Articles tagged with Image Processing
A new approach uses artificial intelligence to turn low-quality images into high-quality ones, enhancing the image quality of metalens cameras. This technology could make these cameras viable for intricate microscopy applications and mobile devices.
Researchers at Delft University of Technology developed a drone that flies autonomously using neuromorphic image processing and control based on the workings of animal brains. The drone's deep neural network processes data up to 64 times faster and consumes three times less energy than when running on a GPU.
Researchers have developed new optical tweezers that can stably trap large and irregularly shaped particles using contour-tracking technology. This advancement could expand light-based trapping to a wider range of objects, including groups of cells, bacteria, and microplastics.
Engineers at Stanford University have developed a prototype augmented reality headset that uses holographic imaging to overlay full-color, 3D moving images on the lenses of regular glasses. The new approach delivers a visually satisfying 3D viewing experience in a compact and comfortable form factor suitable for all-day wear.
A new study found that GPT-4 and Google Gemini performed poorly in breast imaging classification, with a high percentage of discordant assignments that could impact patient management. The researchers emphasize the need for regulation of these large language models in high-level medical scenarios.
Researchers have developed a compact and lightweight single-photon airborne lidar system that can acquire high-resolution 3D images with a low-power laser. The system uses single-photon detection techniques to measure time-of-flight, enabling highly accurate 3D mapping of terrain and objects even in challenging environments.
A new study from the University of Illinois explores the use of hyperspectral imaging and explainable artificial intelligence to assess sweet potato attributes, leading to more informed decision-making and higher-quality products. The results can help industry professionals understand the significance of different features in predictin...
Researchers developed a new method to accelerate high-resolution ultrasound localization microscopy using deep learning, enabling faster and more accurate imaging of microvascular structures. The technique, called LOCA-ULM, improves spatial resolution and processing speed while maintaining sensitivity for functional imaging.
A team of researchers has successfully integrated a metasurface with photonic integrated circuits, enabling fast and tunable control over light manipulation. The device can shape any wavefront in reconfigurable arbitrary polarization states at speeds of up to 1.4 gigahertz.
Researchers at NIH developed a novel AI-based method called P-GAN to improve next-generation imaging of cells in the retina. The technique reduces imaging acquisition and processing time by 100-fold, yielding greater contrast and improving image quality.
This novel approach combines data-driven deep learning with polarization imaging to extract higher-dimensional features and improve imaging quality. The integration of polarization information expands the application scope of existing intensity-based deep learning algorithms.
Research at Schöningen reveals sophisticated woodworking techniques used by early humans to process spruce and pine wood. The findings demonstrate the importance of wood as a raw material in human evolution, with evidence of extensive processing and recycling.
A new approach uses a smartphone screen to create full-color 3D holographic images by leveraging computer-generated holography (CGH) and an optical component called a spatial light modulator. The method has the potential to enhance near-eye displays in virtual reality headsets, creating more realistic and interactive user experiences.
Researchers at University of Missouri are developing software that allows drones to fly independently, perceiving and interacting with their environment while achieving specific goals. This technology has the potential to assist in mapping and monitoring applications, such as 3D or 4D advanced imagery for disaster response.
Researchers developed a new tool to improve the accuracy of electronic devices that measure plant leaf color to assess health. The software uses polarization to account for variations in light and reduce errors caused by glare.
Researchers from Kobe University developed an AI image recognition algorithm that can predict mouse behavior based on brain functional imaging data, achieving 95% accuracy. The model identified critical cortical regions for behavioral classification and demonstrated near real-time speeds.
Researchers identify promising methods for early and accurate plant stress detection using AI and imaging sensor technologies. The review highlights the need for precise AI algorithms, diverse datasets, and accessible sensors like RGB cameras.
Researchers at Rice University have developed a custom-built miniaturized chemical vapor deposition (CVD) system that can observe and record the growth of 2D MoS2 crystals in real-time. Through advanced image processing and machine learning algorithms, they were able to extract valuable insights into the growth processes of these mater...
TaskMatrix.AI uses APIs to connect general-purpose foundation models with specialized models for specific tasks. The tool can perform digital and physical tasks, provide interpretable responses, and learn continuously.
A small, wearable ultrasound sticker can monitor organ stiffness and detect subtle changes that signal disease progression. The device has been shown to identify early signs of acute liver failure in rats and may one day help doctors diagnose internal organ failure more effectively.
