Demonstrating the Suitability of Neuromorphic, Event-Based, Dynamic Vision Sensors for In Process Monitoring of Metallic Additive Manufacturing and Welding

Alessandro Cattaneo; Amber Black; Andre Green; Ashlee Liao; David Mascare\~nas; Garrett Kenyon; John Bernardin; Michael Torrez

REVIEW 4 major objections 8 minor 21 references

Demonstrating the Suitability of Neuromorphic, Event-Based, Dynamic Vision Sensors for In Process Monitoring of Metallic Additive Manufacturing and Welding

T0 review · 4 major / 8 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Event-based, neuromorphic imagers can observe both gas-tungsten-arc and laser welding melt pools without saturating, at a data cost about 35 times smaller than a conventional imager.

desk verdict A useful existence proof, honestly framed, but the suitability claim needs a validation pass before the memory-savings and defect-prediction extrapolations are taken seriously. read the letter →

arxiv 2411.13108 v1 pith:JMMO7E3P submitted 2024-11-20 eess.IV cs.CV

David Mascareñas , Andre Green , Ashlee Liao , Michael Torrez , Alessandro Cattaneo , Amber Black , John Bernardin , Garrett Kenyon This is my paper · ORCID

classification eess.IVcs.CV

keywords additivemanufacturingdynamicvisionsensorevent-basedimagingin-processmonitoringneuromorphicweldingmeltpoolkeyholedigitalcodedexposure

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper sets out to establish that event-based dynamic vision sensors are suitable for in-process monitoring of metallic additive manufacturing and welding. It reports that such a sensor can observe gas-tungsten-arc and laser melt pools without pixel saturation when a welding shade is placed in the optical path, and that it captures melt-pool dynamics on timescales near 100 microseconds. The central quantitative result is a roughly 35-fold reduction in memory needed to store comparable observations relative to a conventional imager. If the claim holds, low-power, low-bandwidth neuromorphic imagers become a practical route to melt-pool monitoring, anomaly detection, and digital twins of metal builds.

What carries the argument

The central object is the event-based dynamic vision sensor (DVS), a neuromorphic imager whose pixels independently emit an event (timestamp, position, polarity) only when local log-light intensity crosses a threshold. This per-pixel change detection gives the sensor a nominal dynamic range of roughly 120 dB versus about 48 dB for a conventional 8-bit imager, and temporal resolution on the order of tens of microseconds. The argument runs on two mechanisms: the high dynamic range lets the sensor avoid saturation in the intense light of a melt pool (with a welding shade in the path), and the event-driven adaptive sampling means data volume tracks the dynamics, producing the measured roughly 35-fold memory saving compared with uniform frame capture. A third mechanism, digital coded exposure, shapes the temporal frequency content of frames formed by summing events over sliding windows.

What would settle it

A synchronized conventional high-speed camera recording the same arc and laser melt pools through the same optical path, compared frame-by-frame with event-based frames, would settle whether the event outlines match the melt-pool and keyhole geometry over time; a clear mismatch would undermine the physical basis of the suitability claim.

Watch

Extended reading notes

Core claim

The paper's central discovery is that event-based imagers, which report only pixel-level changes in log light intensity, have the dynamic range and speed to see melt pools from both electric arcs (GTAW/TIG) and lasers without saturation, provided a welding shade reduces the total light. From the event data the outline of the melt pool and the tungsten electrode are visible in the GTAW case, and the changing aspect ratio of the laser keyhole is visible over time. The paper also demonstrates that these imagers generate roughly 35 times less memory than a conventional imager for comparable observations of a laser welding melt pool, and that event data adaptively samples dynamics, as shown by a popping-balloon experiment. The team hypothesizes that bright moving anomalies in the GTAW melt pool correspond to high-emissivity contaminants, though this is explicitly left to future work.

Load-bearing premise

The claim collapses if the recorded events come from arc flicker, spatter, electromagnetic interference, or weld-shade artifacts instead of melt-pool radiance changes, or if event frames do not faithfully encode melt-pool geometry such as the keyhole outline.

