Articles tagged with Image Processing
A team of scientists has developed an efficient large-scale phase retrieval technique for realizing high-fidelity complex-domain phase imaging. The new method combines conventional optimization algorithms with deep learning techniques, achieving robustness to measurement noise and strong generalization. By comparing the reported method...
Researchers developed a method to overlay a virtual scale on acquired endoscope images in real-time, allowing accurate estimation of colorectal polyp sizes. The approach uses triangulation principles and minimal image processing, enabling cost-effective diagnosis without adding extra instrumentation.
A study by Anglia Ruskin University found that weaker internal connections between the brain and organs are linked to negative body image. Adults with less efficient brains at detecting internal messages are more likely to experience body shame and weight preoccupation.
A team of scientists from Osaka University developed a machine learning method for classifying the type of building and its primary façade color using deep learning models applied to street-level images. This work may assist in fostering neighborhood cohesion and support urban renewal by providing tailored street-view datasets.
Using complex-valued layers can improve performance against adversarial attacks without sacrificing efficiency. This technique, combined with gradient regularization, allows neural networks to resist small perturbations and maintain accuracy.
A team of researchers at Washington University in St. Louis has received a $3.12 million NIH grant to study neurovascular recovery after stroke. They aim to develop new neurovascular imaging technology using two-photon fluorescence microscopy and photoacoustic microscopy to visualize blood oxygen delivery in response to neuronal activity.
Researchers at UCLA have developed a computational technique powered by artificial intelligence that transforms images of tissue previously stained with H&E into new ones with added special stains. The process takes less than one minute per tissue sample, significantly improving diagnoses in medical conditions such as organ transpl...
The team used machine learning technique generative adversarial networks to digitally remove clouds from aerial images, generating accurate datasets of building image masks. This work may help automate computer vision jobs critical to civil engineering, enabling the detection of buildings in areas without labeled training data.
Researchers have developed a dye-free method to visualize blood flow in the brain, allowing for detailed mapping of small capillaries and assessing blood flow rates. The technique has potential applications in understanding cardiovascular diseases, tumor growth, and targeted drug delivery.
Scientists at Nara Institute of Science and Technology create a projected touchscreen system using just one camera and projector, eliminating the need for additional detectors. The system uses slope disparity gating to capture touch data with high efficiency, enabling portable projection systems for large interactive displays.
Scientists have developed a new live analysis system for plant stomata, allowing for rapid and affordable identification of desirable traits. This innovation has the potential to accelerate crop development for climate-resistance, addressing future food shortages.
A recent study published in Biosystems Engineering explores the potential of smartphone cameras to assess soil organic matter and evaluate soil fertility. The technique uses advanced image analysis and machine learning to predict SOM values rapidly and with high correlation to traditional soil analysis.
Researchers at Duke University developed a robotic imaging tool that can automatically detect and scan patients' eyes for eye diseases, producing clear images in under 50 seconds. The system uses optical coherence tomography and is designed to be safe and accessible for optometrist offices, primary-care clinics, and emergency departments.
A retrospective study of over 3,900 patients found that warming iohexol 350 contrast media did not significantly reduce adverse reactions or extravasations. The results suggest that maintaining the agent at room temperature is non-inferior to warming it to body temperature before injection.
A study published in Nature Communications reveals the visual system retains information from moving images, providing a consistent representation of surroundings. The researchers found that neuronal responses in deeper layers of the visual cortex exhibit 'perceptual constancy' and 'intrinsic persistence', ensuring stable encoding and ...
Researchers have developed a novel spectral-volumetric compressed ultrafast photography system that captures 5D information in a single snapshot. This breakthrough imaging technique enables new insights into ultrafast phenomena in physics and biochemistry.
A new imaging technique called DEEP enables researchers to image complex biological systems at high resolution and speeds previously impossible. This breakthrough may lead to new understandings of brain function and other biological processes.
Researchers at NIST developed a method using radio signals to image hidden and speeding objects, enabling real-time imaging around corners and through walls. The technique has potential applications in public safety, tracking hypersonic objects, and improving space debris detection.
Researchers developed algorithms to solve curvature-dependent image processing problems using convex optimization, inspired by Euler's elastic curves and Gestalt laws of perception. The new models perform as well as current deep-learning algorithms but provide a deeper understanding of the structures learned.