A new depth from focus/defocus approach, DDFS, combines model-based and learning-based strategies to achieve notable improvements in performance and applicability. The proposed method outperformed state-of-the-art methods in various metrics for several image datasets.
Researchers developed a physical image denoiser using diffractive layers to process noisy input images at the speed of light, preserving desired spatial features with minimal distortions. The all-optical approach offers ultra-high speed, compact size, and low power consumption, paving the way for various inverse problems in imaging and...
A new machine learning approach combines computer vision with deep-learning algorithms to pinpoint problem areas in concrete structures. The system enables efficient identification and inspection of cracks using autonomous robots, reducing the overall inspection workload.
Researchers at UTSA's Department of Physics and Astronomy have used deconvolution algorithms to enhance images of galaxy NGC 5728 obtained by the James Webb Space Telescope. The study reveals a faint extended feature that could be part of an outflow from a supermassive black hole interacting with the host galaxy.
Researchers have developed a new approach to monitor ultrafast charge motion in strongly correlated solids, demonstrating phase transitions within femtoseconds. The technique offers sub-cycle temporal resolution and opens up new avenues for investigating ultrafast phenomena in correlated materials.
Researchers developed a carbon-based tunable metasurface absorber with an ultrawide, tunable bandwidth in the THz range. The absorber boasts high absorption efficiency and insensitivity to polarization angles, paving the way for advanced technological applications.
MIT researchers have developed a new method to track cell differentiation and study long-term processes like cancer progression or embryonic development. They used noninvasive Raman spectroscopy to monitor embryonic stem cells as they differentiated into multiple cell types over several days.
Scientists use new photography technique to observe interaction between biological cell and shock wave, revealing high-speed phenomena. The technology has potential applications in science, medicine, and industry.
A research team developed an innovative optical technique, 'spectrum shuttle,' to produce and shape GHz burst pulses. The method facilitates ultrafast imaging within subnanosecond timescales, enabling analysis of rapid phenomena.
Novel Dice loss functions, t-vMF Dice loss and Adaptive t-vMF Dice loss, have been developed to improve image segmentation accuracy in medical images. These new functions outperform conventional formulations and show great potential for critical fields like medical imaging and diagnosis.
Researchers used brain imaging and machine learning to identify distinct patterns of brain connectivity in people with autism spectrum disorder (ASD), taking into account individual differences. The study reveals that certain brain features are shared across subtypes, while others are unique to specific individuals.
Researchers at MIT have developed an alternative method to study molecular signals in cells, allowing them to track up to seven different molecules simultaneously. The technique uses fluorescent proteins that flicker on and off at different rates, enabling the tracking of specific cellular functions over time.
Researchers developed a deep convolutional neural network to pinpoint cardiac catheter tip locations in photoacoustic images, achieving high precision and recall. The approach has the potential to replace fluoroscopy during cardiac interventions, leading to safer procedures.
A team of researchers has developed a novel experimental system to simultaneously measure the mechanical properties and internal structure of rubber-like materials. The study found that strain within these materials is non-uniform, depending on the shape and size of composite particles.
Researchers developed an imaging sensor capable of detecting UV light, using it to differentiate between cancer cells and normal cells with 99% confidence. The technology leverages the unique tiered structure of butterfly photoreceptors and perovskite nanocrystals.
Researchers developed a new OCT approach to directly image coordination of tiny hair-like structures in live organisms, giving a powerful tool to investigate cilia's role in the female reproductive system. The technique revealed unexpected behaviors that contradict current views and suggested new roles for cilia.
A new method called TWC-Swin effectively restores holographic images even under low spatial coherence and arbitrary turbulence, surpassing traditional convolutional network-based methods. The study demonstrates strong generalization capabilities, extending its application to unseen scenes.
A team of researchers proposes a novel approach to generate three-dimensional holograms directly from regular 2D color images captured using ordinary cameras. This approach utilizes deep learning to transform the image into data that can be used to display a 3D scene or object as a hologram.
A novel strategy utilizing phosphorus nanolayers mitigates electrode-level heterogeneity in fast-charging lithium-ion batteries. The graphite-phosphorus composite exhibits consistent cycle retention, high Coulombic efficiency, and improved lithiation uniformity.
Osaka University researchers created a radial-coded mask that replaces conventional masks, yielding sharp images at various distances. The optimized mask design extends the depth of field, enabling better focus on both foreground and background objects.
