SEARCH · Engineering Papers

Results for “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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 829 records · Page 46

All sources NASA NTRS DOE OSTI

DOE OSTI2018

Quench Detection for High-Temperature Superconductor Conductors Using Acoustic Thermometry

Detecting local heat-dissipating zones in high-temperature superconductor (HTS) magnets is a challenging task due to slow propagation of such zones in HTS conductors. For long conductor lengths, voltage-based methods may not provide a sufficient sensitivity or redundancy, and therefore nonvoltage-based detection alternatives are being sought. One of those is the recently proposed method of Eigen Frequency Thermometry (EFT), which is an active acoustic technique for a fast and nonintrusive detection of 'hot spots,' utilizing temperature dependence of the conductor elastic moduli. In this work, we demonstrate the efficiency of EFT for detecting localized heating in a 1.2-m-long sample of REBCO tape immersed in liquid nitrogen, and benchmark sensitivity of the acoustic detection with respect to voltage, hot spot temperature, and power dissipation in the conductor. Modifying the original technique for differential mode of operation enables a much improved sensitivity, and adds a hot spot localization capability. Furthermore, we adapt this technique to subscale coils wound with REBCO CORC conductor built in the framework of U.S. Magnet Development Program. A successful thermal-based detection of dissipation onset at the critical current for a two-layer canted CORC dipole assembly is discussed.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

DOE OSTI2025

Detecting Masquerade Attacks in Controller Area Networks Using Graph Machine Learning

Modern vehicles rely on a myriad of electronic control units (ECUs) interconnected via controller area networks (CANs) for critical operations. Despite their ubiquitous use and reliability, CANs are susceptible to sophisticated cyberattacks, particularly masquerade attacks, which inject false data that mimic legitimate messages at the expected frequency. These attacks pose severe risks such as unintended acceleration, brake deactivation, and rogue steering. Traditional intrusion detection systems (IDS) often struggle to detect these subtle intrusions due to their seamless integration into normal traffic. This paper introduces a novel framework for detecting masquerade attacks in the CAN bus using graph machine learning (ML). We hypothesize that the integration of shallow graph embeddings with time series features derived from CAN frames enhances the detection of masquerade attacks. We show that by representing CAN bus frames as message sequence graphs (MSGs) and enriching each node with contextual statistical attributes from time series, we can enhance detection capabilities across various attack patterns compared to using graph-based features only. Our method ensures a comprehensive and dynamic analysis of CAN frame interactions, improving robustness and efficiency. Extensive experiments on the ROAD dataset validate the effectiveness of our approach, demonstrating statistically significant improvements in the detection rates of masquerade attacks compared to a baseline that uses graph-based features only as confirmed by Mann-Whitney U and Kolmogorov-Smirnov tests (p < 0.05) .

Marfo, William [Univ. of Texas, El Paso, TX (Unite↗

DOE OSTI2016

Adaptive Impact-Driven Detection of Silent Data Corruption for HPC Applications

For exascale HPC applications, silent data corruption (SDC) is one of the most dangerous problems because there is no indication that there are errors during the execution. We propose an adaptive impact-driven method that can detect SDCs dynamically. The key contributions are threefold. (1) We carefully characterize 18 real-world HPC applications and discuss the runtime data features, as well as the impact of the SDCs on their execution results. (2) We propose an impact-driven detection model that does not blindly improve the prediction accuracy, but instead detects only influential SDCs to guarantee user-acceptable execution results. (3) Our solution can adapt to dynamic prediction errors based on local runtime data and can automatically tune detection ranges for guaranteeing low false alarms. Experiments show that our detector can detect 80-99.99% of SDCs with a false alarm rate less that 1% of iterations for most cases. The memory cost and detection overhead are reduced to 15% and 6.3%, respectively, for a large majority of applications.

