Conventional Camera Research Articles

Depth error correction for plenoptic cameras based on an innovative AR system combining geometrical waveguide and micro-lens array

Plenoptic cameras have been increasingly applied in three-dimensional (3D) optical imaging and measurements. Micro-lens array is the key component to capture the images at different depths simultaneously to realize the 3D reconstruction of the scenery. However, only a part of the captured information will be used eventually when focusing on specific objects, which implies a waste of time and resources that are incompetent for fast-response tasks like gesture recognition. Furthermore, it is still quite challenging to accurately calibrate the depths under complex circumstances, even sometimes losing a part of the target information due to the low signal-to-noise ratio. In this paper, a new hybrid optical system is developed combining a geometrical waveguide and micro-lens array to project a virtual scale network for quantitative depth calibration and fast-tracking of targets. Such a methodology based on the augmented reality (AR) mechanism helps to rapidly scope the targeted objects/features without reconstructing the full range model, significantly saving the processing time for a low-latency response. By establishing a geometrical model to quantitatively correlate the images in the auxiliary coordinate system and the virtual scale coordinate system, the depth error caused by the plenoptic camera can be calibrated and corrected. The coefficient of determination R2is used to evaluate the depth accuracy of 3D images and acts as the threshold to control the depth correction iterations. The closer the value of R2 is to 1, the more accurate the depth information is. Experiments proved that even under complex backgrounds or insufficient light, in virtue of the virtual coordinate networks, backgrounds could be rapidly filtered, and targets were effectively identified with accurate depth information. The algorithm flow chart for the correction of depth error is given. In such a way, the system can achieve faster and more accurate capturing of 3D objects in a real-time manner than conventional plenoptic cameras.

Handheld NIR-to-NIR Platform for on-site evaluating protective neutralizing antibody against SARS-CoV-2 ancestral strain and Omicron variant after vaccination or infection

Lateral flow assays (LFAs) are promising points-of-care tests, playing a vital role in diseases screening, diagnosis and surveillance. However, development of portable, cheap, and smart LFAs platform for sensitive and accurate quantification of disease biomarkers in complex media is challenging. Here, a cheap handheld device was developed to realize on-site detection of disease biomarkers by Nd3+/Yb3+ co-doped near-infrared (NIR)-to-NIR downconversion nanoparticles (DCNPs) based LFA. Its sensitivity is at least 8-fold higher for detecting NIR light signal from Nd3+/Yb3+ co-doped nanoparticles than conventional expensive InGaAs camera based detection platform. Additionally, we enhance NIR quantum yield of Nd3+/Yb3+ co-doped nanoparticles up to 35.5% via simultaneous high dopant of sensitizer ions Nd3+ and emitter ions Yb3+. Combination of NIR-to-NIR handheld detection device and ultra-bright NIR emitting NaNbF4:Yb60%@NaLuF4 nanoparticle probe allows the detection sensitivity of SARS-CoV-2 ancestral strain and Omicron variants specific neutralizing antibodies LFA up to the level of commercial enzyme linked immunosorbent assay kit. Furthermore, by this robust method, enhanced neutralizing antibodies against SARS-CoV-2 ancestral strain and Omicron variants are observed in healthy participants with Ad5-nCoV booster on top of two doses of inactivated vaccine. This NIR-to-NIR handheld platform provides a promising strategy for on-site evaluating protective humoral immunity after SARS-CoV-2 vaccination or infection.

Precise Integration of Polymeric Sensing Functional Materials within 3D Printed Microfluidic Devices

This work presents a new architecture concept for microfluidic devices, which combines the conventional 3D printing fabrication process with the stable and precise integration of polymeric functional materials in small footprints within the microchannels in well-defined locations. The approach solves the assembly errors that normally occur during the integration of functional and/or sensing materials in hybrid microfluidic devices. The method was demonstrated by embedding four pH-sensitive ionogel microstructures along the main microfluidic channel of a complex 3D printed microfluidic device. The results showed that this microfluidic architecture, comprising the internal integration of sensing microstructures of diverse chemical compositions, highly enhanced the adhesion force between the microstructures and the 3D printed microfluidic device that contains them. In addition, the performance of this novel 3D printed pH sensor device was investigated using image analysis of the pH colour variations obtained from photos taken with a conventional camera. The device presented accurate and repetitive pH responses in the 2 to 12 pH range without showing any type of device deterioration or lack of performance over time.