Researchers developed a new multi-modal image fusion method based on supervised deep learning to enhance image clarity, reduce redundant features, and support batch processing. The method achieves state-of-the-art performance in visual quality and quantitative evaluation metrics, improving medical diagnosis accuracy.
Researchers developed a dual hybrid camera system that enables 360-degree monitoring while capturing high-resolution images, addressing the trade-off between field-of-view and resolution. The system uses an omnidirectional camera and a separate pan-tilt camera to capture highly accurate object identification.
A new algorithm corrects motion blur in single-photon images, allowing for high-quality pictures even with multiple objects moving independently. The approach accurately estimates individual object motion and groups pixels by similar motion, enabling deblurring of each region independently.
Scientists at EMBL combined AI algorithms with advanced microscopy techniques to reduce image processing time from days to mere seconds while maintaining accuracy. This breakthrough enables researchers to track and measure fast biological processes in 3D.
Researchers found that highly sensitive individuals' brains show activity suggesting depth of processing after emotionally evocative tasks. This trait is associated with a heightened appreciation of beauty and deeper bonds with others.
Researchers at Duke University and other institutions are developing a 'super camera' capable of capturing all types of information carried by light, including polarization, depth, phase, coherence, and incidence angle. The project aims to create an imaging system that can handle sensing and processing simultaneously, streamlining size...
Researchers developed a nanoprinted AI optical circuit that processes information at the speed of light without power consumption. The circuit's neural density is comparable to human brain density, enabling efficient processing of near-infrared inference on CMOS chips.
An international team has developed a way to image the interface between 2D and 3D materials, revealing details of atomic configurations and orientations. This breakthrough enables control over the electronic properties of atomically thin materials.
Researchers developed a novel camera-based system for automated drone landing using 2D symbol detection, allowing drones to safely land on uneven terrain. The system consists of a commercial flight controller, Raspberry Pi module, and image-processing algorithm that detects a distinctive landing symbol in real-time.
The new journal, Biological Imaging, provides an interdisciplinary forum for quantitative and computational imaging in life sciences, covering topics like microscopy, image acquisition, and machine learning. The journal aims to drive cross-fertilization across research applications and inspire innovative work.
A breakthrough in optical neural networks accelerates computing speed and processing power to over 1000 times that of previous processors. The system can process record-sized images and achieve full facial image recognition.
Researchers are using Cold War spy satellite images to detect environmental changes, including deforestation in Romania and ecological damage from bombs in Vietnam. The study reveals previously unseen changes in the environment and provides new findings on the extent of large-scale deforestation in Romania.
Researchers have developed an ultracompact metalens array that enables wide-field microscope imaging with large field of view and high resolution. The metalens-integrated imaging device (MIID) achieves compact architecture and working imaging distance in the hundreds of micrometers, paving the way for real-world applications.
Researchers at Sandia National Laboratories have developed a new device that more efficiently processes information using non-volatile computer memory. The breakthrough could revolutionize technologies like voice recognition, image processing, and autonomous driving by reducing energy consumption.
Researchers developed a new technology called PLUS to produce high-resolution 3D images of defects in metallic structures. This innovation uses non-destructive techniques to study structural integrity and identify potential flaws.
Researchers use NASA's IRIS mission to spot nanojets in the solar corona, revealing a potential coronal heating candidate: nanoflares. The observations confirm that nanojets are a telltale signature of magnetic reconnection and nanoflares contributing to coronal heating.
Researchers analyzed ancient texts from 600 BCE, concluding that many inhabitants of the kingdom of Judah could read and write. The study suggests that literacy was not exclusive to royal scribes, but rather a widespread skill among the population.
A novel modality for computational light-field imaging using a diffuser as an encoder has been developed, enabling lensless imaging with adjustable spatio-angular resolutions. This approach avoids the resolution limitation of traditional sensors, allowing for viewpoint shifting, post-capture refocusing and depth sensing capabilities.
Researchers at Texas A&M University have developed a new image processing technique that enhances the resolution of low-quality electron micrographs using deep neural networks. The method leverages pairs of low- and high-resolution images to train AI algorithms, allowing for detailed analysis without damaging specimen samples.
New research using ultra-high-field MRI reveals the cerebellum has a surface area equal to 80% of the cerebral cortex's surface area, challenging the long-held idea that it is smaller. This expansion is linked to human behavior and cognition evolution, enabling the processing of complex concepts like language and abstract reasoning.
Researchers have developed new techniques that can significantly reduce the time needed to process complex images from cutting-edge microscopes. These methods use deconvolution algorithm modifications, parallelization, and neural networks to speed up processing time by several thousand-fold.