Exascale HPC↗

DOE OSTI2014

SU-E-J-191: Automated Detection of Anatomic Changes in H'N Patients

Purpose: To develop a novel statistics-based method for automated detection of anatomical changes using cone-beam CT data. A method was developed that can provide a reliable and automated early warning system that enables a “just-in-time” adaptation of the treatment plan. Methods: Anatomical changes were evaluated by comparing the original treatment planning CT with daily CBCT images taken prior treatment delivery. The external body contour was computed on a given CT slice and compared against the corresponding contour on the daily CBCT. In contrast to threshold-based techniques, a statistical approach was employed to evaluate the difference between the contours using a given confidence level. The detection tool used the two-sample Kolmogorov-Smirnov test, which is a non-parametric technique that compares two samples drawn from arbitrary probability distributions. 11 H'N patients were retrospectively selected from a clinical imaging database with a total of 186 CBCT images. Six patients in the database were confirmed to have anatomic changes during the course of radiotherapy. Five of the H'N patients did not have significant changes. The KS test was applied to the contour data using a sliding window analysis. The confidence level of 0.99 was used to moderate false detection. Results: The algorithm was able to correctly detect anatomical changes in 6 out of 6 patients with an excellent spatial accuracy as early as at the 14th elapsed day. The algorithm provided a consistent and accurate delineation of the detected changes. The output of the anatomical change tool is easy interpretable, and can be shown overlaid on a 3D rendering of the patient's anatomy. Conclusion: The detection method provides the basis for one of the key components of Adaptive Radiation Therapy. The method uses tools that are readily available in the clinic, including daily CBCT imaging, and image co-registration facilities.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2014

TU-A-9A-05: First Experimental Demonstration of the Anisotropic Detection Principle in X-Ray Fluorescence Computed Tomography

Purpose: To improve the sensitivity of X-ray fluorescence computed tomography (XFCT) for in vivo molecular imaging. Is the maximum sensitivity achieved with an isotropic (4π) detector configuration? We prove that this is not necessarily true, and that a greater sensitivity is possible with anisotropic detector configuration. Methods: An XFCT imaging system was constructed consisting of 1) a collimated pencil beam x-ray source using a fluoroscopy grade x-ray tube; 2) a CdTe x-ray photon counting detector to detect fluorescent x-rays; and 3) a rotation/translation stage for tomographic imaging. We created a 6.5-cm diameter water phantom with 2-cm inserts of low gold concentration (0.25%–1%) to simulate tumors targeted by gold nano-particles. The placement of x-ray fluorescence detector were chosen to minimize scatter x-rays. XFCT imaging was performed at three different detector positions (60°, 90°, 145°) to determine the impact of forward-scatter, side-scatter, and back-scatter on imaging performance. The three data sets were also combined to estimate the imaging performance with an isotropic detector. Results: The highest imaging performance was achieved when the XF detector was in the backscatter 145° configuration. The signal-to-noise ratio (SNR) was 5.5 for the 0.25% gold concentration compared to SNRs of 1.4, 0, and 2.4 for 60°, 90°, and combined (60°+90°+145°) datasets. Only the 145° detector arrangement alone could detect the 0.25% concentration. The imaging dose was 14 mGy for each detector arrangement experiment. Conclusion: This study experimentally proves, for the fist time, the Anisotropic Detection Principle in XF imaging, which holds that optimized anisotropic x-ray fluorescence detection provides greater sensitivity than isotropic detection. The optimized detection arrangement was used to improve the sensitivity of the XFCT experiment. The achieved XFCT sensitivity is the highest ever for a phantom at least this large using a benchtop x-ray source, which is an important step toward clinical XFCT molecular imaging. This work was supported by the NCI fellowship grant R25T-CA118681 and by the NIH (1R01-EB016777) and NIBIB (1K99-EB016059)

60 APPLIED LIFE SCIENCES↗

DOE OSTI2014

TU-F-18C-08: Micro-Calcification Detectability Using Spectral Breast CT Based On a Si Strip Detector