Editorial extensions

If this is right

In-process monitoring of arc and laser melt pools becomes feasible with a sensor that does not saturate in bright melt-pool light, enabling weld-quality and additive-manufacturing process control.
The roughly 35-fold reduction in memory and bandwidth makes long build histories and digital twins of additively manufactured components more practical.
Because event rate naturally adapts to dynamics, quiescent periods produce almost no data while fast transients such as keyhole instabilities produce data bursts, easing storage and analysis.
Frames formed from event data can be shaped by digital coded exposure to control motion-blur-like temporal content, reducing the vulnerability of downstream classifiers to motion-blur adversarial effects.
The observed outline of the melt pool and the changing keyhole aspect ratio motivate pursuing quantitative 3D melt-pool geometry measurement and anomaly detection, classification, and prediction with additional engineering.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

If events faithfully encode melt-pool edges, then event-rate statistics alone could serve as a low-cost early-warning signal for keyhole collapse or spatter ejection, a testable extension the paper does not perform.
The memory-savings factor is likely understated when comparing against conventional imagery that stores high-dynamic-range intensity information, since event data cost is independent of dynamic range, a point the paper raises qualitatively.
The reliance on a welding shade suggests a natural next experiment: a short-wave-infrared event sensor, which the paper notes is becoming available, could observe melt pools without a shade and with less sensitivity to arc glare.
The popping-balloon demonstration implies that the same adaptive-sampling benefit transfers to other fast, transient manufacturing phenomena such as spatter, which could be identified and counted from event bursts alone.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Plainly: this is a short, honest feasibility study. What you should know is that the core observation—an event-based camera, placed behind a welding shade, can see something that looks like a GTAW melt pool and a laser keyhole without saturating—is real and illustrated with clear figures. That matters because melt-pool monitoring is a hard practical problem, and the dynamic-range argument for event sensors is plausible. The paper also does a nice job with the popping-balloon example, which makes the adaptive-sampling idea concrete, and it is transparent about being a derivative of the authors' IMAC 2024 paper.

The soft spots are proportional to the claim. The paper frames itself as demonstrating 'suitability' for in-process monitoring, but the evidence is interpretive. Figures 4 and 6 show event frames and assert that they encode the melt pool outline and keyhole aspect ratio. There is no synchronized ground-truth camera, no quantitative comparison of the extracted boundaries, and no check on whether arc flicker, spatter, or the weld shade's own contrast are contributing events. For the moving anomalies the authors explicitly say the hypothesis needs future work; the central outlines get no such caveat, and that asymmetry is the real weakness. The 35x memory-savings number is another soft spot: it is presented without a protocol for what 'comparable' means, what camera settings were used, and whether the comparison includes the event timestamp overhead. It reads as an estimate, not a measurement.

That said, I don't think the paper is circular or dishonest. The authors are careful in most places; the prior-work citations are legitimate, and the free parameter set is empty. The problem is simply that a feasibility demonstration should not push as hard as the conclusion does toward 3D geometry and defect prediction.

Who gets value: readers in welding diagnostics, event-based vision applied to industrial settings, and anyone who wants to see whether event cameras actually work in this environment. It is a useful existence proof, not a methods contribution. I'd accept it for peer review because the empirical observation deserves to be recorded and checked, but I'd ask the authors to add even a minimal validation—a side-by-side frame from a conventional camera, or a quantitative tracking result—and to scale the conclusions to what the data show.

Referee Report

4 major / 8 minor

Summary. This manuscript reports a feasibility demonstration of event-based dynamic vision sensors (DVS) for observing melt pools in gas tungsten arc welding (GTAW) and laser welding. The authors present qualitative event frames showing what they interpret as the melt-pool outline and keyhole geometry, a balloon-pop experiment to illustrate adaptive sampling, a memory-saving estimate of roughly 35x compared with a conventional imager, and a discussion of digital coded exposure for converting event streams to frames. The paper concludes that event-based imagers are suitable for in-process monitoring of welding and additive manufacturing, with future potential for 3D melt-pool geometry and anomaly prediction.

