Plant Health Monitoring Using Multimodal Imaging
Explore the source record for details and available documents.
Aubrie O'Rourke↗
SEARCH · Engineering Papers
Results for “multimodal imaging”
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
High Resolution Multimode Fiber Image Recovery
The research emphasis is on developing a cost-effective method of recovering image information from small, closely confined spaces using multimode fibers. The state-of-the-art good quality-viewing fiber, which can currently be used for performing this function, is a 0.5 mm diameter bundle containing 6000 pixels at a cost of $10,000 per fiber bundle. However, these fiber bundles are very fragile and can easily break during surgical use, thereby making instrument reliability and replacement cost,a major impediment to their routine use in many applications. The advantage of working with a single multimode fiber is that it is significantly less expensive and mechanically more robust. In addition, careful choice of numerical aperture allows a higher image resolution (roughly 750,000 pixels) with a 0.5 mm diameter multimode fiber.
Jackson, Deborah↗
Multimode-Optical-Fiber Imaging Probe
Currently, endoscopic surgery uses single-mode fiber-bundles to obtain in vivo image information inside the orifices of the body. This limits their use to the larger natural orifices and to surgical procedures where there is plenty of room for manipulation. The knee joint, for example, can be easily viewed with a fiber optic viewer, but joints in the finger cannot. However, there are a host of smaller orifices where fiber endoscopy would play an important role if a cost effective fiber probe were developed with small enough dimensions (less than or equal to 250 microns). Examples of beneficiaries of micro-endoscopes are the treatment of the Eustatian tube of the middle ear, the breast ducts, tear ducts, coronary arteries, fallopian tubes, as well as the treatment of salivary duct parotid disease, and the neuro endoscopy of the ventricles and spinal canal. This work describes an approach for recovering images from tightly confined spaces using multimode. The concept draws upon earlier works that concentrated on image recovery after two-way transmission through a multimode fiber as well as work that demonstrated the recovery of images after one-way transmission through a multimode fiber. Both relied on generating a phase conjugated wavefront, which was predistorted with the characteristics of the fiber. The approach described here also relies on generating a phase conjugated wavefront, but utilizes two fibers to capture the image at some intermediate point (accessible by the fibers, but which is otherwise visually inaccessible).
Jackson, Deborah↗
Multimode-Optical-Fiber Imaging Probe
Currently, endoscopic surgery uses single-mode fiber-bundles to obtain in vivo image information inside orifices of the body. This limits their use to the larger natural bodily orifices and to surgical procedures where there is plenty of room for manipulation. The knee joint, for example can be easily viewed with a fiber optic viewer, but joints in the finger cannot. However, there are a host of smaller orifices where fiber endoscopy would play an important role if a cost effective fiber probe were developed with small enough dimensions (< 250 microns). Examples of beneficiaries of micro-endoscopes are the treatment of the Eustatian tube of the middle ear, the breast ducts, tear ducts, coronary arteries, fallopian tubes, as well as the treatment of salivary duct parotid disease, and the neuro endoscopy of the ventricles and spinal canal. To solve this problem, this work describes an approach for recovering images from. tightly confined spaces using multimode fibers and analytically demonstrates that the concept is sound. The proof of concept draws upon earlier works that concentrated on image recovery after two-way transmission through a multimode fiber as well as work that demonstrated the recovery of images after one-way transmission through a multimode fiber. Both relied on generating a phase conjugated wavefront which was predistorted with the characteristics of the fiber. The described approach also relies on generating a phase conjugated wavefront, but utilizes two fibers to capture the image at some intermediate point (accessible by the fibers, but which is otherwise visually unaccessible).
Jackson, Deborah↗
Using Electronic Noses to Detect Tumors During Neurosurgery
It has been proposed to develop special-purpose electronic noses and algorithms for processing the digitized outputs of the electronic noses for determining whether tissue exposed during neurosurgery is cancerous. At present, visual inspection by a surgeon is the only available intraoperative technique for detecting cancerous tissue. Implementation of the proposal would help to satisfy a desire, expressed by some neurosurgeons, for an intraoperative technique for determining whether all of a brain tumor has been removed. The electronic-nose technique could complement multimodal imaging techniques, which have also been proposed as means of detecting cancerous tissue. There are also other potential applications of the electronic-nose technique in general diagnosis of abnormal tissue. In preliminary experiments performed to assess the viability of the proposal, the problem of distinguishing between different types of cultured cells was substituted for the problem of distinguishing between normal and abnormal specimens of the same type of tissue. The figure presents data from one experiment, illustrating differences between patterns that could be used to distinguish between two types of cultured cancer cells. Further development can be expected to include studies directed toward answering questions concerning not only the possibility of distinguishing among various types of normal and abnormal tissue but also distinguishing between tissues of interest and other odorous substances that may be present in medical settings.
