Search NASA⌕ Search

SEARCH · Search NASA

Results for “image detection”

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.

At least 397 records · Page 22
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Using AI Segmentation Models to Improve Foreign Body Detection and Triage from Ultrasound Images

Medical imaging can be a critical tool for triaging casualties in trauma situations. In remote or military medicine scenarios, triage is essential for identifying how to use limited resources or prioritize evacuation for the most serious cases. Ultrasound imaging, while portable and often available near the point of injury, can only be used for triage if images are properly acquired, interpreted, and objectively triage scored. Here, we detail how AI segmentation models can be used for improving image interpretation and objective triage evaluation for a medical application focused on foreign bodies embedded in tissues at variable distances from critical neurovascular features. Ultrasound images previously collected in a tissue phantom with or without neurovascular features were labeled with ground truth masks. These image sets were used to train two different segmentation AI frameworks: YOLOv7 and U-Net segmentation models. Overall, both approaches were successful in identifying shrapnel in the image set, with U-Net outperforming YOLOv7 for single-class segmentation. Both segmentation models were also evaluated with a more complex image set containing shrapnel, artery, vein, and nerve features. YOLOv7 obtained higher precision scores across multiple classes whereas U-Net achieved higher recall scores. Using each AI model, a triage distance metric was adapted to measure the proximity of shrapnel to the nearest neurovascular feature, with U-Net more closely mirroring the triage distances measured from ground truth labels. Overall, the segmentation AI models were successful in detecting shrapnel in ultrasound images and could allow for improved injury triage in emergency medicine scenarios.

47 OTHER INSTRUMENTATION↗
DOE OSTI2021

Haze Mitigation in High-Resolution Satellite Imagery using Enhanced Style-Transfer Neural Network and Normalization Across Multiple GPUs

Despite recent advances in deep learning approaches, haze mitigation in large satellite images is still a challenging problem. Due to amorphous nature of haze, object detection or image segmentation approaches are not applicable. Also it is practically infeasible to obtain ground truths for training. Bounded memory capacity of GPUs is another constraint that limits the size of image to be processed. In this paper, we propose a style transfer based neural network approach to mitigate haze in a large overhead imagery. The network is trained without paired ground truths; further, perception loss is added to restore vivid colors, enhance contrast and minimize artifacts. The paper also illustrates our use of multiple GPUs in a collective way to produce a single coherent clear image where each GPU dehazes different portions of a large hazy image.

Park, Byung↗
NASA NTRS2000

Real Time Intelligent Target Detection and Analysis with Machine Vision

We present an algorithm for detecting a specified set of targets for an Automatic Target Recognition (ATR) application. ATR involves processing images for detecting, classifying, and tracking targets embedded in a background scene. We address the problem of discriminating between targets and nontarget objects in a scene by evaluating 40x40 image blocks belonging to an image. Each image block is first projected onto a set of templates specifically designed to separate images of targets embedded in a typical background scene from those background images without targets. These filters are found using directed principal component analysis which maximally separates the two groups. The projected images are then clustered into one of n classes based on a minimum distance to a set of n cluster prototypes. These cluster prototypes have previously been identified using a modified clustering algorithm based on prior sensed data. Each projected image pattern is then fed into the associated cluster's trained neural network for classification. A detailed description of our algorithm will be given in this paper. We outline our methodology for designing the templates, describe our modified clustering algorithm, and provide details on the neural network classifiers. Evaluation of the overall algorithm demonstrates that our detection rates approach 96% with a false positive rate of less than 0.03%.

Howard, Ayanna↗
NASA NTRS2013

Multiple-Frame Detection of Subpixel Targets in Thermal Image Sequences

The new technology in this approach combines the subpixel detection information from multiple frames of a sequence to achieve a more sensitive detection result, using only the information found in the images themselves. It is taken as a constraint that the method is automated, robust, and computationally feasible for field networks with constrained computation and data rates. This precludes simply downloading a video stream for pixel-wise co-registration on the ground. It is also important that this method not require precise knowledge of sensor position or direction, because such information is often not available. It is also assumed that the scene in question is approximately planar, which is appropriate for a high-altitude airborne or orbital view.

