Article
Computer Science, Information Systems
Nihat Arslan, Kali Gurkahraman
Summary: Parametric curves like Bezier and B-splines, originally developed for automobile body design, are now also used in image processing and computer vision. They can be used to reconstruct object shapes in images, including translations, scales, and rotations. These curves can be generated using a point set from the object's outer boundary.
CMC-COMPUTERS MATERIALS & CONTINUA
(2023)
Article
Mathematics
Muhammad Ammad, Md Yushalify Misro, Muhammad Abbas, Abdul Majeed
Summary: This paper introduces a new approach for the fabrication of generalized developable cubic trigonometric Bezier surfaces with shape parameters and discusses the influence of shape parameters on the surfaces.
Article
Mathematics, Applied
Wujie Liu, Xin Li
Summary: This paper proposes a method to construct a G(3) cubic spline curve from any given open control polygon by defining inner Bezier points and junction points to achieve G(2) continuity. The curvature combs and plots demonstrate the advantage of the G(3) cubic spline curve over the traditional C-2 cubic spline curve.
JOURNAL OF COMPUTATIONAL MATHEMATICS
(2021)
Article
Multidisciplinary Sciences
Chi Zhang, Arthur Porto, Sara Rolfe, Altan Kocatulum, A. Murat Maga
Summary: Manually collecting landmarks for complex morphological phenotypes can be tedious and prone to errors. We introduce a fast and open-source automated landmarking pipeline (MALPACA) that uses multiple templates to accommodate large-scale variations. Our results show that MALPACA outperforms single-template methods and we also provide a K-means method for template selection. MALPACA is an efficient and reproducible method that can handle large morphological variability.
Article
Biology
Sanguk Park, Minyoung Chung
Summary: The segmentation of cardiac structures in CT images is crucial for diagnosing cardiovascular diseases. This study introduces a novel model focusing on shape and boundary-aware features to improve accuracy between proximate organs. The proposed network outperforms state-of-the-art models by enhancing the attention on edges between substructures.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Computer Science, Artificial Intelligence
Yi Zhou, Guillermo Gallego, Xiuyuan Lu, Siqi Liu, Shaojie Shen
Summary: Identifying independently moving objects in dynamic scenes is a crucial task for scene understanding. Traditional cameras may face limitations such as motion blur or exposure artifacts, while event-based cameras offer advantages to overcome these limitations by reporting pixel-wise intensity changes asynchronously. This research develops a method to solve the event-based motion segmentation problem by casting it as an energy minimization problem involving fitting multiple motion models. The experiments demonstrate the versatility of the method in scenes with different motion patterns and number of moving objects, achieving state-of-the-art results without the need to determine the expected number of moving objects beforehand. The software and dataset are released under an open source license to encourage research in event-based motion segmentation.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Environmental Sciences
Jiabin Xv, Fei Deng
Summary: In this study, a 3D point cloud instance segmentation network considering the global shape contour was proposed. A Transformer module was designed to capture the shape contour information of instances in a scene, addressing the problem of distinguishing spatially distributed similar instances.
Article
Computer Science, Information Systems
Zhaobin Wang, Xiong Gao, Runliang Wu, Jianfang Kang, Yaonan Zhang
Summary: This study proposes a fully automated image segmentation method based on FCN and Graph Cuts. It utilizes color histograms and mathematical morphological operations to generate seed regions and performs iterative segmentation using superpixel-level Graph Cuts, resulting in higher segmentation accuracy.
MULTIMEDIA SYSTEMS
(2022)
Article
Computer Science, Interdisciplinary Applications
Junaid Malik, Serkan Kiranyaz, Riyadh Al-Raoush, Olivier Monga, Patricia Garnier, Sebti Foufou, Abdelaziz Bouras, Alexandros Iosifidis, Moncef Gabbouj, Philippe C. Baveye
Summary: This study presents a novel automatic segmentation technique, QCuts-3D, for binary segmentation of volumetric images of porous media. By drawing parallels with natural image segmentation and utilizing state-of-the-art spectral clustering technique, the proposed method overcomes the drawbacks of existing techniques. Additionally, a dataset of 68 multiphase volumetric images with ground truth annotations is provided for comparative evaluations, showcasing the superiority of QCuts-3D in accuracy and computational complexity over the current state-of-the-art. Statistical analysis also demonstrates its robustness against compositional variations of porous media.
