Reduce Motion Blur Research Articles

Cardiac motion correction with a deep learning network for perfusion defect assessment in SPECT myocardial perfusion imaging

BackgroundIn myocardial perfusion imaging (MPI) with SPECT ungated studies are used for evaluation of perfusion defects despite motion blur. We investigate the potential benefit of motion correction using a deep-learning (DL) network for evaluating perfusion defects. MethodsWe employed a DL network for cardiac motion correction in ECG-gated SPECT-MPI images, wherein the image data from different cardiac phases are combined with respect to a reference gate to reduce motion blur. For training the DL network 197 cases were used. Given the variability of gated images during the cardiac cycle, we investigated the detectability of perfusion defects in two distinct reference gates. To assess perfusion defect detection, we performed receiver-operating-characteristic (ROC) analyses on the motion-corrected images using a separate test dataset of clinical 194 subjects, in which studies were created from actual patient data with inserted simulated-lesions as ground truth. The reconstructed images were assessed by the quantitative-perfusion SPECT (QPS) software. We also evaluated the performance on reduced-count studies (by two and four folds). ResultsThe quantitative results, measured by area-under-the-ROC curve (AUC), demonstrated that DL motion correction improves the detectability of perfusion defects significantly on both standard- and reduced-count studies, and that the detectability can vary with reference cardiac phases. A joint assessment from two reference-phases achieved AUC=0.841 on the quarter-count data, higher than with ungated full-count data (AUC=0.795, p-value=0.0054). ConclusionsDL motion correction can benefit assessment of perfusion defects in standard- and reduced-count SPECT-MPI studies. It can also be beneficial to evaluate perfusion images over multiple cardiac phases.

Evaluating Factors Shaping Real-Time Internet-of-Things-Based License Plate Recognition Using Single-Board Computer Technology

Reliable and cost-efficient license plate recognition (LPR) systems enhance security, traffic management, and automated toll collection in real-world applications. This study addresses optimal unique configurations for enhancing LPR system accuracy and reliability by evaluating the impact of camera angle, object velocity, and distance on the efficacy of real-time LPR systems. The Internet of Things (IoT) LPR framework is proposed and utilized on single-board computer (SBC) technology, such as the Raspberry Pi 4 platform, with a high-resolution webcam using advanced OpenCV and OCR–Tesseract algorithms applied. The research endeavors to simulate common deployment scenarios of the real-time LPR system and perform thorough testing by leveraging SBC computational capabilities and the webcam’s imaging capabilities. The testing process is not just comprehensive, but also meticulous, ensuring the system’s reliability in various operational settings. We performed extensive experiments with a hundred repetitions at diverse angles, velocities, and distances. An assessment of the data’s precision, recall, and F1 score indicates the accuracy with which Thai license plates are identified. The results show that camera angles close to 180° significantly reduce perspective distortion, thus enhancing precision. Lower vehicle speeds (<10 km/h) and shorter distances (<10 m) also improve recognition accuracy by reducing motion blur and improving image clarity. Images captured from shorter distances (approximately less than 10 m) are more accurate for high-resolution character recognition. This study substantially contributes to SBC technology utilizing IoT-based real-time LPR systems for practical, accurate, and cost-effective implementations.

Detecting people in sprinting motion using HPRDenoise: Point cloud denoising with hidden point removal

<p>LiDARs are utilized in various applications, such as self-driving vehicles and robotics, to aid in sensing the environment. However, LiDARs do not provide instantaneous images and they generate noise, adding to measurement errors. This noise, often referred to as motion blur phenomenon also observed in other imaging sensors results in decreased sensing accuracy for moving objects. This study introduces HPRDenoise, a noise reduction method based on hidden point removal, specifically designed to reduce motion blur during sprinting motion. This method capitalizes on the occlusion produced by a fixed-position LiDAR. We propose a comprehensive denoising approach to filter points from a point cloud without resorting to supervised learning, unlike most existing denoising algorithms. The number of correct frames and accuracy were compared for Raw, ScoreDenoise, which is the state-of-the-art method for random point cloud denoising, and HPRDenoise (Ours). Accuracy is defined as the ratio of the number of correct frames to the total number of frames. Experimental results demonstrate that the detection accuracy of point clouds processed with HPRDenoise is 72.73%, achieving better accuracy than those using conventional methods.</p>