Mathematical aspects of distortion calibration for digital cameras

Nowadays, most human processes are automated by means of computerization. This process has not spared the automotive industry. The latest developments in this field give promise that in the near future, cars will be completely autonomous. However, before that, there is an urgent need to address a number of issues, such as increasing the angle of view for greater coverage of the road with minimal space curvature. It is known that when the angle of view increases, so does the distortion (a mismatch in geometric similarity between an object and its image). This mismatch significantly reduces the accuracy of recognition algorithms. To solve this problem, this paper proposes a calibration method for cameras. The paper deals with mathematical aspects of distortion and the method of calibration of automobile chambers, and calibration errors. The method of static calibration without the use of points at infinity using static templates is proposed. A conventional camera has a small field of view (about 80 degrees). Such angle of view does not provide full scanning of the space; as a result, some of the necessary information remains outside the field of view, which is crucial for automotive industry because road signs, pedestrians, traffic lights might be missed. Therefore, in case of car cameras, the angle of view is increased by introducing distortion aberration. In this case, the field of view increases, yet the geometric similarity between the object and its image is affected. This, in turn, can affect the accuracy of computer vision algorithms. In order to keep the field of view angle large and the similarity between the subject and the object, it is necessary to calibrate the camera. In this paper, a calibration method for wide-angle car cameras is proposed and considered.

Multi-center study of inter-rater reproducibility, image quality, and diagnostic accuracy of CZT versus conventional SPECT myocardial perfusion imaging

BackgroundCadmium-zinc-telluride (CZT)-based detectors exhibit higher diagnostic sensitivity in myocardial perfusion imaging (MPI) than conventional Anger-MPI for detection of coronary artery disease (CAD); however, reduced specificity and diagnostic accuracy of CZT-MPI were observed. This study aims to compare these different camera systems and to examine the degree of inter-rater reproducibility among readers with varying experience in MPI. Methods83 patients who underwent double stress/rest examinations using both a CZT and conventional SPECT cameras within one visit were included. Anonymized and randomized MPI-images were distributed to 15 international readers using a standardized questionnaire. Subsequent coronary angiography findings of ten patients served as a reference for analysis of sensitivity and specificity. ResultsImage quality was significantly better in CZT-MPI with significantly lower breast attenuation (P < 0.05). CZT-MPI exhibited higher sensitivity than Anger-MPI (87.5% vs. 62.5%) and significantly reduced specificity (40% vs. 100%). Readers experienced with both camera systems had the highest inter-rater agreement indicating higher reproducibility (CZT 0.54 vs. conv. 0.49, P < 0.05). ConclusionsHigher diagnostic sensitivity of CZT-MPI offers advantages in detection of CAD yet potentially of at the cost of reduced specificity, therefore it requires special training and a differentiated evaluation approach, especially for non-experienced readers with such camera systems.

MHealth hyperspectral learning for instantaneous spatiospectral imaging of hemodynamics.