Significance. If the central observations are validated, the paper provides a useful feasibility result: event-based sensors' high dynamic range and microsecond-scale temporal resolution are well suited to extreme arc and laser light environments, and the per-pixel adaptive sampling idea is physically sound. The balloon demonstration is a nice qualitative illustration of adaptive sampling, and the authors are appropriately cautious about the moving anomalies in the melt pool. However, the quantitative memory-saving claim is currently not reproducible, the 120 dB-to-bytes conversion is erroneous, and the identification of melt-pool and keyhole boundaries rests entirely on visual interpretation without external ground truth. These deficiencies do not invalidate the qualitative feasibility, but they must be fixed before the paper can support its stated claims.

major comments (4)

[Sections 2.1, 2.2 and Figs. 4, 6] The central claim that event frames display the melt-pool outline and keyhole aspect-ratio changes is supported only by visual interpretation. No synchronized conventional camera or quantitative geometry comparison is reported, so the possibility that the event structure is driven by arc flicker, spatter, plasma oscillation, weld-shade attenuation gradients, or vibration is not excluded. Because the paper's feasibility conclusion rests on this identification, add a validation experiment (e.g., simultaneous filtered conventional imaging or a calibrated shadowgraph) and report a quantitative agreement metric, such as edge distance or an aspect-ratio time series with uncertainty.
[Section 3.1 and Fig. 9] The claim that event-based imagers 'require about 35 times less memory' is not reproducible from the information given. The manuscript does not specify the conventional imager's bit depth, frame rate, compression, or the event camera's event representation (timestamp width, address bits), nor does it report the protocol used to compute the factor 35 or trial-to-trial variability. In addition, the sentence '120 dB corresponds to 20 bytes' is dimensionally wrong: 120 dB corresponds to approximately 20 bits at 6 dB/bit, and the subsequent '2.5 to 3 times more' argument must be corrected and derived from the actual bit widths. Provide the full data budget and derivation for the memory comparison.
[Sections 3.1 and 3.2] The paper does not state which frame-formation method (naive event summation, digital coded exposure, or something else) and which accumulation window were used to produce the welding frames in Figures 4 and 6, nor how the cumulative event-ratio plot in Figure 8 is normalized and time-based. This matters because the visibility of the melt-pool boundary and keyhole shape could depend strongly on the integration window and event-polarity handling, and it prevents reproduction or assessment of motion blur. Specify the exact frame-formation parameters for all displayed frames and provide a quantitative comparison with the conventional high-speed camera in the balloon experiment.
[Section 3.2] The digital coded exposure concept is introduced but never applied to the welding data; it remains a proposal. The section is presented under 'Results' but contains no experimental results using this method. Either demonstrate digital coded exposure on the welding event streams or re-label the section as a methods/future-work discussion so that it is clear this technique was not used to generate Figures 4 and 6.

minor comments (8)

[Abstract and Introduction] The text contains 'Event-driven driven' (duplicate word) and 'In addition event based imagers' (missing comma); these should be corrected.
[Section 3.1] The phrase '120 dB corresponds to 20 bytes' should be '20 bits', and 'Contempory' is misspelled.
[Sections 2.1 and 2.2] The sensor models are given as 'DVS 240 C' and 'Davis 346'; state the exact manufacturer and model for these and for the DVXplorer used in Section 3.1, and note whether the same event camera was used in all welding experiments.
[Figures 4 and 6] These event frames need scale bars, axis labels, and a statement of the time window used to form each frame; without these, the reader cannot assess the spatial scale or temporal integration of the displayed melt-pool features.
[Section 3.2] The discussion of adversarial motion blur and digital coded exposure is interesting but is not demonstrated on the welding data; it should be clearly framed as proposed future work.
[Section numbering] The manuscript jumps from Section 3.2 to Section 5 (Conclusions); a Section 4 is missing, and the numbering should be corrected.
[References] Reference [6] appears to duplicate Reference [21] (both are IMAC 2024 papers on the same demonstration); the authors should consolidate or clarify the relationship.
[Figure 9] The axes in Figure 9 are unlabeled and the underlying parameters (bit widths, frame rates, time intervals) are not given; the caption should list all parameters used for the memory comparison.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central claims rest on direct event-camera recordings, not on fitted parameters or load-bearing self-citations.