COMPUTERS & GEOSCIENCES
(2022)
Article
Mathematical & Computational Biology
Ke Bi, Yue Tan, Ke Cheng, Qingfang Chen, Yuanquan Wang
Summary: This paper proposes an approach to extract the Left Ventricle (LV) in a sequence of cardiac Magnetic Resonance Images (MRI) by introducing a novel shape similarity constraint called sequential shape similarity (SSS). The proposed method segments all images in the sequence simultaneously and assumes that the shape of the LV in each image is similar to nearby images. By combining the Active Contour Model and the previously proposed Gradient Vector Convolution (GVC) external force with the SSS constraint, the snake contour accurately delineates the LV boundaries. Evaluation on two cardiac MRI datasets using Mean Absolute Distance (MAD) and Hausdorff Distance (HD) metrics shows that the proposed approach performs well in segmenting the boundaries of the LV.
MATHEMATICAL BIOSCIENCES AND ENGINEERING
(2022)
Article
Computer Science, Information Systems
Amirhossein Fallahdizcheh, Sandeep Laroia, Chao Wang
Summary: In this article, a two-stage active contour method is proposed to accurately and efficiently segment ascites. A morphological-driven thresholding method is used to automatically locate the initial contour of the ascites, and a novel sequential active contour algorithm is applied to accurately segment the ascites from the background. The experimental results show the superiority of the proposed method in both accuracy and time efficiency.
IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS
(2023)
Article
Mathematics, Applied
Osama Ala'yed, Rania Saadeh, Ahmad Qazza
Summary: In this study, a collocation method based on cubic B-spline functions is developed to solve Lane-Emden type equations arising in physics, star structure, and astrophysics. The singularity behavior of the considered system at tau=0 is overcome using the L'Hopital rule. The proposed method is shown to have a second-order convergence through convergence analysis and demonstrated its effectiveness, accuracy, simplicity, and practicality through numerical solutions to five test problems.
Article
Engineering, Electrical & Electronic
Mingchun Li, Dali Chen, Shixin Liu, Fang Liu
Summary: The novel instance segmentation network proposed in this article, named prior mask R-CNN, effectively combines prior knowledge with data-driven deep learning for automatic precipitation detection in transmission electron microscopes (TEM). Through improvements at different stages and the design of an effective measurement extraction module, the method achieves satisfactory results in experiments.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Radiology, Nuclear Medicine & Medical Imaging
Zahra Sedghi Gamechi, Andres M. Arias-Lorza, Zaigham Saghir, Daniel Bos, Marleen de Bruijne
Summary: The study proposes a fully automatic method based on an optimal surface graph-cut algorithm for accurate segmentation of the pulmonary arteries and aorta in noncontrast computed tomography (CT) scans. Evaluation results show that the method performs well in segmentation accuracy and diameter measurements.
Article
Computer Science, Information Systems
J. H. Gagan, Harshit S. Shirsat, Yogish S. Kamath, Neetha I. R. Kuzhuppilly, J. R. Harish Kumar
Summary: In this paper, we propose a fully automated method for segmenting the optic disc in retinal fundus images using a basis-spline-based active contour. The method achieves segmentation by optimizing the energy of the active contour with respect to five free parameters using gradient descent technique and Green's theorem. The use of these techniques reduces computational cost and speeds up the segmentation task. The method is validated on multiple databases and demonstrates high segmentation accuracy.