Purpose: To investigate the feasibility of micro-calcification (μCa) detectability by using an energy-resolved photon-counting Si strip detector for spectral breast computed tomography (CT). Methods: A bench-top CT system was constructed using a tungsten anode x-ray source with a focal spot size of 0.8 mm and a single line 256-pixel Si strip photon counting detector with a pixel pitch of 100 μm. The slice thickness was 0.5 mm. Five different size groups of calcium carbonate grains, from 105 to 215 μm in diameter, were embedded in a cylindrical resin phantom with a diameter of 16 mm to simulate μCas. The phantoms were imaged at 65 kVp with an Entrance Skin Air Kerma (ESAK) of 1.2, 3, 6, and 8 mGy. The images were reconstructed using a standard filtered back projection (FBP) with a ramp filter. A total of 200 μCa images (5 different sizes of μCas × 4 different doses × 10 images for each setting) were combined with another 200 control images without μCas, to ultimately form 400 images for the reader study. The images were displayed in random order to three blinded observers, who were asked to give a binary score on each image regarding the presence of μCas. The μCa detectability for each image was evaluated in terms of binary decision theory metrics. The sensitivity, specificity, and accuracy were calculated to study the size and dose-dependence for μCa detectability. Additionally, the influence of the partial volume effect on the μCa detectability was investigated by simulation. Results: For a μCa larger than 140 μm in diameter, detection accuracy of above 90 % was achieved with the investigated prototype spectral CT system at ESAK of 1.2 mGy. Conclusion: The proposed Si strip detector is expected to offer superior image quality with the capability to detect μCas for low dose breast imaging.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2014

A system to detect cerebral aneurysms in multimodality angiographic data sets

Purpose: The early detection of cerebral aneurysms plays a major role in preventing subarachnoid hemorrhage. The authors present a system to automatically detect cerebral aneurysms in multimodal 3D angiographic data sets. The authors’ system is parametrizable for contrast-enhanced magnetic resonance angiography (CE-MRA), time-of-flight magnetic resonance angiography (TOF-MRA), and computed tomography angiography (CTA). Methods: Initial volumes of interest are found by applying a multiscale sphere-enhancing filter. Several features are combined in a linear discriminant function (LDF) to distinguish between true aneurysms and false positives. The features include shape information, spatial information, and probability information. The LDF can either be parametrized by domain experts or automatically by training. Vessel segmentation is avoided as it could heavily influence the detection algorithm. Results: The authors tested their method with 151 clinical angiographic data sets containing 112 aneurysms. The authors reach a sensitivity of 95% with CE-MRA data sets at an average false positive rate per data set (FP{sub DS}) of 8.2. For TOF-MRA, we achieve 95% sensitivity at 11.3 FP{sub DS}. For CTA, we reach a sensitivity of 95% at 22.8 FP{sub DS}. For all modalities, the expert parametrization led to similar or better results than the trained parametrization eliminating the need for training. 93% of aneurysms that were smaller than 5 mm were found. The authors also showed that their algorithm is capable of detecting aneurysms that were previously overlooked by radiologists. Conclusions: The authors present an automatic system to detect cerebral aneurysms in multimodal angiographic data sets. The system proved as a suitable computer-aided detection tool to help radiologists find cerebral aneurysms.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2015

Characteristic image quality of a third generation dual-source MDCT scanner: Noise, resolution, and detectability

