Article
Engineering, Electrical & Electronic
Qinyan Huang, Weiwen Zhou, Minjie Wan, Xin Chen, Kan Ren, Qian Chen, Guohua Gu
Summary: In this paper, a multi-feature driven active contour segmentation model is proposed to handle infrared images with intensity inhomogeneity by combining global and local feature information. Experimental results demonstrate that the presented method outperforms typical models in terms of precision rate and overlapping rate in infrared test images.
OPTICAL AND QUANTUM ELECTRONICS
(2021)
Article
Engineering, Biomedical
A. Smolders, A. Lomax, D. C. Weber, F. Albertini
Summary: Fast and accurate contouring of daily 3D images is essential for online adaptive radiotherapy. This study aims to incorporate patient-specific information into convolutional neural networks (CNNs) to improve their segmentation accuracy. Patient-specific fine-tuning of CNNs significantly improves contour accuracy compared to standard CNNs, outperforming rigid registration and a commercial DL segmentation software, and yielding similar contour quality as deformable registration (DIR) while being 7-10 times faster.
PHYSICS IN MEDICINE AND BIOLOGY
(2023)
Review
Chemistry, Multidisciplinary
Mariusz Martyniuk, K. K. M. B. Dilusha Silva, Gino Putrino, Hemendra Kala, Dhirendra Kumar Tripathi, Gurpreet Singh Gill, Lorenzo Faraone
Summary: This paper reviews the MEMS optical filter technologies developed for the important infrared and terahertz wavelength bands, and demonstrates a portable, spectroscopic, chem/bio sensing solution suitable for remote sensing and imaging applications.
ADVANCED FUNCTIONAL MATERIALS
(2022)
Article
Mathematics, Interdisciplinary Applications
Yanzhu Zhang, Lijun Yang, Yan Li
Summary: A novel adaptive fractional differential active contour image segmentation method is proposed in this paper to improve the accuracy of segmentation results when the image is affected by strong noise and uneven intensity.
FRACTAL AND FRACTIONAL
(2022)
Article
Chemistry, Multidisciplinary
Tao Liu, Xingchen Lv, Min Jin
Summary: In this study, a detection method based on adaptive chord inclination angle was proposed to accurately measure the critical dimensions of flat parts using machine vision. The method utilized bicubic interpolation to increase the size of the part image and extracted single-pixel contours with more detailed information. By employing an adaptive step size, the front and back chord inclination angles of the contour were obtained. Complementing the front and back angles helped to avoid negative effects caused by contour jaggedness. The method successfully obtained segmentation points and accurately fitted individual graphic elements to determine the key dimensions of the closed part. The relative error was less than 0.6%.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Information Systems
Hui Chen, Hongyan Wu, Ning Yang, Heping Huang, Weibin Liang
Summary: Surface shape feature is an important index for monitoring objects. Existing slicing methods often have low volume measurement accuracy for point clouds due to fuzzy boundaries and static slicing. This study proposes an improved method based on dynamic adaptation which is suitable for calculating the volume of point clouds on irregular object surfaces with multiple contour boundaries. The method involves preprocessing the point cloud, separating the multiple contour boundaries using a two-step clustering method, introducing an adaptive mechanism for slicing, and calculating the point cloud volume.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Statistics & Probability
Shuyu Chu, Huijing Jiang, Zhengliang Xue, Xinwei Deng
Summary: In the pricing of customized products, accurately predicting the purchase behavior of different customers for personalized requests is challenging, requiring the construction of distinct models for data analysis. An adaptive convex clustering method is proposed to segment data and fit models simultaneously, ensuring data points with similar model structures are grouped together.