Homer, Margie L.↗
Multimode-Optical-Fiber Imaging Probe
This work describes an approach for recovering images from tightly confined spaces using multimode.
phase↗
Parallel Information Processing (Image Transmission Via Fiber Bundle and Multimode Fiber
Growing demand for visual, user-friendly representation of information inspires search for the new methods of image transmission. Currently used in-series (sequential) methods of information processing are inherently slow and are designed mainly for transmission of one or two dimensional arrays of data. Conventional transmission of data by fibers requires many fibers with array of laser diodes and photodetectors. In practice, fiber bundles are also used for transmission of images. Image is formed on the fiber-optic bundle entrance surface and each fiber transmits the incident image to the exit surface. Since the fibers do not preserve phase, only 2D intensity distribution can be transmitted in this way. Each single mode fiber transmit only one pixel of an image. Multimode fibers may be also used, so that each mode represent different pixel element. Direct transmission of image through multimode fiber is hindered by the mode scrambling and phase randomization. To overcome these obstacles wavelength and time-division multiplexing have been used, with each pixel transmitted on a separate wavelength or time interval. Phase-conjugate techniques also was tested in, but only in the unpractical scheme when reconstructed image return back to the fiber input end. Another method of three-dimensional imaging over single mode fibers was demonstrated in, using laser light of reduced spatial coherence. Coherence encoding, needed for a transmission of images by this methods, was realized with grating interferometer or with the help of an acousto-optic deflector. We suggest simple practical holographic method of image transmission over single multimode fiber or over fiber bundle with coherent light using filtering by holographic optical elements. Originally this method was successfully tested for the single multimode fiber. In this research we have modified holographic method for transmission of laser illuminated images over commercially available fiber bundle (fiber endoscope, or fiberscope).
Kukhtarev, Nicholai↗
Real time quantitative imaging for semiconductor crystal growth, control and characterization
A quantitative real time image processing system has been developed which can be software-reconfigured for semiconductor processing and characterization tasks. In thermal imager mode, 2D temperature distributions of semiconductor melt surfaces (900-1600 C) can be obtained with temperature and spatial resolutions better than 0.5 C and 0.5 mm, respectively, as demonstrated by analysis of melt surface thermal distributions. Temporal and spatial image processing techniques and multitasking computational capabilities convert such thermal imaging into a multimode sensor for crystal growth control. A second configuration of the image processing engine in conjunction with bright and dark field transmission optics is used to nonintrusively determine the microdistribution of free charge carriers and submicron sized crystalline defects in semiconductors. The IR absorption characteristics of wafers are determined with 10-micron spatial resolution and, after calibration, are converted into charge carrier density.
Wargo, Michael J.↗
Use of anomolous thermal imaging effects for multi-mode systems control during crystal growth
Real time image processing techniques, combined with multitasking computational capabilities are used to establish thermal imaging as a multimode sensor for systems control during crystal growth. Whereas certain regions of the high temperature scene are presently unusable for quantitative determination of temperature, the anomalous information thus obtained is found to serve as a potentially low noise source of other important systems control output. Using this approach, the light emission/reflection characteristics of the crystal, meniscus and melt system are used to infer the crystal diameter and a linear regression algorithm is employed to determine the local diameter trend. This data is utilized as input for closed loop control of crystal shape. No performance penalty in thermal imaging speed is paid for this added functionality. Approach to secondary (diameter) sensor design and systems control structure is discussed. Preliminary experimental results are presented.
Wargo, Michael J.↗
Technique for Recovering Images from Tightly Confined Spaces
This work describes an approach for recovering images from tightly confined spaces using multimode and analytically demonstrates that the concept is sound.
confined spaces recovering images multimode↗
A light intensity function suitable for multimode fiber-optic sensors
Intensity distribution functions on the image plane are derived under the assumption that a multimode optic fiber is illuminated by a light source which can be approximated by infinite uniangular beams carrying equal power. The application of the theory to a Lambertian source is discussed. Experimental results, based on the observation of the light intensity distribution imaged on an opaque screen and collected by a digital camera, are presented. The theoretical and experimental data are compared to verify that the intensity function is an accurate model for describing the realistic approach to sensor design. The limitations of the uniangular and uniform intensity distribution models are discussed on the basis of theoretical and experimental results.