Purpose: The purpose of this work was to assess the inherent image quality characteristics of a new multidetector computed tomography system in terms of noise, resolution, and detectability index as a function of image acquisition and reconstruction for a range of clinically relevant settings. Methods: A multisized image quality phantom (37, 30, 23, 18.5, and 12 cm physical diameter) was imaged on a SOMATOM Force scanner (Siemens Medical Solutions) under variable dose, kVp, and tube current modulation settings. Images were reconstructed with filtered back projection (FBP) and with advanced modeled iterative reconstruction (ADMIRE) with iterative strengths of 3, 4, and 5. Image quality was assessed in terms of the noise power spectrum (NPS), task transfer function (TTF), and detectability index for a range of detection tasks (contrasts of approximately 45, 90, 300, −900, and 1000 HU, and 2–20 mm diameter) based on a non-prewhitening matched filter model observer with eye filter. Results: Image noise magnitude decreased with decreasing phantom size, increasing dose, and increasing ADMIRE strength, offering up to 64% noise reduction relative to FBP. Noise texture in terms of the NPS was similar between FBP and ADMIRE (<5% shift in peak frequency). The resolution, based on the TTF, improved with increased ADMIRE strength by an average of 15% in the TTF 50% frequency for ADMIRE-5. The detectability index increased with increasing dose and ADMIRE strength by an average of 55%, 90%, and 163% for ADMIRE 3, 4, and 5, respectively. Assessing the impact of mA modulation for a fixed average dose over the length of the phantom, detectability was up to 49% lower in smaller phantom sections and up to 26% higher in larger phantom sections for the modulated scan compared to a fixed tube current scan. Overall, the detectability exhibited less variability with phantom size for modulated scans compared to fixed tube current scans. Conclusions: Image quality increased with increasing dose and decreasing phantom size. The CT system exhibited nonlinear noise and resolution properties, especially at very low-doses, large phantom sizes, and for low-contrast objects. Objective image quality metrics generally increased with increasing dose and ADMIRE strength, and with decreasing phantom size. The ADMIRE algorithm could offer comparable image quality at reduced doses or improved image quality at the same dose. The use of tube current modulation resulted in more consistent image quality with changing phantom size.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2015

TU-EF-204-11: Impact of Using Multi-Slice Training Sets On the Performance of a Channelized Hotelling Observer in a Low-Contrast Detection Task in CT

Purpose: To investigate using multiple CT image slices from a single acquisition as independent training images for a channelized Hotelling observer (CHO) model to reduce the number of repeated scans for CHO-based CT image quality assessment. Methods: We applied a previously validated CHO model to detect low contrast disk objects formed from cross-sectional images of three epoxy-resin-based rods (diameters: 3, 5, and 9 mm; length: ∼5cm). The rods were submerged in a 35x 25 cm2 iodine-doped water filled phantom, yielding-15 HU object contrast. The phantom was scanned 100 times with and without the rods present. Scan and reconstruction parameters include: 5 mm slice thickness at 0.5 mm intervals, 120 kV, 480 Quality Reference mAs, and a 128-slice scanner. The CHO’s detectability index was evaluated as a function of factors related to incorporating multi-slice image data: object misalignment along the z-axis, inter-slice pixel correlation, and number of unique slice locations. In each case, the CHO training set was fixed to 100 images. Results: Artificially shifting the object’s center position by as much as 3 pixels in any direction relative to the Gabor channel filters had insignificant impact on object detectability. An inter-slice pixel correlation of >∼0.2 yielded positive bias in the model’s performance. Incorporating multi-slice image data yielded slight negative bias in detectability with increasing number of slices, likely due to physical variations in the objects. However, inclusion of image data from up to 5 slice locations yielded detectability indices within measurement error of the single slice value. Conclusion: For the investigated model and task, incorporating image data from 5 different slice locations of at least 5 mm intervals into the CHO model yielded detectability indices within measurement error of the single slice value. Consequently, this methodology would Result in a 5-fold reduction in number of image acquisitions. This project was supported by National Institutes of Health grants R01 EB017095 and U01 EB017185 from the National Institute of Biomedical Imaging and Bioengineering.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2015

TU-G-BRD-08: In-Vivo EPID Dosimetry: Quantifying the Detectability of Four Classes of Errors

