Articles tagged with Image Processing
Researchers developed a novel, calibration-free, and reconstruction-free imaging technique that directly obtains a clear image from a single shot of speckle images. The method uses real-time video imaging to build a three-dimensional image, taking slices through the speckles.
RIT Professor Roger Easton and his team used multispectral imaging to reveal detailed drawings on an ancient Greek palimpsest, reconstructing Hipparchus' Star Catalogue. The discovery showcases the technique's growing role in historical preservation, with RIT supporting locations across US and Europe.
Researchers use drone technology to detect changes in building facades and assess technical condition, enabling accurate classification and identification. The system can also be used for urban heritage site monitoring and planning urban development.
This study proposes DSD-MatchingNet, a novel network leveraging sparse-to-dense hypercolumn matching and deformable convolutional networks for robust local feature matching. The proposed framework generates multilevel dense feature maps and incorporates pixel-level correspondence estimation error to improve accuracy.
A new AI tool developed by Brigham and Women's Hospital improves the accuracy of time-critical pathology diagnostics during surgery. The tool, which leverages deep-learning technology, translates frozen tissue samples into high-quality images, increasing diagnostic accuracy and reducing the need for lengthy laboratory tests.
A team led by Abhinav Jha is developing a novel low-count quantitative single photon emission computed tomography (LC-QSPECT) method to measure the concentration of radiopharmaceutical material in tumors and vital organs. This technique aims to provide accurate measurements of absorbed dose from radiation therapy, helping plan treatmen...
Researchers developed a laser-based approach to perform microbiopsies, enabling fast, painless tissue sampling with minimal damage. The novel technique uses laser ablation to extract tiny tissue volumes, which can be analyzed using virtual H&E imaging and other techniques in minutes, not hours.
Researchers at Incheon National University have developed an IoT-enabled, real-time object detection system for autonomous vehicles. The YOLOv3-based model achieved high accuracy (>96%) in detecting 2D and 3D objects, outperforming other state-of-the-art detection models.
Researchers developed a new 3D imaging model to analyze cellular geometry and mechanics, helping biologists quickly see how plants respond to environmental changes. The model identified unexpected guard cell behavior, shedding light on plant adaptation to drought.
Researchers from the University of Johannesburg deployed Few Shot Learning (FSL) for NIALM, a non-intrusive appliance load monitoring system. FSL requires only 7 test images to recognize appliances with 97.83% accuracy, making it faster and more cost-effective than traditional Machine Learning.
Researchers are developing a comprehensive cell atlas for the human brain, which would cover billions of cells and provide a holistic description of its properties. Advanced visualization techniques have been developed to catalog brain regions and cell types, enabling better resolution and accuracy in studying neural circuits.
A novel multi-modal image retrieval system, DenseBert4Ret, has been developed by researchers from Gwangju Institute of Science and Technology (GIST) using deep learning algorithms. The system outperforms state-of-the-art models in retrieving images based on both image and text features.
Researchers develop hybrid brightfield-darkfield transport of intensity approach, expanding accessible sample spatial frequencies and achieving 5-fold resolution increase. This method enables precise detection and quantitative analysis of subcellular features in large-scale cell studies.
Researchers from North Carolina State University have developed a new method for identifying genes relevant to the aging process in the C. elegans roundworm model. By exposing thousands of worms to random genetic mutations, they can pinpoint which genes are associated with protein aggregation and reduced lifespan.
Researchers from the University of Kassel developed an approach to extend the limits of interferometric topography measurements for optical resolution below small structures. Microsphere assistance enables fast and label-free imaging without requiring extensive sample preparation.
Researchers from the University of Tokyo's Interactive Intelligent Systems Laboratory developed a new system called LookHere that incorporates natural hand gestures into the teaching process. This approach eliminates extraneous details and provides better input data for machines to create models, resulting in improved efficiency and ac...
The Keck School of Medicine's Stevens INI has received a nearly $2 million grant to upgrade its storage system, increasing its capacity to handle vast amounts of brain imaging data. This will enable the lab to maximize scientific productivity and fulfill its promise to collaborate with researchers globally.
Researchers at UTHealth Houston will create a coordinating unit for biostatistics, informatics, and engagement to advance knowledge about human brain neurons. The project aims to produce an open-access digital brain cell reference atlas to improve understanding of neurological functions and disorders.
A team of international scientists found that many densely populated coastal cities worldwide are at risk from sea level rise due to land subsidence. The study used satellite images to estimate the rate of land sinking in 48 cities, with median speeds ranging from 16.2mm/year to 43mm/year.