Article
Engineering, Electrical & Electronic
Luiz Felipe da Silveira Coelho, Lisandro Lovisolo, Michel Pompeu Tcheou
Summary: Implementing FIR adaptive filters with SOPOT arithmetic can lead to simpler hardware design and reduced power consumption. This paper evaluates the effects of SOPOT arithmetic on recursion algorithms of adaptive filters, focusing on convergence rate, numerical stability, and accuracy. SOPOT approximations obtained through the MPGBP algorithm show notable cost-performance trade-off and low computational complexity in the implementation of LMS, NLMS, and RLS algorithms for system identification and change detection.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS
(2022)
Article
Chemistry, Analytical
Sen Yang, Xiaobao Wang, Qijuan Yang, Enzeng Dong, Shengzhi Du
Summary: This paper proposes a method to improve instance segmentation algorithms by replacing the single BN with an adaptive weight loss layer. By enhancing the input feature expression ability, the proposed method improves stability and convergence speed, independent of the batch size.
Article
Engineering, Biomedical
Y. Zhang, J. Duan, Y. Sa, Y. Guo
Summary: In this paper, a multi-atlas based adaptive active contour model is proposed for automatic segmentation of OARs in brain MR images. Experimental results show that the proposed method achieved more accurate segmentation results in brainstem, eyes, and lens.
Article
Instruments & Instrumentation
Yushaa Shafqat Malik, Maria Tamoor, Asma Naseer, Aamir Wali, Ayesha Khan
Summary: The study proposes a fully automated initialization algorithm that can be used with active contour model to effectively segment regions of interest in medical images. Evaluation results on a skin lesion dataset demonstrate its superiority compared to other methods.
JOURNAL OF X-RAY SCIENCE AND TECHNOLOGY
(2022)
Article
Engineering, Electrical & Electronic
Xiaofang Zhang, Suxiao Li, Bin Zhang, Jie Dong, Shujun Zhao, Xiaomin Liu
Summary: Automatic detection and segmentation of lung nodules can assist doctors in diagnosing and treating lung cancer. This study proposes a method that achieves better performance than previous approaches in terms of Jaccard index and Dice similar coefficient, by applying a series of operations for lung tissue segmentation and nodule classification in different locations.
INTERNATIONAL JOURNAL OF IMAGING SYSTEMS AND TECHNOLOGY
(2021)
Article
Computer Science, Information Systems
Mohamed S. Elmahdy, Laurens Beljaards, Sahar Yousefi, Hessam Sokooti, Fons Verbeek, Uulke A. Van der Heide, Marius Staring
Summary: This paper introduces a method that combines medical image registration and segmentation tasks using Multi-Task Learning in medical image analysis, demonstrating superior performance in contour generation quality and generalization compared to Single-Task Learning. The proposed method shows high accuracy and fast inference speed, making it promising for automatic re-contouring in adaptive radiotherapy follow-up scans.
Article
Computer Science, Information Systems
Takashi Ohhira, Akira Shimada, Toshiyuki Murakami
Summary: This paper introduces a new method for designing disturbance observers (DOB) that can estimate system state and unknown disturbances simultaneously while reducing the influence of noises. The proposed method uses a Kalman filter-based DOB and adaptive KF to achieve simultaneous estimation of system state and unknown disturbances. Simulation results show the effectiveness of this method in various noise environments.
Article
Engineering, Electrical & Electronic
Li Gengsheng, Liu Guojun, Ma Wentao
Summary: An adaptive active contour model based on the coupling of local and global information was proposed in this paper to address the shortcomings of traditional variational level set segmentation algorithm. By adjusting the proportions of local and global region fitting energy using a weight function, the algorithm is able to effectively segment uneven gray images. Experimental results demonstrate the algorithm's robustness to noise and insensitivity to the initial position of contours.
LASER & OPTOELECTRONICS PROGRESS
(2022)
Article
Instruments & Instrumentation
He Zhang, Weixian Qian, Minjie Wan, Kaimin Zhang
Summary: This paper proposes an infrared image enhancement method using local entropy mapping histogram adaptive segmentation, which effectively solves the problems of over-enhancement and noise amplification in traditional histogram equalization algorithm.
INFRARED PHYSICS & TECHNOLOGY
(2022)