Article
Agriculture, Multidisciplinary
Gafur Gozukara, Alfred E. Hartemink, Yakun Zhang
Summary: This study explored the effect of different illumination levels on the prediction accuracy of soil organic carbon (SOC) using smartphone-based digital images. Lower illumination levels improved the prediction accuracy of SOC content, with the random forest model achieving relatively high accuracy (r2 = 0.42 to 0.70) under 300 lx illumination level. The a*, R, and H color parameters were the best variables for predicting SOC in the 300 lx dataset.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2023)
Article
Computer Science, Artificial Intelligence
Junqing Huang, Michael Ruzhansky, Qianying Zhang, Haihui Wang
Summary: This article presents a novel intrinsic image transfer (IIT) algorithm for image illumination manipulation, which creates a local image translation between two illumination surfaces. The model is constructed based on an optimization framework and utilizes illumination, reflectance, and content losses. These losses are defined on sub-layers factorized by an intrinsic image decomposition and reduced under the illumination-invariant reflectance prior knowledge. The article also demonstrates the algorithm's versatility in illumination compensation, image enhancement, tone mapping, and HDR image compression, showcasing high-quality results on natural image datasets.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Yongjie Zhu, Chen Li, Si Li, Boxin Shi, Yu-Wing Tai
Summary: In this paper, we propose HyFRIS-Net to jointly estimate the hybrid reflectance and illumination models, as well as the refined face shape in a pre-defined texture space from a single unconstrained face image. Our method ensures efficient learning of photometric face appearance modeling in both parametric and non-parametric spaces by enforcing reflectance consistency and face identity constraints. We recover an occlusion-free face albedo with disambiguated color from the illumination color. The network is trained in a self-evolving manner to achieve general applicability on real-world data. Comprehensive evaluations demonstrate the advantages of HyFRIS-Net in modeling photo-realistic face albedo, illumination, and shape.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)
Article
Computer Science, Artificial Intelligence
Heng Guo, Fumio Okura, Boxin Shi, Takuya Funatomi, Yasuhiro Mukaigawa, Yasuyuki Matsushita
Summary: Multispectral photometric stereo (MPS) aims to recover the surface normal of a scene measured under multiple light sources with different wavelengths. This paper proposes a new method that alleviates restrictive assumptions in existing MPS methods and investigates surfaces with spatially-varying reflectance. Experimental results demonstrate the effectiveness of the method and show its potential applicability for multispectral heritage preservation.
INTERNATIONAL JOURNAL OF COMPUTER VISION
(2022)
Article
Engineering, Electrical & Electronic
Medha Gupta, Vishwas Rathi, Puneet Goyal
Summary: This paper proposes an adaptive and progressive MSFA demosaicking method for multispectral images, which improves image quality and computational efficiency through the design of MSFA patterns and interpolation of the middle band.
IEEE TRANSACTIONS ON COMPUTATIONAL IMAGING
(2022)
Article
Archaeology
Yoonjo Choi, Yun-Jung Yang, Hong-Gyoo Sohn
Summary: Recent developments in geospatial information technology have enabled researchers to utilize geospatial information for decision making in disaster situations. Preservation of cultural properties is crucial as they reflect local characteristics and knowledge, and a method using geospatial information to enhance cultural heritage resilience in disasters has been proposed in this paper. The proposed method has been shown to minimize damage to cultural assets and enhance their resilience in disasters.
JOURNAL OF CULTURAL HERITAGE
(2021)
Article
Green & Sustainable Science & Technology
Karol Krol, Dariusz Zdonek
Summary: New communication channels and methods for retrieving information can provide precise data describing how cultural heritage is perceived, protected, promoted, and shared. This study aimed to identify the context and frequency of cultural heritage topics in Google search engine queries. The research revealed that Polish and English users were interested in similar phrases, suggesting that language might not impact the thought processes of internet users.
Article
Optics
Yuhao Niu, Yubin Liu, Guang Chen, Jieming Zhao, Lin Deng, Ye Sa, Zhan Gao
Summary: In this study, we proposed a new method that allows imaging of multiple objects with strong background illumination. By isolating the speckle pattern of each object through image difference process and eliminating background noise interference with principal components analysis (PCA) technique, our method can provide better observations of the objects in bright field scenarios.