Hyperspectral imaging acquires data in both the spatial and frequency domains to offer abundant physical or biological information. However, conventional hyperspectral imaging has intrinsic limitations of bulky instruments, slow data acquisition rate, and spatiospectral trade-off. Here we introduce hyperspectral learning for snapshot hyperspectral imaging in which sampled hyperspectral data in a small subarea are incorporated into a learning algorithm to recover the hypercube. Hyperspectral learning exploits the idea that a photograph is more than merely a picture and contains detailed spectral information. A small sampling of hyperspectral data enables spectrally informed learning to recover a hypercube from a red-green-blue (RGB) image without complete hyperspectral measurements. Hyperspectral learning is capable of recovering full spectroscopic resolution in the hypercube, comparable to high spectral resolutions of scientific spectrometers. Hyperspectral learning also enables ultrafast dynamic imaging, leveraging ultraslow video recording in an off-the-shelf smartphone, given that a video comprises a time series of multiple RGB images. To demonstrate its versatility, an experimental model of vascular development is used to extract hemodynamic parameters via statistical and deep learning approaches. Subsequently, the hemodynamics of peripheral microcirculation is assessed at an ultrafast temporal resolution up to a millisecond, using a conventional smartphone camera. This spectrally informed learning method is analogous to compressed sensing; however, it further allows for reliable hypercube recovery and key feature extractions with a transparent learning algorithm. This learning-powered snapshot hyperspectral imaging method yields high spectral and temporal resolutions and eliminates the spatiospectral trade-off, offering simple hardware requirements and potential applications of various machine learning techniques.

Environmental DNA metabarcoding from flowers reveals arthropod pollinators, plant pests, parasites, and potential predator–prey interactions while revealing more arthropod diversity than camera traps

AbstractArthropods can strongly impact ecosystems through pollination, herbivory, predation, and parasitism. As such, characterizing arthropod biodiversity is vital to understanding ecosystem health, functions, and services. Emerging environmental DNA (eDNA) methods targeting trace arthropod eDNA left behind on flowers have the potential to track arthropod biodiversity and interactions. The goal of this study was to determine the extent to which eDNA metabarcoding can identify plant‐arthropod and arthropod‐arthropod interactions and assess eDNA metabarcoding compared to conventional sampling. We deployed camera traps to document arthropod activity on specific flowers, sampled eDNA from those same flowers, then performed a metabarcoding analysis that targets a partial fragment of the cytochrome c oxidase subunit I gene (COI) to determine all arthropod eDNA present. We found that our eDNA metabarcoding analysis detected small arthropod pollinators, plant pests, and parasites, and shed light on potential predator–prey interactions while detecting 55 species compared to just 21 species from conventional camera trapping. The camera trapping survey, however, detected larger, more conspicuous nectarivores more successfully. We also explored the ecology of residual arthropod eDNA, finding that rainfall had a significant negative effect on the ability to detect residual arthropod eDNA. Preliminary evidence also indicates flower species may impact the amount of arthropod eDNA that can be detected. We found that eDNA metabarcoding can provide clues to potential predator–prey interactions on flowers and highlights the potential insights that can be gained from future eDNA metabarcoding studies. We show that eDNA metabarcoding is a valuable tool for not only detecting pollinator communities but for revealing potential interactions among plants, pollinators, pests, parasites, and predators. Future research should focus on how to improve the detection of large pollinators/nectivores and studying the ecology of residual arthropod eDNA to further explore this method's utility.

Research on Omnidirectional Stereo Measurement Using Convex Mirrors and Vertical Disparity

We propose an omnidirectional measurement method without blind spots by using a convex mirror, which in principle does not cause chromatic aberration, and by using vertical disparity by installing cameras at the top and bottom of the image. In recent years, there has been significant research in the fields of autonomous cars and robots. In these fields, three-dimensional measurements of the surrounding environment have become indispensable. Depth sensing with cameras is one of the most important sensors for recognizing the surrounding environment. Previous studies have attempted to measure a wide range of areas using fisheye and full spherical panoramic cameras. However, these approaches have limitations such as blind spots and the need for multiple cameras to measure all directions. Therefore, this paper describes a stereo camera system that uses a device capable of taking an omnidirectional image with a single shot, enabling omnidirectional measurement with only two cameras. This achievement was challenging to attain with conventional stereo cameras. The results of experiments confirmed an improvement in accuracy of up to 37.4% compared to previous studies. In addition, the system succeeded in generating depth image that can recognize distances in all directions in a single frame, demonstrating the possibility of omnidirectional measurement with two cameras.