Purpose: EPID dosimetry is an emerging method for treatment verification and QA. Given that the in-vivo EPID technique is in clinical use at some centers, we investigate the sensitivity and specificity for detecting different classes of errors. We assess the impact of these errors using dose volume histogram endpoints. Though data exist for EPID dosimetry performed pre-treatment, this is the first study quantifying its effectiveness when used during patient treatment (in-vivo). Methods: We analyzed 17 patients; EPID images of the exit dose were acquired and used to reconstruct the planar dose at isocenter. This dose was compared to the TPS dose using a 3%/3mm gamma criteria. To simulate errors, modifications were made to treatment plans using four possible classes of error: 1) patient misalignment, 2) changes in patient body habitus, 3) machine output changes and 4) MLC misalignments. Each error was applied with varying magnitudes. To assess the detectability of the error, the area under a ROC curve (AUC) was analyzed. The AUC was compared to changes in D99 of the PTV introduced by the simulated error. Results: For systematic changes in the MLC leaves, changes in the machine output and patient habitus, the AUC varied from 0.78–0.97 scaling with the magnitude of the error. The optimal gamma threshold as determined by the ROC curve varied between 84–92%. There was little diagnostic power in detecting random MLC leaf errors and patient shifts (AUC 0.52–0.74). Some errors with weak detectability had large changes in D99. Conclusion: These data demonstrate the ability of EPID-based in-vivo dosimetry in detecting variations in patient habitus and errors related to machine parameters such as systematic MLC misalignments and machine output changes. There was no correlation found between the detectability of the error using the gamma pass rate, ROC analysis and the impact on the dose volume histogram. Funded by grant R18HS022244 from AHRQ.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2015

WE-G-204-07: Automated Characterization of Perceptual Quality of Clinical Chest Radiographs: Improvements in Lung, Spine, and Hardware Detection

Purpose: To develop and validate more robust methods for automated lung, spine, and hardware detection in AP/PA chest images. This work is part of a continuing effort to automatically characterize the perceptual image quality of clinical radiographs. [Y. Lin et al. Med. Phys. 39, 7019–7031 (2012)] Methods: Our previous implementation of lung/spine identification was applicable to only one vendor. A more generalized routine was devised based on three primary components: lung boundary detection, fuzzy c-means (FCM) clustering, and a clinically-derived lung pixel probability map. Boundary detection was used to constrain the lung segmentations. FCM clustering produced grayscale- and neighborhood-based pixel classification probabilities which are weighted by the clinically-derived probability maps to generate a final lung segmentation. Lung centerlines were set along the left-right lung midpoints. Spine centerlines were estimated as a weighted average of body contour, lateral lung contour, and intensity-based centerline estimates. Centerline estimation was tested on 900 clinical AP/PA chest radiographs which included inpatient/outpatient, upright/bedside, men/women, and adult/pediatric images from multiple imaging systems. Our previous implementation further did not account for the presence of medical hardware (pacemakers, wires, implants, staples, stents, etc.) potentially biasing image quality analysis. A hardware detection algorithm was developed using a gradient-based thresholding method. The training and testing paradigm used a set of 48 images from which 1920 51×51 pixel{sup 2} ROIs with and 1920 ROIs without hardware were manually selected. Results: Acceptable lung centerlines were generated in 98.7% of radiographs while spine centerlines were acceptable in 99.1% of radiographs. Following threshold optimization, the hardware detection software yielded average true positive and true negative rates of 92.7% and 96.9%, respectively. Conclusion: Updated segmentation and centerline estimation methods in addition to new gradient-based hardware detection software provide improved data integrity control and error-checking for automated clinical chest image quality characterization across multiple radiography systems.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2016

Mediastinal lymph node detection and station mapping on chest CT using spatial priors and random forest