OPTICS AND LASER TECHNOLOGY
(2024)
Article
Computer Science, Software Engineering
Ivan Sipiran, Patrick Lazo, Cristian Lopez, Milagritos Jimenez, Nihar Bagewadi, Benjamin Bustos, Hieu Dao, Shankar Gangisetty, Martin Hanik, Ngoc-Phuong Ho-Thi, Mike Holenderski, Dmitri Jarnikov, Arniel Labrada, Stefan Lengauer, Roxane Licandro, Dinh-Huan Nguyen, Thang-Long Nguyen-Ho, Luis A. Perez Rey, Bang-Dang Pham, Minh-Khoi Pham, Reinhold Preiner, Tobias Schreck, Quoc-Huy Trinh, Loek Tonnaer, Christoph von Tycowicz, The-Anh Vu-Le
Summary: This paper presents the methods and results of the SHREC'21 track on a dataset of cultural heritage objects, introducing challenges in shape retrieval and cultural retrieval. Ten groups participated with results showing the dominance of learning methods in characterizing 3D objects, but the challenges remain unsolved.
COMPUTERS & GRAPHICS-UK
(2021)
Article
Computer Science, Software Engineering
Masaru Ohkawara, Issei Fujishiro
Summary: We present a novel compositing framework for synthesizing full-length human figure images with high visual coherence. The framework consists of five stages, including photometric information estimation, 3D reconstruction, global illumination simulation, lighting transfer, and compositing. Through empirical evaluation, we demonstrate that the proposed framework functions effectively as a group portrait compositor and outperforms the results obtained by rendering multiple avatars together.
Article
Optics
Takuma Morimoto, Cong Zhang, Kazuho Fukuda, Keiji Uchikawa
Summary: This article presents a spectral dataset of daylights and surface reflectances and transmittances of natural objects measured in Japan. The study found that colors of natural objects were spread in a limited area of color space, and daylight spectra had a larger variation across time, weather, and local environments than previously thought.
Article
Computer Science, Artificial Intelligence
Tao Fan, Hao Wang, Sanhong Deng
Summary: Designing an efficient Intangible Cultural Heritage (ICH) image classification model is crucial for recognizing and preserving ICH. Existing research on ICH image classification mainly focuses on visual features, neglecting textual descriptions. In this study, we propose a multimodal approach that combines textual descriptions with ICH images to improve classification accuracy. Our proposed model, MICMLF, utilizes multimodal attention and hierarchical fusion to capture interactions between ICH images and textual descriptions. Experimental results on Chinese nation-level ICH projects show the superiority of MICMLF over state-of-the-art methods, even when dealing with incomplete data.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Geochemistry & Geophysics
Zhenqiang Qin, Xian Li, Yanfeng Gu
Summary: In this article, a physics-based method is proposed to estimate and compensate illumination for the radiometric correction of UAV multispectral images (MSIs). The proposed method builds an illumination estimation model based on intraimage and interimage consistency hypotheses, and utilizes the physical imaging principle to alleviate the influence of varying illumination. Numerical experiments on three UAV datasets demonstrate that the proposed method outperforms current methods and significantly reduces the normalized root mean square error (NRMSE).
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Biodiversity Conservation
Franziska Wolff, Tiina H. M. Kolari, Miguel Villoslada, Teemu Tahvanainen, Pasi Korpelainen, Pedro A. P. Zamboni, Timo Kumpula
Summary: Plant communities of mires are important for ecological processes such as carbon storage and gas fluxes. Mapping mire vegetation using UAVs can provide valuable information for ecosystem assessment. However, accurate mapping of plant communities remains challenging due to overlapping spectral signatures of plant species.