Researchers developed a lensless camera that captures 3D information with a single exposure using a thin microlens array and new image processing algorithms. The camera can produce 3D images in real-time, enabling applications such as industrial part inspection, gesture recognition, and 3D display systems.
Researchers at George Washington University found that people perceive objects differently based on their prior knowledge and experience, with manipulable objects perceived faster but with less detail, while non-manipulable objects are perceived slower but with higher detail.
Researchers from Singapore University of Technology and Design (SUTD) have developed a new Brain-Inspired Replay model that enables continual learning in edge computing systems without storing data. This approach achieves state-of-the-art accuracy and high energy efficiency, overcoming the stability-plasticity issue in traditional models.
Researchers have developed a novel computational imaging framework, Compact Light-field Photography (CLIP), allowing the camera system to acquire wide and deep panoramic views. The technology enables the detection of hidden objects around occlusions and has potential applications in autonomous vehicles and medical imaging.
Researchers found that 7-month-old infants quickly discriminate between symmetrical and asymmetrical mosaics, indicating a robust ability to detect structural symmetry. This skill coincides with those found in studies using other stimuli, suggesting babies are good at extracting structure and rules from various media.
Researchers at MIT have developed a machine-learning system that uses computer vision to monitor the 3D printing process and correct errors in real-time. The system successfully printed objects more accurately than other 3D printing controllers, enabling engineers to incorporate novel materials into their prints with ease.
A team of researchers at Osaka University has created a machine learning system that can virtually remove buildings from a live view, streaming in real-time on a mobile device. This technology can help accelerate the process of urban renewal based on community agreement, reducing conflicts and delays.
A new study suggests that voluntary imagination, rather than speech, is crucial for combinatorial language acquisition in children with language deficiencies. The research found that children who engaged in a voluntary imagination intervention showed significant improvement in language comprehension.
Researchers developed a lidar-based system for smart cars to recognize objects more accurately than cameras. The system uses a grid map to divide the field of view into regions containing individual objects.
Researchers at Duke University have developed a new design for plasmonic metasurfaces that greatly expands their frequency range while also making them more robust against the elements. The new fabrication process allows for the use of a wide variety of shapes, opening up new possibilities for applications such as super cameras.
Researchers developed a novel convolutional neural network for facial expression recognition, outperforming conventional models while being computationally less expensive. The new model achieved an accuracy of 72.4% using only 58,000 parameters.
A research team reanalyzed M87 data using standard tools, revealing a core at the galactic center instead of the ring structure observed by the Event Horizon Telescope. The study validates EHT results and provides new insights into the astrophysical jet extending from the core.
By using the brain's visual response as feedback, researchers can reconstruct images of simple objects in real-time. The technique has potential applications in augmenting human capabilities and could one day be used to bring together human and artificial intelligence.
Researchers at the University of Tokyo have developed a new method to detect deepfakes, using self-blended images that improve detection accuracy by 5-12%. The team created novel synthesized images with controlled artifacts to train algorithms and found significant improvements in detecting deepfake images and videos.
Researchers have discovered the process of incorporating selenium into 25 specialized proteins, essential for various cellular and metabolic processes. The study provides critical insights into the workings of these vital mechanisms, which could lead to the development of new medical therapies.
The device can instantly recognize what it sees, like automatic descriptions of pictures taken by a camera or phone, and outperforms the eye in the number of wavelengths it can see. It's uniqueness comes from its ability to integrate three different operations into one, making it many times faster than current technology.
Researchers found that individual variation in primary visual cortex size and brain tissue can predict contrast sensitivity, with larger V1 areas and more cortical tissue leading to better vision at specific locations. The study reveals a new link between brain structure and behavior.
Researchers at Carnegie Mellon University developed an AI-powered method for robots to recognize and pour transparent liquids like water. The technique uses contrastive learning for unpaired image-to-image translation, enabling robots to see through different backgrounds and pour accurately.
Scientists create genetically engineered mouse model that changes color in response to light, allowing them to isolate background noise from blood flow and enhance imaging techniques. This breakthrough enables researchers to observe internal physiology with unprecedented accuracy, paving the way for new treatments and therapies.
A team from the Institute for X-ray Physics at the University of Göttingen has developed a new method for X-ray microscopy that uses imperfect lenses to achieve higher image quality and sharpness. The researchers used a lens consisting of finely structured layers deposited on a thin wire and adjusted it between the object to be imaged ...