Single-shot wavelength-multiplexed phase microscopy under Gabor regime in a regular microscope embodiment

Phase imaging microscopy under Gabor regime has been recently reported as an extremely simple, low cost and compact way to update a standard bright-field microscope with coherent sensing capabilities. By inserting coherent illumination in the microscope embodiment and producing a small defocus distance of the sample at the input plane, the digital sensor records an in-line Gabor hologram of the target sample, which is then numerically post-processed to finally achieve the sample’s quantitative phase information. However, the retrieved phase distribution is affected by the two well-known drawbacks when dealing with Gabor’s regime, that is, coherent noise and twin image disturbances. Here, we present a single-shot technique based on wavelength multiplexing for mitigating these two effects. A multi-illumination laser source (including 3 diode lasers) illuminates the sample and a color digital sensor (conventional RGB color camera) is used to record the wavelength-multiplexed Gabor hologram in a single exposure. The technique is completed by presenting a novel algorithm based on a modified Gerchberg–Saxton kernel to finally retrieve an enhanced quantitative phase image of the sample, enhanced in terms of coherent noise removal and twin image minimization. Experimental validations are performed in a regular Olympus BX-60 upright microscope using a 20X 0.46NA objective lens and considering static (resolution test targets) and dynamic (living spermatozoa) phase samples.

Multi-Stage Network for Event-Based Video Deblurring with Residual Hint Attention

Video deblurring aims at removing the motion blur caused by the movement of objects or camera shake. Traditional video deblurring methods have mainly focused on frame-based deblurring, which takes only blurry frames as the input to produce sharp frames. However, frame-based deblurring has shown poor picture quality in challenging cases of video restoration where severely blurred frames are provided as the input. To overcome this issue, recent studies have begun to explore the event-based approach, which uses the event sequence captured by an event camera for motion deblurring. Event cameras have several advantages compared to conventional frame cameras. Among these advantages, event cameras have a low latency in imaging data acquisition (0.001 ms for event cameras vs. 10 ms for frame cameras). Hence, event data can be acquired at a high acquisition rate (up to one microsecond). This means that the event sequence contains more accurate motion information than video frames. Additionally, event data can be acquired with less motion blur. Due to these advantages, the use of event data is highly beneficial for achieving improvements in the quality of deblurred frames. Accordingly, the results of event-based video deblurring are superior to those of frame-based deblurring methods, even for severely blurred video frames. However, the direct use of event data can often generate visual artifacts in the final output frame (e.g., image noise and incorrect textures), because event data intrinsically contain insufficient textures and event noise. To tackle this issue in event-based deblurring, we propose a two-stage coarse-refinement network by adding a frame-based refinement stage that utilizes all the available frames with more abundant textures to further improve the picture quality of the first-stage coarse output. Specifically, a coarse intermediate frame is estimated by performing event-based video deblurring in the first-stage network. A residual hint attention (RHA) module is also proposed to extract useful attention information from the coarse output and all the available frames. This module connects the first and second stages and effectively guides the frame-based refinement of the coarse output. The final deblurred frame is then obtained by refining the coarse output using the residual hint attention and all the available frame information in the second-stage network. We validated the deblurring performance of the proposed network on the GoPro synthetic dataset (33 videos and 4702 frames) and the HQF real dataset (11 videos and 2212 frames). Compared to the state-of-the-art method (D2Net), we achieved a performance improvement of 1 dB in PSNR and 0.05 in SSIM on the GoPro dataset, and an improvement of 1.7 dB in PSNR and 0.03 in SSIM on the HQF dataset.