Thompson, David R.↗
NASA NTRS1975

The detection of geothermal areas from Skylab thermal data

Skylab-4 X-5 thermal data of the geysers area was analyzed to determine the feasibility of using midday Skylab images to detect geothermal areas. The hottest ground areas indicated on the Skylab image corresponded to south-facing barren or sparsely vegetated slopes. A geothermal area approximately 15 by 30 m coincided with one of the hottest areas indicated by Skylab. This area could not be unambiguously distinguished from the other areas which are believed to be hotter than their surroundings as a result of their topography, and micrometeorological conditions. A simple modification of a previous thermal model was performed and the predicted temperatures for the hottest slopes using representative values was in general agreement with the observed data. It is concluded that data from a single midday Skylab pass cannot be used to locate geothermal areas.

Siegal, B. S.↗
NASA NTRS1994

A model-based approach for detection of runways and other objects in image sequences acquired using an on-board camera

This research was initiated as a part of the Advanced Sensor and Imaging System Technology (ASSIST) program at NASA Langley Research Center. The primary goal of this research is the development of image analysis algorithms for the detection of runways and other objects using an on-board camera. Initial effort was concentrated on images acquired using a passive millimeter wave (PMMW) sensor. The images obtained using PMMW sensors under poor visibility conditions due to atmospheric fog are characterized by very low spatial resolution but good image contrast compared to those images obtained using sensors operating in the visible spectrum. Algorithms developed for analyzing these images using a model of the runway and other objects are described in Part 1 of this report. Experimental verification of these algorithms was limited to a sequence of images simulated from a single frame of PMMW image. Subsequent development and evaluation of algorithms was done using video image sequences. These images have better spatial and temporal resolution compared to PMMW images. Algorithms for reliable recognition of runways and accurate estimation of spatial position of stationary objects on the ground have been developed and evaluated using several image sequences. These algorithms are described in Part 2 of this report. A list of all publications resulting from this work is also included.

Kasturi, Rangachar↗
DOE OSTI2025

One-shot gas detection with transformer paired neural networks in Mako collected longwave infrared hyperspectral imagery

To date, careful data treatment workflows and statistical detectors are used to perform hyperspectral image (HSI) detection of any gas contained in a spectral library, which is often expanded with physics models to incorporate different spectral characteristics. In general, surrounding evidence or known gas-release parameters are used to provide confidence in or confirm detection capability, respectively. This makes quantifying detection performance difficult as it is nearly impossible to develop an absolute ground truth for gas target pixel presence in collected HSI. Consequently, developing and comparing new detection methods, especially machine learning (ML)-based methods, is susceptible to subjectivity in derived detection map quality. Here, in this work, we demonstrate the first use of transformer-based paired neural networks (PNNs) for one-shot gas target detection for multiple gases while providing quantitative classification and detection metrics for their use on labeled data. Terabytes of training data are generated from a database of long-wave infrared HSI obtained from historical Mako sensor campaigns over Los Angeles. By incorporating labels, singular signature representations, and a model development pipeline, we can tune and select PNNs to detect multiple gas targets that are not seen in training on a quantitative basis. We additionally assess our test set detections using interpretability techniques widely employed with ML-based predictors, but less common with detection methods relying on learned latent spaces.

Hyperspectral imaging↗
NASA NTRS1975

Temporal registration of multispectral digital satellite images using their edge images

An algorithm is described which will form an edge image by detecting the edges of features in a particular spectral band of a digital satellite image. It is capable also of forming composite multispectral edge images. In addition, an edge image correlation algorithm is presented which performs rapid automatic registration of the edge images and, consequently, the grey level images.