full rationale

The paper is an empirical demonstration rather than a derivation. The claim that event-based imagers can observe arc and laser melt pools rests on the recorded event data shown in Figures 3-6; no equation is fitted to a subset of that data and then used to predict the same quantity. The 35x memory-saving figure is a reported ratio from the authors' own laser-welding observation, not a prediction, though the conventional-imager baseline is not fully specified. The self-citations ([6], [17], [18], [20], [21]) concern prior demonstrations, future compression approaches, and frame-formation methods; none is invoked as the sole justification for the central suitability claim, and none imports a uniqueness theorem or ansatz that forces the conclusions. The interpretive assumption that recorded events correspond to melt-pool/keyhole boundaries rather than arc flicker or spatter is an unvalidated physical assumption, but that is a correctness and validation risk, not a circularity: the conclusion does not reduce to the assumption by construction.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

No parameters are fitted to data and no new physical entities are postulated. The paper is an experimental demonstration, so the main unstated premises are the datasheet specifications, the physical interpretation of the events, and the fairness of the data-savings comparison. These are reasonable for a feasibility study but should be validated in follow-up work.

assumptions (3)

domain assumption Event camera datasheets are accurate: DVS 240 C, Davis 346, and DVXplorer offer roughly 100 microsecond temporal precision and approximately 120 dB dynamic range.
The suitability argument relies on sensor specifications from references 10 and 14; the paper does not independently verify these characteristics.
domain assumption Events recorded during welding are dominated by melt pool radiance changes, not by arc flicker, spatter, electromagnetic interference, or weld-shade artifacts.
Sections 2.1 and 2.2 interpret event frames as melt pool and keyhole geometry without ground-truth comparison. If this assumption fails, the central suitability claim weakens.
domain assumption The 35x memory-saving comparison between event-based and conventional imagers is fair and uses comparable settings.
Section 3.1 reports a factor of 35 but does not specify frame rates, bit depths, resolutions, or whether the conventional imager used identical conditions.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Demonstrating the Suitability of Neuromorphic, Event-Based, Dynamic Vision Sensors for In Process Monitoring of Metallic Additive Manufacturing and Welding." pith.science (2026). https://pith.science/paper/JMMO7E3P

@misc{pith2026241113108,
  author       = {Pith},
  title        = {Pith review of: Demonstrating the Suitability of Neuromorphic, Event-Based, Dynamic Vision Sensors for In Process Monitoring of Metallic Additive Manufacturing and Welding},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JMMO7E3P}},
  note         = {Machine review of arXiv:2411.13108}
}

read the original abstract

We demonstrate the suitability of high dynamic range, high-speed, neuromorphic event-based, dynamic vision sensors for metallic additive manufacturing and welding for in-process monitoring applications. In-process monitoring to enable quality control of mission critical components produced using metallic additive manufacturing is of high interest. However, the extreme light environment and high speed dynamics of metallic melt pools have made this a difficult environment in which to make measurements. Event-based sensing is an alternative measurement paradigm where data is only transmitted/recorded when a measured quantity exceeds a threshold resolution. The result is that event-based sensors consume less power and less memory/bandwidth, and they operate across a wide range of timescales and dynamic ranges. Event-driven driven imagers stand out from conventional imager technology in that they have a very high dynamic range of approximately 120 dB. Conventional 8 bit imagers only have a dynamic range of about 48 dB. This high dynamic range makes them a good candidate for monitoring manufacturing processes that feature high intensity light sources/generation such as metallic additive manufacturing and welding. In addition event based imagers are able to capture data at timescales on the order of 100 {\mu}s, which makes them attractive to capturing fast dynamics in a metallic melt pool. In this work we demonstrate that event-driven imagers have been shown to be able to observe tungsten inert gas (TIG) and laser welding melt pools. The results of this effort suggest that with additional engineering effort, neuromorphic event imagers should be capable of 3D geometry measurements of the melt pool, and anomaly detection/classification/prediction.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

21 extracted references · 21 canonical work pages

[1]

Charge Readout,

D. A. Rowlands, "Charge Readout," in Field Guide to Photographic Science , Bellingham, Washington, SPIE Press, 2020, p. 41

work page 2020
[2]

Full Well Capacity and Pixel SaturationEducational Notes,

Teledyne Princeton Instruments, "Full Well Capacity and Pixel SaturationEducational Notes," Teledyne Princeton Instruments, [Online]. Available: https://www.princetoninstruments.com/learn/camera-fundamentals/full-well- capacity-pixel-saturation. [Accessed 2 April 2024]

work page 2024
[3]