ECOLOGICAL INDICATORS
(2023)
Article
Computer Science, Artificial Intelligence
Ilija Domislovic, Donik Vrsnjak, Marko Subasic, Sven Loncaric
Summary: This paper presents a novel method for estimating illumination in situations where the scene is illuminated by multiple different illuminants and the illumination is non-uniform. The method uses a lightweight convolutional neural network and performs patch-by-patch illumination estimation.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Engineering, Electrical & Electronic
K. Szelag, P. Kurowski, P. Bolewicki, R. Sitnik
OPTO-ELECTRONICS REVIEW
(2019)
Article
Environmental Sciences
Jakub Michonski, Marcin Witkowski, Bozena Glinkowska, Robert Sitnik, Wojciech Glinkowski
INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH
(2019)
Article
Chemistry, Analytical
Marcin Adamczyk, Pawel Liberadzki, Robert Sitnik
Article
Chemistry, Multidisciplinary
Michal Kozbial, Lukasz Markiewicz, Robert Sitnik
APPLIED SCIENCES-BASEL
(2020)
Article
Chemistry, Analytical
Krzysztof Krasowicz, Jakub Michonski, Pawel Liberadzki, Robert Sitnik
Article
Computer Science, Information Systems
Marta Nowak, Jakub Michonski, Robert Sitnik
Summary: This paper introduces a cavity reconstructing algorithm for 3D surface scans (CRASS) developed for filling cavities in point clouds representing human body surfaces. The method uses Bezier patches to reconstruct missing data and was tested on simulated and scanned 3D input data, with results showing quality assessment lower than 0.5 of average sampling density.
MULTIMEDIA TOOLS AND APPLICATIONS
(2021)
Article
Chemistry, Analytical
Sunita Saha, Jacek Martusewicz, Noelle L. W. Streeton, Robert Sitnik
Summary: This study introduces a novel change-based segmentation method for cultural heritage imaging to monitor surface geometry changes and aid decision-making and quantification in the field of conservation and restoration.
Article
Chemistry, Analytical
Wiktor Krajnik, Lukasz Markiewicz, Robert Sitnik
Summary: This paper presents a survey of current state-of-the-art solutions for three-dimensional shape estimation of the human body. It introduces a novel approach called sSfS that combines voxel-based Shape-from-Silhouette method with silhouette segmentation to achieve better results, especially in concave areas.
Article
Chemistry, Analytical
Piotr Osinski, Jakub Markiewicz, Jaroslaw Nowisz, Michal Remiszewski, Albert Rasinski, Robert Sitnik
Summary: This paper presents a new approach to designing and planning multi-view imaging networks for dynamic 3D scene reconstruction. The proposed method, based on the Monte Carlo algorithm and a set of prediction functions, determines the camera positions and orientations to achieve the required completeness, accuracy, and resolution of the final 3D reconstruction.
Article
Humanities, Multidisciplinary
Sunita Saha, Amalia Siatou, Alamin Mansouri, Robert Sitnik
Summary: This paper presents a supervised segmentation method for detecting surface changes on cultural heritage metal surfaces. The method utilizes Reflectance Transformation Imaging (RTI) reconstruction coefficients (PTM and HSH) to track changes over time and assigns each detected category a single color for visualization. The results show accurate segmentation, especially for HSH, and the method shows promise for tracking changes and quantifying the degree of change.
Article
Humanities, Multidisciplinary
Sunita Saha, Anna Tomkowska, Jacek Martusewicz, Robert Sitnik
Summary: This paper presents a method for replicating a ceramic Venus figurine from a mold stored in the Museum of Ceramic Techniques in Kolo, Poland. The process involves understanding old manufacturing technology, conservation supervision, and state-of-the-art 3D scanning, data processing, and 3D printing technologies. By integrating 3D scanning, optimizing and correcting the printed fragments, a realistic replica of the Venus figurine was created. This method saves time and produces indistinguishable copies suitable for educational purposes.
Article
Chemistry, Multidisciplinary
Wojciech Michal Glinkowski, Tomasz Miscior, Robert Sitnik
Summary: “The study developed the Ortho_Kinect_OR application to improve usability and satisfaction for surgeons in the operating room through contactless control of medical image access. Field tests demonstrated high usability and user satisfaction, without distracting the surgeons during procedures.”
APPLIED SCIENCES-BASEL
(2023)
Article
Chemistry, Multidisciplinary
Marta Nowak, Robert Sitnik
APPLIED SCIENCES-BASEL
(2020)
Review
Surgery
Maciej Kosieradzki, Wojciech Lisik, Radoslaw Gierwialo, Robert Sitnik
ANNALS OF TRANSPLANTATION
(2020)
Proceedings Paper
Archaeology
R. Sitnik, K. Lech, E. Bunsch, J. Michonski
OPTICS FOR ARTS, ARCHITECTURE, AND ARCHAEOLOGY VII
(2019)