Multitemporal and Multiscale Applications of Geomatic Techniques to Medium-Sized Archaeological Sites—Case Study of Marroquíes Bajos (Jaén, Spain)

This study describes a methodology for obtaining a geometric documentation of a medium-sized archaeological area by applying various geomatic techniques. The procedure considers the obtainment of products at several scales, from the entire site to small artifacts, and at several dates, in order to model the evolution of the archaeological work. The methodology includes both LiDAR and photogrammetry, using the LiDAR point clouds to support the geometry obtained using photogrammetry and adding texture from this source. The technique used was adapted to the circumstances of the scene by considering the scale level (resolution and accuracy), complexity, and other requirements of the project. In the case of LiDAR, terrestrial laser scanning and structured-light scanning were used, and the aerial photogrammetry used two types of RPAS (medium and low flight height), close range photogrammetry with a conventional camera, and very close-range photogrammetry with a conventional camera mounted with a macro lens. The methodology demonstrated its feasibility for performing these types of studies, providing products adapted to the required scale level. All results were integrated into a website, including a map that allows user interaction and displays products at a selected zoom level, according to their scale level. The website also displays 3D models of the scenes and objects studied.

Imaging through Random Media Using Coherent Averaging (Laser Photonics Rev. 17(3)/2023)

Coherent Averaging In article number 2200673, Jung-Hoon Park and colleagues describe a new scheme to recover information hidden in randomly distorted images based on a simple coherent averaging algorithm. The method successfully recovers diffraction limited resolution images from images distorted by dynamic turbulence or multiple scattering. Since this work does not require any additional hardware and just utilizes images taken with conventional cameras or microscopes, it may open new avenues to realize high resolution imaging in challenging environments.

Predictive performance of random forest on the identification of mangrove species in arid environments

Machine learning methods are the most popular approaches for carrying out classification in remote sensing studies. Of the methods available, random forest (RF) is the one most often used due to its high predictive performance. The objective of this study was to assess the predictive performance of RF in identifying (classifying) mangrove species in an arid environment using two cameras: one conventional (visible part of the light, RGB), the other specialized (Green, Red, Near-infrared, GRN). The RGB and GRN bands were used with derived vegetation indexes (for each camera), the canopy height model (derived from photogrammetry), and distance to water (derived from raster analysis) to classify the study area in eight classes (including three mangrove species) using RF. Results suggest only slight differences in predictive performance (validation) between the products derived from the GRN and RGB cameras, the accuracy values ranged from 0.58 to 0.77 and from 0.53 to 0.72 for RGB and GRN, respectively. The most important variables were the distance to water and canopy height model for both cameras, followed by specific bands and vegetation indices. The study concludes that conventional cameras mounted in commercial drones can be used efficiently to identify mangrove species in arid environments when the classification model uses physical variables of the species (tree height) and the system (distance to water). Results of this study can be applied to describe spatial distributions by species in small or large patches of mangroves in arid environments, thus improving our ecological knowledge of this ecosystem.

Convolutional Neural Network-Based Soil Water Content and Density Prediction Model for Agricultural Land Using Soil Surface Images

For appropriate managing fields and crops, it is essential to understand soil properties. There are drawbacks to the conventional methods currently used for collecting a large amount of data from agricultural lands. Convolutional neural network is a deep learning algorithm that specializes in image classification, and developing soil property prediction techniques using this algorithm will be extremely beneficial to soil management. We present the convolution neural network models for estimating water content and dry density using soil surface images. Soil surface images were taken with a conventional digital camera. The range of water content and dry density were determined considering general upland soil conditions. Each image was divided into segmented images and used for model training and validation. The developed model confirmed that the model can learn soil features through appropriate image argumentation of few of original soil surface images. Additionally, it was possible to predict the soil water content in a situation where various soil dry density conditions were considered.