Visible Welding® WeldWatch® HD Weld Camera Kit,

Interest, "Visible Welding® WeldWatch® HD Weld Camera Kit," [Online]. Available: https://intertest.com/products/visible-welding%C2%AE-weldwatch%C2%AE-hd-weld-camera-kit-with-ultra- dynamic-range. [Accessed 12 Feb 2023]

work page 2023
[4]

Common Welding Methods and the Use of Laser Illumination in Welding Imaging,

Cavitar, "Common Welding Methods and the Use of Laser Illumination in Welding Imaging," [Online]. Available: https://www.cavitar.com/library/common-welding-methods-and-the-use-of-laser-illumination-in-welding- imaging/. [Accessed 12 Feb 2023]. 10 of 10

work page 2023
[5]

A 128× 128 120 dB 15 μs Latency Asynchronous Temporal Contrast Vision Sensor,

P. Lichtsteiner, C. Posch and T. Delbruck, "A 128× 128 120 dB 15 μs Latency Asynchronous Temporal Contrast Vision Sensor," IEEE Journal of Solid-State Circuits , vol. 43, no. 2, pp. 566-576, 2008

work page 2008
[6]

Demonstration of the Utility of Neuromorphic Event -Based Imagers for Optical Measurement of Melt Pools for Additive Manufacturing and Welding Diagnostics,

A. G. David Mascarenas, "Demonstration of the Utility of Neuromorphic Event -Based Imagers for Optical Measurement of Melt Pools for Additive Manufacturing and Welding Diagnostics," in International Modal Analysis Conference, Orlando, FL, 2024

work page 2024
[7]

Truchas,

Los Alamos National Laboratory, "Truchas," [Online]. Available: https://github.com/lanl/truchas. [Accessed 8 2 2023]

work page 2023
[8]

Truchas Reference Manual,

Computational Physics and Methods Group, "Truchas Reference Manual," Los Alamos National Laboratory, Los Alamos, NM, 2021

work page 2021

Show all 21 references

[9]

Truchas – a multi-physics tool for casting simulation,

D. Korzekwa, "Truchas – a multi-physics tool for casting simulation," Int. J. Cast Met. Res., vol. 22, p. 187–191, 2009

2009
[10]

Specifications - Current Models,

Inivation, "Specifications - Current Models," 2022. [Online]. Available: https://inivation.com/wp- content/uploads/2023/02/2022-09-iniVation-devices-Specifications.pdf. [Accessed 12 May 2023]

2022
[11]

Mead and M

C. Mead and M. Ismail, Analog VLSI and Neural Systems, New York, NY: The Springer International Series in Engineering and Computer Science, 1989

1989
[12]

Neuromorphic Electronic Systems,

C. Mead, "Neuromorphic Electronic Systems," Proceedings of the IEEE , vol. 78, no. 10, pp. 1629 - 1636, 1990

1990
[13]

Ultra-Low-Power Vision and Sensory Processing for Edge Devices,

D. Muir, "Ultra-Low-Power Vision and Sensory Processing for Edge Devices," SynSense, 2023

2023
[14]

Prophesee, Metavision for Machines,

"Prophesee, Metavision for Machines," [Online]. Available: https://www.prophesee.ai/. [Accessed 3 January 2024]

2024
[15]

Quantum Imaging, launches its next-generation short-wave infrared (SWIR) camera – MIRA 02Y-E,

Laser Focus World, "Quantum Imaging, launches its next-generation short-wave infrared (SWIR) camera – MIRA 02Y-E," European Defence Review, pp. https:// www.edrmagazine.eu/quantum-imaging-launches-its-next- generation-short-wave-infrared-swir-camera-mira-02y-e , 27 April 2023

2023
[16]

Integrated melt pool and microstructure control for Ti –6Al–4V thin wall additive manufacturing.,

J. Gockel, J. B. J. Fox and R. Hafley, "Integrated melt pool and microstructure control for Ti –6Al–4V thin wall additive manufacturing.," Materials Science and Technology, vol. 31, no. 8, pp. 912-916, 2015

2015
[17]

Efficient Full -Field Vibration Measurements and Operational Modal Analysis Using Neuromorphic Event -Based Imaging,

C. Dorn, S. Dasari, Y. Yang, C. Farrar, P . W. Garrett Kenyon and D. Mascarenas, "Efficient Full -Field Vibration Measurements and Operational Modal Analysis Using Neuromorphic Event -Based Imaging," Journal of Engineering Mechanics, vol. 144, no. 7, 2018

2018
[18]

Imager -Based Techniques for Analyzing Metallic Melt Pools for Additive Manufacturing,

C. Hayes, C. Schelle, G. Taylor, B. Martinez, G. Kenyon, T. Lienert, Y. Yang and D. Mascareñas, "Imager -Based Techniques for Analyzing Metallic Melt Pools for Additive Manufacturing," in Conference Proceedings for the Society for Experimental Mechanics- International Modal An...

