducible models of brain function from stroke images is challenging due to the variability of lesion frequency and patterns. Large datasets and automated image post-processing tools are essential for analysis. We present a public dataset of 2,888 multimodal clinical MRIs of stroke patients, with manual lesion segmentation, to support the development of artificial intelligence tools for lesion analysis and related studies.
Article
Engineering, Electrical & Electronic
Zishu Gao, Guodong Yang, En Li, Zize Liang
Summary: The study proposes an improved detection network for small insulator defects with BN-CBAM and a feature fusion module, achieving better performance than other state-of-the-art approaches. Additionally, a data augmentation method based on the fusion of target segment and background is introduced to enhance network generalizability.
IEEE SENSORS JOURNAL
(2021)
Article
Computer Science, Artificial Intelligence
Yuxi Xie, Shaofan Li, C. T. Wu, Zhipeng Lai, Miao Su
Summary: In this paper, a novel Hypergraph Convolution Network (HCN-WDI) is proposed for automatic wafer defect identification. Experimental results have shown that the HCN-WDI model outperforms traditional image classifiers in terms of classification accuracy.
JOURNAL OF INTELLIGENT MANUFACTURING
(2022)
Article
Computer Science, Artificial Intelligence
Hongyu Wang, Dandan Zhang, Songtao Ding, Zhanyi Gao, Jun Feng, Shaohua Wan
Summary: In this paper, a novel rib segmentation framework based on Unpaired Sample Augmentation and Multi-Scale Network is proposed to improve the accuracy of ribs segmentation with limited labeled samples. The algorithm learns pneumonia-related texture changes and generates augmented samples. The deep separation module and comprehensive loss function achieve refined segmentation results for each organ. Experimental results demonstrate that the proposed algorithm outperforms other methods in the segmentation performance of overlapping and fuzzy regions of multiple organs.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Automation & Control Systems
Chang Chen, Chen Lin, Zhen Meng, Jing Ni, Jiteng Sun, Zuji Li
Summary: This paper proposes a small sample end mill wear area segmentation method based on transfer learning and generative adversarial networks to address the issue of insufficient samples of end mill wear images. By using WGAN to generate wear images and employing transfer learning to improve the segmentation network's generalization ability, small sample training is achieved. This approach increases mPA by 4.46% and mIOU by 8.97% compared to merely using the semantic segmentation network for small sample training. The method not only has high stability and segmentation accuracy but also solves the problem of insufficient end mill wear image samples, making it effective for intelligent detection of tool wear state.
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2023)
Article
Agriculture, Multidisciplinary
Tingting Yang, Suyin Zhou, Zhijie Huang, Aijun Xu, Junhua Ye, Jianxin Yin
Summary: In this paper, a comprehensive urban street tree dataset is proposed, consisting of 41,467 high-resolution images of 50 tree species, with 22,872 annotated images. The dataset includes subdatasets for leaves, trunks, branches, flowers, and fruits, captured under various conditions. Various vision algorithms were evaluated for tree species identification and instance segmentation tasks.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2023)
Article
Computer Science, Artificial Intelligence
Ilija Domislovic, Donik Vrsnak, Marko Subasic, Sven Loncaric
Summary: In this paper, a new large-scale publicly available color constancy dataset called Shadows & Lumination dataset is introduced. The dataset consists of 2500 minimally processed images from various indoor, outdoor, and night-time scenes. Unlike other datasets, it includes real-world images with two illuminants for multi-illuminant estimation, along with binary segmentation masks dividing the image into regions illuminated by each illuminant. The paper also benchmarks illumination estimation methods and image segmentation methods using the dataset.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Rui Lv, Dingheng Wang, Jiangbin Zheng, Zhao-Xu Yang
Summary: In this paper, the authors investigate tensor decomposition for neural network compression. They analyze the convergence and precision of tensor mapping theory, validate the rationality of tensor mapping and its superiority over traditional tensor approximation based on the Lottery Ticket Hypothesis. They propose an efficient method called 3D-KCPNet to compress 3D convolutional neural networks using the Kronecker canonical polyadic (KCP) tensor decomposition. Experimental results show that 3D-KCPNet achieves higher accuracy compared to the original baseline model and the corresponding tensor approximation model.
Article
Computer Science, Artificial Intelligence
Xiangkun He, Zhongxu Hu, Haohan Yang, Chen Lv
Summary: In this paper, a novel constrained multi-objective reinforcement learning algorithm is proposed for personalized end-to-end robotic control with continuous actions. The approach trains a single model using constraint design and a comprehensive index to achieve optimal policies based on user-specified preferences.
Article
Computer Science, Artificial Intelligence
Zhijian Zhuo, Bilian Chen, Shenbao Yu, Langcai Cao
Summary: In this paper, a novel method called Expansion with Contraction Method for Overlapping Community Detection (ECOCD) is proposed, which utilizes non-negative matrix factorization to obtain disjoint communities and applies expansion and contraction processes to adjust the degree of overlap. ECOCD is applicable to various networks with different properties and achieves high-quality overlapping community detection.