KAIST researchers developed an algorithm for automated 3D brain imaging data analysis, enabling precise and quantitative mapping of complex neural circuits. The new technology allows for accurate comparison of brain data from different animals and improves the accuracy and consistency of analysis results.
A team of researchers from the University of Tokyo has successfully captured high-speed atomic video at 1,600 frames per second using a powerful electron microscope and highly sensitive camera. This achievement is 100 times faster than previous experiments and enables the observation of previously inaccessible details.
A new study found that low-intensity exercise activates brain networks involved in cognitive control and attention processing, while high-intensity exercise primarily affects emotion processing. The results suggest a potential therapeutic strategy for neurological and psychiatric disorders.
Researchers at Queen's University Belfast are developing cutting-edge millimetre-wave radar systems with machine learning capabilities to enhance security scanners' effectiveness. The technology hopes to process images in under a tenth of a second, reducing false alarms and queues.
A team of researchers, led by Uri Manor at the Salk Institute, used deep learning to develop a new approach for super-resolution microscopy. By training a neural network on high-resolution images, they were able to improve the resolution of microscope images, enabling better understanding of brain cells and their behavior.
HDR images offer a wider dynamic range of luminance, simulating human eye adaptation to various lighting conditions. A new method stabilizes the camera response function while coupling colour channels, producing artefact-free results.
Researchers developed an innovative method to measure the complexity of image representations in deep neural networks, shedding light on their processing stages. The study found that classification accuracy depends on the network's ability to simplify information, with more accurate results from simplified representations.
Researchers developed an automated platform using exclusive liquid repellency microdrops for lossless single-cell isolation, identification and retrieval. The system combines a robotic liquid handler, microscopic imaging system and real-time image-processing software to enable rapid hands-free isolation of rare cellular samples.
Researchers found that minute facial sweating can measure stress during knowledge production, while presenting views to management is more stressful than producing them. The study also revealed that spell checkers save time in long writings and that neurotic individuals perform better with email distractions.
Researchers analyzed 80,000 chords in 745 classic U.S. Billboard pop songs to find the right combination of uncertainty and surprise that makes music pleasing. The study suggests that musical pleasure depends on the dynamic interplay between prospective and retrospective states of expectation.
A team led by John Tsotsos disproved a long-standing theory of how the human vision system processes images. The study found that salience is not needed for quickly deciding what an image depicts and that current AI algorithms fall short in matching human performance.
Researchers found that the brain's inability to process information quickly enough can lead to a bottleneck in visual processing, resulting in missed stimuli. By reducing interference between feedforward and feedback signals, they observed improved detection and categorization performance.
Researchers developed a millimeter-sized chip that can image through walls or detect tumors using microwave signals. The chip-based imager uses optical processing and includes over 1,000 photonic components.
A new unified shock sensor developed by researchers at Yokohama National University can quickly and accurately detect and dissipate shock waves. The sensor combines imaging processing with compressible flow theory to predict the behavior of shock waves, offering improved efficiency and precision in computational fluid dynamics.
The new camera system uses a high-powered laser to capture reflected light from objects around the corner, allowing for real-time monitoring of movement in 3D. This breakthrough enables faster and more accurate tracking of objects beyond visible light spectrum, with applications in autonomous cars and robots
Researchers discovered that monovision can cause significant misperceptions of distance and 3D direction, particularly with moving objects. By analyzing the effects of blur differences on visual processing, they found that a darker lens can cancel out the negative impact of monovision.
A bioengineering group from the University of Pittsburgh developed a methodology to analyze imaging data, using approximate Bayesian computation to derive useful quantitative information for experimental design. They found that tissue properties such as force production and adhesion are more likely to vary with initial tissue size.
A new Salk study found that neurons selectively respond to particular combinations of color and shape, rather than extracting them separately. This breakthrough challenges the long-held notion that color and shape are processed in the early stages of vision.
Researchers developed a novel method for producing higher-quality images and scene designs in less time by using a deep-learning-based approach that considerably cuts the noise in images. The method results in sharper images with intricate details, including complex lighting components.
Researchers at NYU Tandon developed an AI-driven imaging system that can spot manipulated photos and videos with a detection rate of over 90%. The system introduces digital watermarks into images during acquisition, which are sensitive to manipulation. This approach optimizes image integrity and authentication, making it difficult for ...