Purpose: To develop an automated system for mediastinal lymph node detection and station mapping for chest CT. Methods: The contextual organs, trachea, lungs, and spine are first automatically identified to locate the region of interest (ROI) (mediastinum). The authors employ shape features derived from Hessian analysis, local object scale, and circular transformation that are computed per voxel in the ROI. Eight more anatomical structures are simultaneously segmented by multiatlas label fusion. Spatial priors are defined as the relative multidimensional distance vectors corresponding to each structure. Intensity, shape, and spatial prior features are integrated and parsed by a random forest classifier for lymph node detection. The detected candidates are then segmented by the following curve evolution process. Texture features are computed on the segmented lymph nodes and a support vector machine committee is used for final classification. For lymph node station labeling, based on the segmentation results of the above anatomical structures, the textual definitions of mediastinal lymph node map according to the International Association for the Study of Lung Cancer are converted into patient-specific color-coded CT image, where the lymph node station can be automatically assigned for each detected node. Results: The chest CT volumes from 70 patients with 316 enlarged mediastinal lymph nodes are used for validation. For lymph node detection, their system achieves 88% sensitivity at eight false positives per patient. For lymph node station labeling, 84.5% of lymph nodes are correctly assigned to their stations. Conclusions: Multiple-channel shape, intensity, and spatial prior features aggregated by a random forest classifier improve mediastinal lymph node detection on chest CT. Using the location information of segmented anatomic structures from the multiatlas formulation enables accurate identification of lymph node stations.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2016

SU-F-J-197: A Novel Intra-Beam Range Detection and Adaptation Strategy for Particle Therapy

Purpose: In-vivo range detection/verification is crucial in particle therapy for effective and safe delivery. The state-of-art techniques are not sufficient for in-vivo on-line range verification due to conflicts among patient dose, signal statistics and imaging time. We propose a novel intra-beam range detection and adaptation strategy for particle therapy. Methods: This strategy uses the planned mid-range spots as probing beams without adding extra radiation to patients. Such choice of probing beams ensures the Bragg peaks to remain inside the tumor even with significant range variation from the plan. It offers sufficient signal statistics for in-beam positron emission tomography (PET) due to high positron activity of therapeutic dose. The probing beam signal can be acquired and reconstructed using in-beam PET that allows for delineation of the Bragg peaks and detection of range shift with ease of detection enabled by single-layered spots. If the detected range shift is within a pre-defined tolerance, the remaining spots will be delivered as the original plan. Otherwise, a fast re-optimization using range-shifted beamlets and accounting for the probing beam dose is applied to consider the tradeoffs posed by the online anatomy. Simulated planning and delivery studies were used to demonstrate the effectiveness of the proposed techniques. Results: Simulations with online range variations due to shifts of various foreign objects into the beam path showed successful delineation of the Bragg peaks as a result of delivering probing beams. Without on-line delivery adaptation, dose distribution was significantly distorted. In contrast, delivery adaptation incorporating detected range shift recovered well the planned dose. Conclusion: The proposed intra-beam range detection and adaptation utilizing the planned mid-range spots as probing beams, which illuminate the beam range with strong and accurate PET signals, is a safe, practical, yet effective approach to address range uncertainty issues in particle therapy.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2016

MO-FG-CAMPUS-TeP1-03: Pre-Treatment Surface Imaging Based Collision Detection

Purpose: Modern radiotherapy increasingly employs large immobilization devices, gantry attachments, and couch rotations for treatments. All of which raise the risk of collisions between the patient and the gantry / couch. Collision detection is often achieved by manually checking each couch position in the treatment room and sometimes results in extraneous imaging if collisions are detected after image based setup has begun. In the interest of improving efficiency and avoiding extra imaging, we explore the use of a surface imaging based collision detection model. Methods: Surfaces acquired from AlignRT (VisionRT, London, UK) were transferred in wavefront format to a custom Matlab (Mathworks, Natick, MA) software package (CCHECK). Computed tomography (CT) scans acquired at the same time were sent to CCHECK in DICOM format. In CCHECK, binary maps of the surfaces were created and overlaid on the CT images based on the fixed relationship of the AlignRT and CT coordinate systems. Isocenters were added through a graphical user interface (GUI). CCHECK then compares the inputted surfaces to a model of the linear accelerator (linac) to check for collisions at defined gantry and couch positions. Note, CCHECK may be used with or without a CT. Results: The nominal surface image field of view is 650 mm × 900 mm, with variance based on patient position and size. The accuracy of collision detections is primarily based on the linac model and the surface mapping process. The current linac model and mapping process yield detection accuracies on the order of 5 mm, assuming no change in patient posture between surface acquisition and treatment. Conclusions: CCHECK provides a non-ionizing method to check for collisions without the patient in the treatment room. Collision detection accuracy may be improved with more robust linac modeling. Additional gantry attachments (e.g. conical collimators) can be easily added to the model.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2016