The new technique, 3D optical coherence refraction tomography (3D OCRT), produces highly detailed images revealing features difficult to observe with traditional OCT. It has the potential for biomedical research and eventually more accurate medical diagnostic imaging.
Researchers at Cornell University have developed a new algorithm for autonomous underwater sonar imaging that significantly improves speed and accuracy for identifying objects such as explosive mines and sunken ships. The new approach, called informative multi-view planning, integrates information about object locations with sonar proc...
Researchers at the Lew lab have created a novel hardware and algorithm that enables visualization of cell membranes and molecular motions in six dimensions. This breakthrough allows for the observation of 3D structures with additional information on molecular orientation, providing new insights into biological systems.
A study led by Flinders University reveals that Australia's Letter-winged kite may not have enhanced vision for nighttime hunting as previously thought. The research found similar eye dimensions to its daytime-hunting relatives, contradicting decades of speculation about the kite's visual system.
A new computer-aided diagnostic tool, Deep-Lung Parenchyma-Enhancing (DLPE), uses artificial intelligence to reveal signs of pulmonary fibrosis in COVID long-haulers, helping to explain respiratory symptoms and improve disease management.
The review highlights advances in fundamental visualization methods for medical images in 3D, including scalar, vector, and tensor data. Medical professionals can quickly locate proper techniques using a taxonomy of medical problems and examples of health applications.
Researchers developed a system to estimate the probability of a majority response being wrong, eliminating unnecessary task assignments and increasing efficiency in crowdsourcing projects. This method can be applied to various binary classification tasks, reducing labor and improving accuracy.
An artificial intelligence model trained on histological images of surgical specimens accurately classified patients with and without Crohn disease recurrence. The model revealed previously unrecognized differences in adipose cells and mast cell infiltration, enabling stratification by prognosis for postoperative Crohn disease patients.
Scientists have developed a transparent device that produces a hidden image when light shines on it, using liquid crystals to recreate an ancient light trick. The technology has the potential to enable reconfigurable displays and stable 3D images.
Artificial Intelligence can now identify legendary batting techniques used by Sir Donald Bradman and modern players. Researchers developed a deep learning computer vision AI model to detect lateral backlift batters from straight ones.
Researchers found that speeding can decrease fuel economy by 7% and result in an extra 28 cents per gallon at current US fuel prices. They also studied human behavior during the early days of COVID-19 and simulated cosmic collisions to better understand X-ray emissions.
Researchers have developed a novel image reconstruction method using Vision Transformer (ViT) architecture to overcome limitations of conventional methods. The proposed algorithm enables the acquisition of high-quality images in a short computing time, suitable for real-time capture and various applications.
Researchers used terahertz imaging to uncover a hidden inscription on a 16th-century lead funerary cross, revealing the Lord's Prayer. The technique allowed for non-destructive examination of the corrosion layer, enabling the team to restore and enhance images containing the text.
Researchers developed a live imaging system to observe collagen synthesis in fibroblasts, revealing the intracellular processing and transportation of collagen fibers. The study found that this step controls the speed of collagen synthesis, providing a new understanding of collagen production.
Two new methods for producing high-resolution visualizations of small artefacts are presented, allowing anyone to create high-quality images and models with minimal effort and cost. The protocols provide detailed workflows for photographic acquisition and processing, enabling replicability and reproducibility in the field of archaeology.
A team of researchers developed a novel chip-based infection model to study invasive aspergillosis, a mold infection that affects the lungs. The model allows for live microscopic observation of damage caused by fungal hyphae and the response of immune cells.
Researchers developed a novel algorithm, 'Joint Space and Frequency Reconstruction' (JSFR-SIM), to accelerate image reconstruction in optically sectioned superresolution structured illumination microscopy. The method achieves 80 times faster execution speed without compromising image quality.
A novel method developed by the University of Tsukuba uses drones and machine learning to estimate the amount of plastic litter in rivers. The approach combines high-resolution optical and thermal images, resulting in more accurate estimates than other methods.
Scientists develop models that complement simulations using reinforcement learning and numerical methods to predict climate change, turbulent flows, and morphogenesis. This approach enables faster and more energy-efficient predictions, solving complex problems in engineering and climate applications.
Researchers at Kyoto University found that chimpanzees exhibit a preference for faces and skulls of their own species, similar to previous studies with African elephants. This suggests the presence of a face module in chimpanzee brains that detects and processes facial cues.
Scientists have discovered a shapeshifting volcano virus with remarkable properties that let it alter its shape. This finding could lead to new ways to deliver drugs and vaccines, with implications for understanding how viruses evolved and potentially creating new technologies.