Active Object Detection and Tracking Using Gimbal Mechanisms for Autonomous Drone Applications

Object recognition, localization, and tracking play a role of primordial importance in computer vision applications. However, it is still an extremely difficult task, particularly in scenarios where objects are attended to using fast-moving UAVs that need to robustly operate in real time. Typically the performance of these vision-based systems is affected by motion blur and geometric distortions, to name but two issues. Gimbal systems are thus essential to compensate for motion blur and ensure visual streams are stable. In this work, we investigate the advantages of active tracking approaches using a three-degrees-of-freedom (DoF) gimbal system mounted on UAVs. A method that utilizes joint movement and visual information for actively tracking spherical and planar objects in real time is proposed. Tracking methodologies are tested and evaluated in two different realistic Gazebo simulation environments: the first on 3D positional tracking (sphere) and the second on tracking of 6D poses (planar fiducial markers). We show that active object tracking is advantageous for UAV applications, first, by reducing motion blur, caused by fast camera motion and vibrations, and, second, by fixating the object of interest within the center of the field of view and thus reducing re-projection errors due to peripheral distortion. The results demonstrate significant object pose estimation accuracy improvements of active approaches when compared with traditional passive ones. More specifically, a set of experiments suggests that active gimbal tracking can increase the spatial estimation accuracy of known-size moving objects, under conditions of challenging motion patterns and in the presence of image distortion.

High-accuracy image-free classification of high-speed rotating objects with fluctuating rotation periods

Image-free classification methods with single-pixel measuring and deep learning show a capacity for long-duration classification of moving objects. However, motion blur restricts the allowable object motion speed of existing image-free classification methods. Aimed at high-speed rotating objects, we propose an image-free classification approach based on single-pixel measuring at the same spatial position of different rotation cycles to reduce motion blur. The proposed approach allows classifying rotating objects with fluctuating rotation periods to better meet the actual application conditions. We verify the proposed method by classifying the digits placed on a rotating disk. In our experiments, when digits rotate at around 960.9 revolutions per minute, corresponding to 10.06 m/s, the classification accuracy reaches 95.9%. In theory, the allowable speed is determined only by the sampling rate of the single-pixel measurements, which can allow for higher speeds than experimentally achieved. The proposed image-free classification method provides a promising way of monitoring high-speed rotating objects in the engineering field.

A 3.0 µm Pixels and 1.5 µm Pixels Combined Complementary Metal-Oxide Semiconductor Image Sensor for High Dynamic Range Vision beyond 106 dB.

We propose a new concept image sensor suitable for viewing and sensing applications. This is a report of a CMOS image sensor with a pixel architecture consisting of a 1.5 μm pixel with four-floating-diffusions-shared pixel structures and a 3.0 μm pixel with an in-pixel capacitor. These pixels are four small quadrate pixels and one big square pixel, also called quadrate-square pixels. They are arranged in a staggered pitch array. The 1.5 μm pixel pitch allows for a resolution high enough to recognize distant road signs. The 3 μm pixel with intra-pixel capacitance provides two types of signal outputs: a low-noise signal with high conversion efficiency and a highly saturated signal output, resulting in a high dynamic range (HDR). Two types of signals with long exposure times are read out from the vertical pixel, and four types of signals are read out from the horizontal pixel. In addition, two signals with short exposure times are read out again from the square pixel. A total of eight different signals are read out. This allows two rows to be read out simultaneously while reducing motion blur. This architecture achieves both an HDR of 106 dB and LED flicker mitigation (LFM), as well as being motion-artifact-free and motion-blur-less. As a result, moving subjects can be accurately recognized and detected with good color reproducibility in any lighting environment. This allows a single sensor to deliver the performance required for viewing and sensing applications.

63‐4: A New Single X‐Wire Active Matrix mini‐LED backlight with Motion Blur Reduction for LCD TVs

We have developed a high‐image‐quality 85‐inch 4K HDR LCD display with 2,304 dimming zones using single X‐wire AM mini‐LED backlight unit with motion blur and residual image reduction technology. Our latest work includes the proposed new single X‐wire AM driving method, consists of one matrix controller (DCON) and X‐wires of mini‐LED driver ICs (dimmers), which eliminates extra scanning, sensing lines and pixel/source driver IC using adjustable addressing dimming method, the flexible layout makes the display size easier to expand with low cost, and the H‐scan driving method which is designed for reducing motion blur and residual image using a scanning black frame insertion technology to improve display quality. The proposed mini‐LED BLU can save about 9% power consumption compared to conventional one in normal display.