2020
[19]

Watch out! Motion is Blurring the Vision of Your Deep Neural Networks,

Q. Guo, F. Juefei-Xu, X. Xie, L. Ma, J. Wang, B. Yu, W. Feng and Y. Liu, "Watch out! Motion is Blurring the Vision of Your Deep Neural Networks," in NeurIPS, 2020

2020
[20]

Digital coded exposure formation of frames from event -based imagery,

A. Gothard, D. Jones, A. Green, M. Torrez, A. Cattaneo and D. Mascareñas, "Digital coded exposure formation of frames from event -based imagery," Digital coded exposure formation of frames from event -based imagery, vol. 2, no. 014005, 2022

2022
[21]

Demonstration of Neuromorphic Event-Based Imagers for Optical Measurement of Melt Pools for Additive Manufacturing and Welding Diagnostics,

D. Mascarenas and A. Green, "Demonstration of Neuromorphic Event-Based Imagers for Optical Measurement of Melt Pools for Additive Manufacturing and Welding Diagnostics," in International Modal Analysis Conference , Orlando, FL, 2024

2024

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.

[1] [1]

Charge Readout,

D. A. Rowlands, "Charge Readout," in Field Guide to Photographic Science , Bellingham, Washington, SPIE Press, 2020, p. 41

work page 2020

[2] [2]

Full Well Capacity and Pixel SaturationEducational Notes,

Teledyne Princeton Instruments, "Full Well Capacity and Pixel SaturationEducational Notes," Teledyne Princeton Instruments, [Online]. Available: https://www.princetoninstruments.com/learn/camera-fundamentals/full-well- capacity-pixel-saturation. [Accessed 2 April 2024]

work page 2024

[3] [3]

Visible Welding® WeldWatch® HD Weld Camera Kit,

Interest, "Visible Welding® WeldWatch® HD Weld Camera Kit," [Online]. Available: https://intertest.com/products/visible-welding%C2%AE-weldwatch%C2%AE-hd-weld-camera-kit-with-ultra- dynamic-range. [Accessed 12 Feb 2023]

work page 2023

[4] [4]

Common Welding Methods and the Use of Laser Illumination in Welding Imaging,

Cavitar, "Common Welding Methods and the Use of Laser Illumination in Welding Imaging," [Online]. Available: https://www.cavitar.com/library/common-welding-methods-and-the-use-of-laser-illumination-in-welding- imaging/. [Accessed 12 Feb 2023]. 10 of 10

work page 2023

[5] [5]

A 128× 128 120 dB 15 μs Latency Asynchronous Temporal Contrast Vision Sensor,

P. Lichtsteiner, C. Posch and T. Delbruck, "A 128× 128 120 dB 15 μs Latency Asynchronous Temporal Contrast Vision Sensor," IEEE Journal of Solid-State Circuits , vol. 43, no. 2, pp. 566-576, 2008

work page 2008

[6] [6]

Demonstration of the Utility of Neuromorphic Event -Based Imagers for Optical Measurement of Melt Pools for Additive Manufacturing and Welding Diagnostics,

A. G. David Mascarenas, "Demonstration of the Utility of Neuromorphic Event -Based Imagers for Optical Measurement of Melt Pools for Additive Manufacturing and Welding Diagnostics," in International Modal Analysis Conference, Orlando, FL, 2024

work page 2024

[7] [7]

Truchas,

Los Alamos National Laboratory, "Truchas," [Online]. Available: https://github.com/lanl/truchas. [Accessed 8 2 2023]

work page 2023

[8] [8]

Truchas Reference Manual,

Computational Physics and Methods Group, "Truchas Reference Manual," Los Alamos National Laboratory, Los Alamos, NM, 2021

work page 2021