Reconfigurable and Polarization‐Dependent Grating Absorber for Large‐Area Emissivity Control Based on the Plasmonic Phase‐Change Material In3SbTe2

AbstractMetasurfaces with perfect infrared absorption promise integrated filters and compact detector elements with narrowband thermal emission. Phase‐change materials (PCMs) are prime candidates for active, non‐volatile absorption tuning. Commonly, the response of the entire metasurface is tuned, while local adaptions remain elusive. In this work, flexible encoding of different absorption/emission properties within a metasurface is shown. The plasmonic PCM In3SbTe2 (IST) is employed to obtain control over the emissivity by patterning an adaptable grating absorber metasurface. Using a commercial direct laser writing setup, the IST is locally switched from an amorphous dielectric into a crystalline metallic state, and cm‐sized stripe gratings are written above a reflecting mirror. Modification of already written patterns is demonstrated by changing the laser power and thus the IST stripe width to encode different polarization‐sensitive patterns with nearly perfect absorption into the same metasurface. Finally, an apparent local temperature pattern due to the large‐area emissivity shaping metasurface is measured with a conventional thermal camera. The results pave the way towards low‐cost, large‐area, and adaptable patterning of metasurfaces with wavelength and polarization‐selective perfect absorption, enabling applications like enhanced thermal detection, infrared camouflage, or encoding anti‐counterfeiting symbols.

A joint validation study on camera-aided illuminance measurement

This joint study conducted in the United States and Japan reviewed existing illuminance measurement methods that use a conventional digital single-lens reflex camera, including (a) per-pixel illuminance calculation using non-fisheye wide-angle lenses, (b) hemispherical projection-based illuminance calculation, (c) the KU method developed by the University of Kansas (KU) and (d) the L-CEPT method developed by the Japanese team for per-pixel illuminance calculation using 180° fisheye lens. Next, laboratory and field measurements were conducted in 32 test scenarios in Kansas and Japan to improve and validate the two similar methods (c) and (d) for per-pixel illuminance calculation and compare them to the simplified method (b) using orthographic projection-based illuminance calculation. Based on the test results, the camera-aided illuminance measurement method could have an error rate of 13.7% ± 15.3% (SD) in all 32 scenarios. Large errors (&gt;25%) were observed in six extremely dynamic gymnasiums with rapidly changing morning sunlight. Alternatively, an average error rate of 7.7% ± 7.5% was observed in the 26 test scenarios without rapidly changing direct sunlight. The simplified orthographic projection-based method could achieve the same level of accuracy as the per-pixel KU/L-CEPT method. Yet, multiple issues on camera-aided illuminance measurement still need further investigation.

Event-based feature tracking in a visual inertial odometry framework.

Introduction: Event cameras report pixel-wise brightness changes at high temporal resolutions, allowing for high speed tracking of features in visual inertial odometry (VIO) estimation, but require a paradigm shift, as common practices from the past decades using conventional cameras, such as feature detection and tracking, do not translate directly. One method for feature detection and tracking is the Eventbased Kanade-Lucas-Tomasi tracker (EKLT), an hybrid approach that combines frames with events to provide a high speed tracking of features. Despite the high temporal resolution of the events, the local nature of the registration of features imposes conservative limits to the camera motion speed. Methods: Our proposed approach expands on EKLT by relying on the concurrent use of the event-based feature tracker with a visual inertial odometry system performing pose estimation, leveraging frames, events and Inertial Measurement Unit (IMU) information to improve tracking. The problem of temporally combining high-rate IMU information with asynchronous event cameras is solved by means of an asynchronous probabilistic filter, in particular an Unscented Kalman Filter (UKF). The proposed method of feature tracking based on EKLT takes into account the state estimation of the pose estimator running in parallel and provides this information to the feature tracker, resulting in a synergy that can improve not only the feature tracking, but also the pose estimation. This approach can be seen as a feedback, where the state estimation of the filter is fed back into the tracker, which then produces visual information for the filter, creating a "closed loop". Results: The method is tested on rotational motions only, and comparisons between a conventional (not event-based) approach and the proposed approach are made, using synthetic and real datasets. Results support that the use of events for the task improve performance. Discussion: To the best of our knowledge, this is the first work proposing the fusion of visual with inertial information using events cameras by means of an UKF, as well as the use of EKLT in the context of pose estimation. Furthermore, our closed loop approach proved to be an improvement over the base EKLT, resulting in better feature tracking and pose estimation. The inertial information, despite prone to drifting over time, allows keeping track of the features that would otherwise be lost. Then, feature tracking synergically helps estimating and minimizing the drift.