Compiler-Directed Soft Error Detection and Recovery to Avoid DUE and SDC via Tail-DMR

This article presents Clover, a compiler-directed soft error detection and recovery scheme for lightweight soft error resilience. The compiler carefully generates soft-error-tolerant code based on idempotent processing without explicit checkpoints. During program execution, Clover relies on a small number of acoustic wave detectors deployed in the processor to identify soft errors by sensing the wave made by a particle strike. To cope with DUEs (detected unrecoverable errors) caused by the sensing latency of error detection, Clover leverages a novel selective instruction duplication technique called tail-DMR (dual modular redundancy) that provides a region-level error containment. Once a soft error is detected by either the sensors or the tail-DMR, Clover takes care of the error as in the case of exception handling. To recover from the error, Clover simply redirects program control to the beginning of the code region where the error is detected. The experimental results demonstrate that the average runtime overhead is only 26%, which is a 75% reduction compared to that of the state-of-the-art soft error resilience technique. In addition, this article evaluates an alternative technique called tail-wait, comparing it to Clover. According to the evaluation with the different processor configurations and the various error detection latencies, Clover turns out to be a superior technique, achieving 1.06 to 3.49× speedup over the tail-wait.

Computer Science↗

DOE OSTI2019

Towards Misregistration-Tolerant Change Detection using Deep Learning Techniques with Object-Based Image Analysis

Co-registrating is a common pre-processing step for existing change detection algorithms, but registering bi-temporal images is nontrivial. The use of image patch as input for deep learning techniques provides a natural avenue to apply them in the OBIA framework, and have shown successful performance in the object-based land cover mapping and change detection applications. Even though attempts of applying deep learning techniques for change detection applications have been made with varying success, its application under OBIA framework for change detection have not been conducted and its tolerance for misregistration among temporal images are neither known. This study performed change detection under OBIA framework using deep learning techniques for the first time, and evaluated its performance regarding their tolerance of image misregistration on training and testing dataset. Our results demonstrate the proposed change detection scheme is surprisingly robust to image misregistration on the testing dataset, while classifiers trained with the training dataset containing image misregistration errors suffer from slight decrease of overall accuracy.

Liu, Tao↗

DOE OSTI2022

Toward the Detection of Polyglot Files

Standardized file types play a key role in the development and use of computer software. However, it is possible to confound standardized file type processing by creating a file that is valid in multiple file types. The resulting polyglot (many languages) file can confuse file type identification, allowing elements of the file to evade analysis. This is especially problematic for malware detection systems that rely on file type identification for feature extraction. Although work has been done to identify file types using more comprehensive methods than file signatures, accurate identification of polyglot files remains an open problem. Since malware detection systems routinely perform file type-specific feature extraction, polyglot files need to be filtered out prior to ingestion by these systems. Otherwise, malicious content could pass through undetected. To address the problem of polyglot detection we assembled a data set using the mitra tool. We then evaluated the performance of the most commonly used file identification tools, including file, polydet, binwalk, and TrID. Our analysis demonstrates that existing file type detection tools fail to provide reliable polyglot detection. We then evaluated the ability of a range of machine and deep learning models to detect polyglot files. The most performant models were MalConv2 and Catboost, which demonstrated the highest recall on our data set with 95.16% and 95.45%, respectively. These models outperformed existing methods and could be incorporated into a malware detector’s file processing pipeline to filter out potentially malicious polyglots before file type-dependent feature extraction takes place.

Koch, Luke↗

DOE OSTI2022

Toward the Detection of Polyglot Files

Koch, Lucas↗