Article
Computer Science, Artificial Intelligence
Chenglong Li, Xiaobin Yang, Guohao Wang, Aihua Zheng, Chang Tan, Jin Tang
Summary: License plate recognition is crucial in various practical applications, however, recognizing license plates of large vehicles is challenging due to low resolution, contamination, low illumination, and occlusion. To address this problem, a novel data generation framework based on the Disentangled Generation Network is proposed to ensure the generation diversity and integrity for robust enlarged license plate recognition.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2024)
Article
Computer Science, Artificial Intelligence
Sara Casao, Alvaro Serra-Gomez, Ana C. Murillo, Wendelin Bohmer, Javier Alonso-Mora, Eduardo Montijano
Summary: This paper presents a hybrid camera system that combines static and mobile cameras, exploiting the cooperation between tracking and control modules to achieve high-level scene understanding. The static camera network provides global awareness, while the mobile cameras enhance the information about the people on the scene.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2024)
Article
Computer Science, Artificial Intelligence
Anh-Dzung Doan, Bach Long Nguyen, Surabhi Gupta, Ian Reid, Markus Wagner, Tat-Jun Chin
Summary: To ensure reliable object detection in autonomous systems, the detector needs to adapt to changes in appearance caused by environmental factors. We propose a selective adaptation approach using domain gap as a criterion to improve the efficiency of the detector's operation.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2024)
Article
Computer Science, Artificial Intelligence
Tianhong Dai, Wei Li, Xilei Cao, Jianzhuang Liu, Xu Jia, Ales Leonardis, Youliang Yan, Shanxin Yuan
Summary: This study proposes a novel frequency-guided deep neural network (FHDRNet) for high dynamic range (HDR) imaging from multiple low dynamic range (LDR) images, aiming to address ghosting artifacts. By conducting HDR fusion in the frequency domain, the network utilizes low-frequency signals to remove specific ghosting artifacts and high-frequency signals to preserve details. Extensive experiments demonstrate that this approach achieves state-of-the-art performance.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2024)
Article
Computer Science, Artificial Intelligence
Guobin Li, Reyer Zwiggelaar
Summary: Breast cancer is the most commonly diagnosed female malignancy worldwide. Recent developments in deep convolutional neural networks have shown promising performance for breast cancer detection and classification. However, biased features can be learned due to variations in appearance and small datasets. To address this issue, a densely connected convolutional network (DenseNet) was trained using texture features representing different physical morphological representations as inputs. The use of connectivity estimation and nearest neighbors improved the network's unbiased prediction. The approach achieved higher diagnostic accuracy and provided visual explanations for model predictions.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2024)
Article
Computer Science, Artificial Intelligence
Yuezun Li, Cong Zhang, Honggang Qi, Siwei Lyu
Summary: Deep Neural Networks (DNNs) are vulnerable to adversarial perturbations, limiting their applicability in safe-critical scenarios. To address this, a new method called AdaNI is proposed to increase feature randomness through adaptive noise injection, improving adversarial robustness. Extensive experiments demonstrate the efficacy of AdaNI against various white-box and black-box attacks, as well as its applicability in DeepFake detection.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2024)
Article
Computer Science, Artificial Intelligence
Chengyin Hu, Weiwen Shi, Ling Tian, Wen Li
Summary: In this study, we introduce a pioneering black-box light-based physical attack called Adversarial Neon Beam (AdvNB). Our method excels in attack modeling, efficient attack simulation, and robust optimization, striking a balance between robustness and efficiency. Through rigorous evaluation, we achieve impressive attack success rates in both digital and real-world scenarios. AdvNB demonstrates its stealthiness through comparisons with baseline samples and consistently achieves high success rates when targeting advanced DNN models.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2024)
Article
Computer Science, Artificial Intelligence
Hang Wang, Zhenyu Ding, Cheng Cheng, Yuhai Li, Hongbin Sun
Summary: Learning-based super resolution has made remarkable progress in improving image quality, but the performance decreases when the degradation kernel changes. Blind SR networks can estimate the degradation kernel and adapt well in realistic scenarios, improving performance and runtime. This paper proposes a design that imposes constraints for the kernel estimation network in both the image domain and kernel domain, resulting in high-quality images and efficient runtime.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2024)
Article
Computer Science, Artificial Intelligence
Yuantao Chen, Runlong Xia, Kai Yang, Ke Zou
Summary: This paper proposes an improved image inpainting network using a multi-scale feature module and improved attention module. The network addresses issues in deep learning-based image inpainting algorithms, such as information loss in deep level features and the neglect of semantic features. The proposed network generates better inpainting results by reducing information loss and enhancing the ability to restore texture and semantic features.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2024)
Article
Computer Science, Artificial Intelligence
Yi-Tung Chan
Summary: This study proposes a novel maritime background subtraction method based on ensemble learning theory to address the challenges posed by dynamic marine environments and noise, improving the detection accuracy and enhancing maritime transportation security for autonomous ships in open waters.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2024)