CFSR: Coarse-to-Fine High-Speed Motion Scene Reconstruction with Region-Adaptive-Based Spike Distinction

As a novel bio-inspired vision sensor, spike cameras offer significant advantages over conventional cameras with a fixed low sampling rate, recording fast-moving scenes by firing a continuous stream of spikes. Reconstruction methods including Texture from ISI (TFI), Texture from Playback (TFP), and Texture from Adaptive threshold (TFA) produce undesirable noise or motion blur. A spiking neural model distinguishes the dynamic and static spikes before reconstruction, but the reconstruction of motion details is still unsatisfactory even with the advanced TFA method. To address this issue, we propose a coarse-to-fine high-speed motion scene reconstruction (CFSR) method with a region-adaptive-based spike distinction (RASE) framework to reconstruct the full texture of natural scenes from the spike data. We utilize the spike distribution of dynamic and static regions to propose the RASE to distinguish the spikes of different moments. After distinction, the TFI, TFP, and patch matching are exploited for image reconstruction in different regions, respectively, which does not introduce unexpected noise or motion blur. Experimental results on the PKU-SPIKE-RECON dataset demonstrate that our CFSR method outperforms the state-of-the-art approaches in terms of objective and subjective quality.

CamDec: Advancing Axis P1435-LE video camera security using honeypot-based deception

The explosion of online video streaming in recent years resulted in advanced services both in terms of efficiency and convenience. However, Internet-connected video cameras are prone to exploitation, leading to information security issues and data privacy concerns. The proliferation of video-capable Internet of Things devices and cloud-managed surveillance systems further extend these security issues and concerns. In this paper, a novel approach is proposed for video camera deception via honeypots, offering increased security measures compared to what is available on conventional Internet-enabled video cameras.

Neuromorphic vision based control for the precise positioning of robotic drilling systems

The manufacturing industry is currently witnessing a paradigm shift with the unprecedented adoption of industrial robots, and machine vision is a key perception technology that enables these robots to perform precise operations in unstructured environments. However, the sensitivity of conventional vision sensors to lighting conditions and high-speed motion sets a limitation on the reliability and work-rate of production lines. Neuromorphic vision is a recent technology with the potential to address the challenges of conventional vision with its high temporal resolution, low latency, and wide dynamic range. In this paper and for the first time, we propose a novel neuromorphic vision based controller for robotic machining applications to enable faster and more reliable operation, and present a complete robotic system capable of performing drilling tasks with sub-millimeter accuracy. Our proposed system localizes the target workpiece in 3D using two perception stages that we developed specifically for the asynchronous output of neuromorphic cameras. The first stage performs multi-view reconstruction for an initial estimate of the workpiece’s pose, and the second stage refines this estimate for a local region of the workpiece using circular hole detection. The robot then precisely positions the drilling end-effector and drills the target holes on the workpiece using a combined position-based and image-based visual servoing approach. The proposed solution is validated experimentally for drilling nutplate holes on workpieces placed arbitrarily in an unstructured environment with uncontrolled lighting. Experimental results prove the effectiveness of our solution with maximum positional errors of less than 0.2 mm, and demonstrate that the use of neuromorphic vision overcomes the lighting and speed limitations of conventional cameras. The findings of this paper identify neuromorphic vision as a promising technology that can expedite and robustify robotic manufacturing processes in line with the requirements of the fourth industrial revolution. • We present a neuromorphic visual controller approach for precise robotic machining. • We devise an event-based 3D reconstruction method for the localization of workpieces. • We develop an event-based method for the detection and tracking of circular objects. • We preform experiments that validate the proposed method in a robotic drilling setup. • The presented system can perform drilling tasks with sub-millimeter precision. • Results demonstrate the performance and robustness advantages of neuromorphic vision.