New plant cell walls exhibit significantly different mechanical properties compared to surrounding parental walls, enabling cells to alter their local shape and influence the growth of plant organs. Researchers have discovered that new cell walls in some plants are 1.5 times stiffer than the parental cell walls.
Researchers have developed high-resolution near-eye displays with integrated light field technology, overcoming limitations of earlier displays. The new designs feature improved resolution, pixel density, and vision correction capabilities, resulting in enhanced visual comfort and immersive VR experiences.
A new study uses brain imaging to explore the neural underpinnings of uncertainty processing in OCD, finding that patients struggle with certainty and decision-making, particularly when outcomes are more certain. This understanding may offer new hope for developing therapies targeting uncertainty processing in OCD.
The Stevens INI is expanding its health disparities research with a new award to examine vascular causes of dementia in Asian Americans. The project aims to use innovative imaging techniques and clinical translations to better understand cerebral small vessel disease (CSVD) and its impact on cognitive impairment and dementia.
Researchers at Shanghai Jiao Tong University have developed a new scattering matrix method that can sculpt light output with minimal optimization time. The method offers unparalleled nonlinear scattered light control, enabling high-resolution scanning microscopy and particle trapping through dense, scattering media.
Scientists at Beijing Institute of Technology have developed an ultrafast quasi-three-dimensional technique, enabling higher dimensions to analyze ultrafast processes. This method breaks through the limitations of original observational dimensions, enhancing our ability to analyze ultra-fast processes comprehensively.
Studies reveal hyper-connectivity of VTA with visual processing regions and hypo-connectivity with cognitive control regions in patients with obesity. This imbalance is associated with food craving and cognition, but not impulsivity or hunger.
Researchers developed a neural network-based system, PAT, for snapshot compressive imaging. It achieves comparable image quality to CASSI and holds strong promise due to advancements in AI processing capabilities.
A new imaging method developed by UC researchers can identify specific lung infections in real time, reducing diagnosis time from 2-3 days to hours. The method uses injectable probes that light up under a nuclear imaging machine, enabling faster treatment for critically ill patients.
Advances in cardiac imaging enable a multifaceted cardiac evaluation, reducing cost and procedural adverse events while maintaining accuracy. This new approach allows for personalized and high-value care, reserving invasive procedures for cases requiring intervention.
Scientists at Max-Planck-Institut für Eisenforschung developed a machine learning model that enhances predictive accuracy in alloy design, uncovering new corrosion-resistant compositions. The model combines numerical and textual data, enabling the identification of optimal alloy formulas.
Researchers have developed a new measurement technique that uses the Kramers-Kronig relation to untangle complex helical light patterns from camera intensity measurements. This allows for single-shot retrieval of orbital angular momentum spectrum information, accelerating and simplifying the process compared to conventional on-axis int...
Quantum ghost imaging allows 3D imaging on a single photon level, enabling the lowest photon dose possible. The technique can be applied to image materials and tissues sensitive to light or drugs without risk of damage.
Researchers created a self-supervised AI model called GedankenNet that learns physics laws and thought experiments to reconstruct microscopic images. The model successfully reconstructed human tissue samples and Pap smears from holograms without relying on real-world experiments or data.
Researchers have discovered a way to utilize nonlinear scattering media for optical computing and machine learning. They created a novel theoretical framework involving third-order tensors, which can represent the complex relationships between input and output signals. This breakthrough has potential applications in real-world settings...
Researchers developed a new method for capturing turbulent flame behavior, providing detailed insights into flame dynamics, ignition processes and combustion efficiency. The high-speed 3D imaging approach can be used to optimize energy production processes and improve fire safety measures.
A newly developed P-VAE framework could speed up computational imaging by reducing the number of measurements required, making it suitable for applications such as scientific discovery and medical imaging. Researchers applied P-VAE to LED microscopy and computed tomography, achieving improved reconstruction with sparse measurements.
A new geometric deep learning model called GFCN has been developed to detect stroke lesions in brain imaging scans. The model leverages rich geometric information to segment brain tissue and achieves higher segmentation performance than other neural network architectures.
Researchers from Edith Cowan University developed software to analyze bone density scans, detecting abdominal aortic calcification (AAC) at high accuracy. The AI can predict cardiovascular disease events and late-life dementia, allowing for early interventions.
Researchers developed a chest imaging protocol using photon-counting CT, allowing for simultaneous evaluation of lung structure, ventilation, vasculature, and perfusion. The protocol showed advantages over standard CT, providing high image quality at lower radiation doses and better spectral resolution.