Nack, M. L.↗
NASA NTRS2023

Plant Disease Detection: Exploring Applications in Hyperspectral Imaging and Machine Learning for Agriculture

The threat of crop disease on food security and agriculture is projected to escalate, leading to reduced crop yields, global economic loss, and endangered food availability to vulnerable populations. Early identification of plant disease is crucial to combatting the crisis but traditional manual methods for disease detection are laborious and may miss early signs of infection. The proposed solution suggests equipping a drone with a hyperspectral camera to collect images and analyzing the data with neural networks trained to flag and classify infected plants. This approach offers a faster, more accurate, and potentially more cost-effective alternative to current practices. While the expensive and complex nature of hyperspectral imaging (HSI) may be an obstacle to the adoption of the drone, rapidly advancing technologies compounded with rental-based usage may make these tools simpler, cheaper, and more accessible to a wider audience. The research further discusses the potential for automated flight paths and the expansion of disease detection to a broader range of crops.

plant disease detection↗
DOE OSTI2023

Computer Vision: EO Imaging & Video Tasks, Metrics, and Datasets [Slides]

Classification is an assignment of an image to a certain category. Localization is identified by the box surrounding the object in the image. Object detection detects instances of semantic objects of a certain class. Semantic segmentation categorizes all the pixels of an image into classes of objects. Instance segmentation identifies each instance of each object in the image. Panoptic (i.e., showing or seeing everything at once) segmentation is a combination of semantic and instance segmentation and classifies all the pixels in the image.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Implementation of heterodyne detection of radio frequency waves using phase contrast imaging

An upgrade to the DIII–D phase contrast imaging (PCI) diagnostic for detection of radio frequency waves has been implemented using low-noise optical heterodyning. Electro-optical modulation allows detection of waves with frequencies up to 500 MHz with wide flexibility restricted only by the driver electronics, with continuous adjustability up to 100 MHz and fixed frequency drivers needed for higher frequencies. The PCI diagnostic images the line-integrated plasma density fluctuations and provides absolutely calibrated measurements with good wavenumber resolution.

Rost, J. C. [Massachusetts Institute of Technology↗
DOE OSTI2023

Towards deep computer vision for in-line defect detection in polymer electrolyte membrane fuel cell materials

Polymer Electrolyte Membrane (PEM) fuel cells are a promising source of alternative energy. However, their production is limited by a lack of well-established methods for quality control of their constituent materials like the membrane-electrode assembly during roll-to-roll manufacturing. One potential solution is the implementation of deep learning methods to detect unwanted defects through their detection in scanned images. Here we explore the detection of defects like scratches, pinholes, and scuffs in a sample dataset of PEM optical images using two deep learning algorithms: Patch Distribution Modeling (PaDiM) for unsupervised anomaly detection and Faster-RCNN for supervised object detection. Both methods achieve scores on performance metrics (ROC-AUC and PRO-AUC for PaDiM and AP for Faster-RCNN) that are comparable to their scores on benchmark datasets. These methods also have the potential to detect a wider range of defects compared to IR thermography and previous optical inspection methods. Overall, deep learning shows promise at detecting relevant defects of interest and has the potential to achieve real-time defect detection.

30 DIRECT ENERGY CONVERSION↗
DOE OSTI2024

Reinforcement Learning as a Parsimonious Alternative to Prediction Cascades: A Case Study on Image Segmentation

Deep learning architectures have achieved state-of-the-art (SOTA) performance on computer vision tasks such as object detection and image segmentation. This may be attributed to the use of over-parameterized, monolithic deep learning architectures executed on large datasets. Although such large architectures lead to increased accuracy, this is usually accompanied by a larger increase in computation and memory requirements during inference. While this is a non-issue in traditional machine learning (ML) pipelines, the recent confluence of machine learning and fields like the Internet of Things (IoT) has rendered such large architectures infeasible for execution in low-resource settings. For some datasets, large monolithic pipelines may be overkill for simpler inputs. To address this problem, previous efforts have proposed decision cascades where inputs are passed through models of increasing complexity until the desired performance is achieved. However, we argue that cascaded prediction leads to sub-optimal throughput and increased computational cost due to wasteful intermediate computations. To address this, we propose PaSeR (Parsimonious Segmentation with Reinforcement Learning) a non-cascading, cost-aware learning pipeline as an efficient alternative to cascaded decision architectures. Through experimental evaluation on both real-world and standard datasets, we demonstrate that PaSeR achieves better accuracy while minimizing computational cost relative to cascaded models. Further, we introduce a new metric IoU/GigaFlop to evaluate the balance between cost and performance. On the real-world task of battery material phase segmentation, PaSeR yields 179% improvement over SOTA MatPhase model and a 196% improvement over IDK Cascades under the IoU/GigaFlop metric. We also demonstrate PaSeR’s adaptability to complementary models trained on a noisy MNIST dataset, where it outperforms all baselines on IoU/GigaFlop by an average of 44%.