He, Gang↗
Geomorphologic Studies of a Very Long Lava Flow in Tharsis, Mars
CEPS has undertaken an extended study of long lava flows on the terrestrial planets, their location, morphology, and potential modes of emplacement. As part of this ongoing investigation, we have concentrated on a single large flow in Tharsis, with noted similarities to several terrestrial analogs. An impressive series of lava flows emerges from the topographic saddle between Ascraeus and Pavonis Mons. The most prominent of these (hereafter referred to as the 'Saddle Flow') has distinct margins that can be traced for over 480 km in the Viking images, although its exact source cannot be identified. A multimodal approach is utilized in the examination of the Saddle Flow, including image interpretation (VIKING and THEMIS, MOLA topographic analysis and flow profiling, downflow behavior statistical analysis, rheologic modeling, and GIS modeling and integration.
Peitersen, M. N.↗
Technology transfer: Imaging tracker to robotic controller
The transformation of an imaging tracker to a robotic controller is described. A multimode tracker was developed for fire and forget missile systems. The tracker locks on to target images within an acquisition window using multiple image tracking algorithms to provide guidance commands to missile control systems. This basic tracker technology is used with the addition of a ranging algorithm based on sizing a cooperative target to perform autonomous guidance and control of a platform for an Advanced Development Project on automation and robotics. A ranging tracker is required to provide the positioning necessary for robotic control. A simple functional demonstration of the feasibility of this approach was performed and described. More realistic demonstrations are under way at NASA-JSC. In particular, this modified tracker, or robotic controller, will be used to autonomously guide the Man Maneuvering Unit (MMU) to targets such as disabled astronauts or tools as part of the EVA Retriever efforts. It will also be used to control the orbiter's Remote Manipulator Systems (RMS) in autonomous approach and positioning demonstrations. These efforts will also be discussed.
Otaguro, M. S.↗
Learning instrument invariant characteristics for generating high-resolution global coral reef maps
Coral reefs are one of the most biologically complex and diverse ecosystems within the shallow marine environment. Unfortunately, these underwater ecosystems are threatened by a number of anthropogenic challenges, including ocean acidification and warming, overfishing, and the continued increase of marine debris in oceans. This requires a comprehensive assessment of the world's coastal environments, including a quantitative analysis on the health and extent of coral reefs and other associated marine species, as a vital Earth Science measurement. However, limitations in observational and technological capabilities inhibit global sustained imaging of the marine environment. Harmonizing multimodal data sets acquired using different remote sensing instruments presents additional challenges, thereby limiting the availability of good quality labeled data for analysis. In this work, we develop a deep learning model for extracting domain invariant features from multimodal remote sensing imagery and creating high-resolution global maps of coral reefs by combining various sources of imagery and limited hand-labeled data available for certain regions. This framework allows us to generate, for the first time, coral reef segmentation maps at 2-meter resolution, which is a significant improvement over the kilometer-scale state-of-the-art maps. Additionally, this framework doubles accuracy and IoU metrics over baselines that do not account for domain invariance.
Domain Adaptation↗
Asymmetric Flow State Switching in the Space Launch System Block 2 Liftoff Environment
An in-depth investigation was conducted at the NASA Langley Research Center 14- by 22-Foot Subsonic Tunnel to evaluate flow state switching in the liftoff flow environment of the Space Launch System Block 2 Crew launch vehicle using force measurements, two-component particle image velocimetry, and tuft flow visualization. Multimodal flow states were observed in the gap flow between the centerbody and solid rocket boosters at a range of incoming flow angles and were characterized for the relative strength of state switches. When flow switches occur, the flow is predominantly bimodal, but trimodal flow states are observed with the launch tower downstream of the vehicle. Tuft visualization indicates three-dimensionality during flow state switching, which initiates at one part of the gap and quickly transitions throughout the length of the booster. Due to the long time scales between switches, statistics such as frequency of flow state switches and converged state probabilities cannot be established without significantly increased acquisition times.
Lee J. Mears↗
Asymmetric Flow State Switching on the Space Launch System (SLS) Block 2 Crew Vehicle
An in-depth investigation was conducted at the NASA Langley Research Center 14- by 22-Foot Subsonic Tunnel to evaluate flow state switching in the liftoff flow environment of the Space Launch System Block 2 Crew launch vehicle using side force measurements, two-component particle image velocimetry, and tufts flow visualization. Multimodal flow states were observed in the gap flow between the centerbody and solid rocket boosters at a range of incoming flow angles and were characterized for the relative strength of state switches. When flow switches occur, the flow is predominantly bimodal, but trimodal flow states are observed with the launch tower downstream of the vehicle. Tufts visualization indicates three-dimensionality during flow state switching, which initiates at one part of the gap and quickly transitions throughout the length of the booster. Due to the long time scales between switches, statistics such as frequency of flow state switches and converged state probabilities cannot be established without significantly increased acquisition times.
Lee J. Mears↗