Article
Computer Science, Artificial Intelligence
Yizhe Zhu, Chunhui Zhang, Jialin Gao, Xin Sun, Zihan Rui, Xi Zhou
Summary: In this work, the authors propose a Contrastive Spatio-Temporal Distilling (CSTD) approach to improve the detection of high-compressed deepfake videos. The approach leverages spatial-frequency cues and temporal-contrastive alignment to fully exploit spatiotemporal inconsistency information.
Review
Computer Science, Artificial Intelligence
Laijin Meng, Xinghao Jiang, Tanfeng Sun
Summary: This paper provides a review of coverless steganographic algorithms, including the development process, known contributions, and general issues in image and video algorithms. It also discusses the security of coverless steganography from theoretical analysis to actual investigation for the first time.
Article
Computer Science, Artificial Intelligence
Yajie Bao, Tianwei Xing, Xun Chen
Summary: Visual question answering requires processing multi-modal information and effective reasoning. Neural-symbolic learning is a promising method, but current approaches lack uncertainty handling and can only provide a single answer. To address this, we propose a confidence based neural-symbolic approach that evaluates NN inferences and conducts reasoning based on confidence.
Article
Computer Science, Artificial Intelligence
Anh H. Vo, Bao T. Nguyen
Summary: Interior style classification is an interesting problem with potential applications in both commercial and academic domains. This project proposes a method named ISC-DeIT, which combines data-efficient image transformer architectures and knowledge distillation, to address the interior style classification problem. Experimental results demonstrate a significant improvement in predictive accuracy compared to other state-of-the-art methods.
Article
Computer Science, Artificial Intelligence
Shashank Kotyan, Danilo Vasconcellos Vargas
Summary: This article introduces a novel augmentation technique called Dynamic Scanning Augmentation to improve the accuracy and robustness of Vision Transformer (ViT). The technique leverages dynamic input sequences to adaptively focus on different patches, resulting in significant changes in ViT's attention mechanism. Experimental results demonstrate that Dynamic Scanning Augmentation outperforms ViT in terms of both robustness to adversarial attacks and accuracy against natural images.
Article
Computer Science, Artificial Intelligence
Hiba Alqasir, Damien Muselet, Christophe Ducottet
Summary: The article proposes a solution to improve the learning process of a classification network by providing shape priors, reducing the need for annotated data. The solution is tested on cross-domain digit classification tasks and a video surveillance application.
Article
Computer Science, Artificial Intelligence
Dexiu Ma, Mei Liu, Mingsheng Shang
Summary: This paper proposes a method using neural dynamics solvers to solve infinity-norm optimization problems. Two improved solvers are constructed and their effectiveness and superiority are demonstrated through theoretical analysis and simulation experiments.
Article
Computer Science, Artificial Intelligence
Francesco Gregoretti, Giovanni Pezzulo, Domenico Maisto
Summary: Active Inference is a computational framework that uses probabilistic inference and variational free energy minimization to describe perception, planning, and action. cpp-AIF is a header-only C++ library that provides a powerful tool for implementing Active Inference for Partially Observable Markov Decision Processes through multi-core computing. It is cross-platform and improves performance, memory management, and usability compared to existing software.
Article
Computer Science, Artificial Intelligence
Zelin Ying, Dawei Cheng, Cen Chen, Xiang Li, Peng Zhu, Yifeng Luo, Yuqi Liang
Summary: This paper proposes a novel stock market trends prediction framework called SMART, which includes a self-supervised stock technical data sequence embedding model S3E. By training with multiple self-supervised auxiliary tasks, the model encodes stock technical data sequences into embeddings and uses the learned sequence embeddings for predicting stock market trends. Extensive experiments on China A-Shares market and NASDAQ market prove the high effectiveness of our model in stock market trends prediction, and its effectiveness is further validated in real-world applications in a leading financial service provider in China.
Article
Computer Science, Artificial Intelligence
Hao Li, Hao Jiang, Dongsheng Ye, Qiang Wang, Liang Du, Yuanyuan Zeng, Liu Yuan, Yingxue Wang, C. Chen
Summary: DHGAT1, a dynamic hyperbolic graph attention network, utilizes hyperbolic metric properties to embed dynamic graphs. It employs a spatiotemporal self-attention mechanism and weighted node representations, resulting in excellent performance in link prediction tasks.
Article
Computer Science, Artificial Intelligence
Jiehui Huang, Zhenchao Tang, Xuedong He, Jun Zhou, Defeng Zhou, Calvin Yu-Chian Chen
Summary: This study proposes a progressive learning multi-scale feature blending model for image deraining tasks. The model utilizes detail dilation and texture extraction to improve the restoration of rainy images. Experimental results show that the model achieves near state-of-the-art performance in rain removal tasks and exhibits better rain removal realism.
Article
Computer Science, Artificial Intelligence
Lizhi Liu, Zilin Gao, Yinhe Wang, Yongfu Li
Summary: This paper proposes a novel discrete-time interconnected model for depicting complex dynamical networks. The model consists of nodes and edges subsystems, which consider the dynamic characteristic of both nodes and edges. By designing control strategies and coupling modes, the stabilization and synchronization of the network are achieved. Simulation results demonstrate the effectiveness of the proposed methods.