Srikshan, Bharat↗
DOE OSTI2020

Detection of defects in atomic-resolution images of materials using cycle analysis

The automated detection of defects in high-angle annular dark-field Z-contrast (HAADF) scanning-transmission-electron microscopy (STEM) images has been a major challenge. Here, we report an approach for the automated detection and categorization of structural defects based on changes in the material’s local atomic geometry. The approach applies geometric graph theory to the already-found positions of atomic-column centers and is capable of detecting and categorizing any defect in thin diperiodic structures (i.e., “2D materials”) and a large subset of defects in thick diperiodic structures (i.e., 3D or bulk-like materials). Despite the somewhat limited applicability of the approach in detecting and categorizing defects in thicker bulk-like materials, it provides potentially informative insights into the presence of defects. The categorization of defects can be used to screen large quantities of data and to provide statistical data about the distribution of defects within a material. This methodology is applicable to atomic column locations extracted from any type of high-resolution image, but here we demonstrate it for HAADF STEM images.

36 MATERIALS SCIENCE↗
NASA NTRS2008

Terahertz Imaging and Backscatter Radiography Probability of Detection Study for Space Shuttle Foam Inspections

This slide presentation reviews the use of terahertz imaging and Backscatter Radiography in a probability of detection study of the foam on the external tank (ET) shedding and damaging the shuttle orbiter. Non-destructive Examination (NDE) is performed as one method of preventing critical foam debris during the launch. Conventional NDE methods for inspection of the foam are assessed and the deficiencies are reviewed. Two methods for NDE inspection are reviewed: Backscatter Radiography (BSX) and Terahertz (THZ) Imaging. The purpose of the Probability of Detection (POD) study was to assess performance and reliability of the use of BSX and or THZ as an appropriate NDE method. The study used a test article with inserted defects, and a sample of blanks included to test for false positives. The results of the POD study are reported.

Ussery, Warren↗
DOE OSTI2020

FCIC FY18 LT-baseline Feedstock Images

Set of images taken during pilot plant dilute-acid hydrolysis preatreatment reactor operation. Images were taken with a GoPro Hero 5 with rectilinear lens facing downwards on a weigh-belt instrument. Image period was every 30 seconds during operation. This image set is connected to a journal article titled "Real-Time Biomass Feedstock Particle Quality Detection Using Image Analysis and Machine Vision", http://dx.doi.org/10.1007/s13399-020-00904-w.

09 BIOMASS FUELS↗
NASA NTRS2012

Thermal Imaging for Inspection of Large Cryogenic Tanks

The end of the Shuttle Program provides an opportunity to evaluate and possibly refurbish launch support infrastructure at the Kennedy Space Center in support of future launch vehicles. One major infrastructure element needing attention is the cryogenic fuel and oxidizer system and specifically the cryogenic fuel ground storage tanks located at Launch Complex 39. These tanks were constructed in 1965 and served both the Apollo and Shuttle Programs and will be used to support future launch programs. However, they have received only external inspection and minimal refurbishment over the years as there were no operational issues that warranted the significant time and schedule disruption required to drain and refurbish the tanks while the launch programs were ongoing. Now, during the break between programs, the health of the tanks is being evaluated and refurbishment is being performed as necessary to maintain their fitness for future launch programs. Thermography was used as one part of the inspection and analysis of the tanks. This paper will describe the conclusions derived from the thermal images to evaluate anomalous regions in the tanks, confirm structural integrity of components within the annular region, and evaluate the effectiveness of thermal imaging to detect large insulation voids in tanks prior to filling with cryogenic fluid. The use of thermal imaging as a tool to inspect unfilled tanks will be important if the construction of additional storage tanks is required to fuel new launch vehicles.

Arens, Ellen↗