A Novel Single X-Wire AM Mini-LED Backlight Unit with Motion Blur Reduction for High-Image-Quality LCD TVs

We have demonstrated a high-image-quality 85-inch 4 K high-dynamic-range (HDR) LCD display with 2304 dimming zones using a single X-wire AM mini-LED backlight unit (BLU) with motion blur and residual image reduction technology. Our latest work proposed the new single X-wire AM driving method, consisting of one matrix control IC/dimming controller (DCON) and X-wires of mini-LED driver ICs (dimmers), which eliminates extra scanning, sensing lines and pixel/source driver IC using an adaptive addressing dimming method. Additionally, a single X-wire controlled system can realize a flexible and expandable layout, which can achieve ultra thin and low cost in ultra-high dimming zones and large size display. Furthermore, the horizontal-scan (H-scan) driving method, which reduces motion blur and residual images using a scanning black frame insertion (BFI) technology, is utilized. The proposed mini-LED BLU can save about 9% power consumption compared to the conventional one in normal displays.

Small-amplitude head oscillations result from a multimodal head stabilization reflex in hawkmoths.

In flying insects, head stabilization is an essential reflex that helps to reduce motion blur during fast aerial manoeuvres. This reflex is multimodal and requires the integration of visual and antennal mechanosensory feedback in hawkmoths, each operating as a negative-feedback-control loop. As in any negative-feedback system, the head stabilization system possesses inherent oscillatory dynamics that depend on the rate at which the sensorimotor components of the reflex operate. Consistent with this expectation, weobserved small-amplitude oscillations in the head motion (or head wobble) of the oleander hawkmoth, Daphnis nerii, which are accentuated when sensory feedback is aberrant. Here, we show that these oscillations emerge from the inherent dynamics of the multimodal reflex underlying gaze stabilization, and that the amplitude of head wobble is a function of both the visual feedback and antennal mechanosensory feedback from the Johnston's organs. Our data support the hypothesis that head wobble results from a multimodal, dynamically stabilized reflex loop that mediates head positioning.

Impact of walking speed and motion adaptation on optokinetic nystagmus-like head movements in the blowfly Calliphora

The optokinetic nystagmus is a gaze-stabilizing mechanism reducing motion blur by rapid eye rotations against the direction of visual motion, followed by slower syndirectional eye movements minimizing retinal slip speed. Flies control their gaze through head turns controlled by neck motor neurons receiving input directly, or via descending neurons, from well-characterized directional-selective interneurons sensitive to visual wide-field motion. Locomotion increases the gain and speed sensitivity of these interneurons, while visual motion adaptation in walking animals has the opposite effects. To find out whether flies perform an optokinetic nystagmus, and how it may be affected by locomotion and visual motion adaptation, we recorded head movements of blowflies on a trackball stimulated by progressive and rotational visual motion. Flies flexibly responded to rotational stimuli with optokinetic nystagmus-like head movements, independent of their locomotor state. The temporal frequency tuning of these movements, though matching that of the upstream directional-selective interneurons, was only mildly modulated by walking speed or visual motion adaptation. Our results suggest flies flexibly control their gaze to compensate for rotational wide-field motion by a mechanism similar to an optokinetic nystagmus. Surprisingly, the mechanism is less state-dependent than the response properties of directional-selective interneurons providing input to the neck motor system.

Integration of visual and antennal mechanosensory feedback during head stabilization in hawkmoths.

During flight maneuvers, insects exhibit compensatory head movements which are essential for stabilizing the visual field on their retina, reducing motion blur, and supporting visual self-motion estimation. In Diptera, such head movements are mediated via visual feedback from their compound eyes that detect retinal slip, as well as rapid mechanosensory feedback from their halteres - the modified hindwings that sense the angular rates of body rotations. Because non-Dipteran insects lack halteres, it is not known if mechanosensory feedback about body rotations plays any role in their head stabilization response. Diverse non-Dipteran insects are known to rely on visual and antennal mechanosensory feedback for flight control. In hawkmoths, for instance, reduction of antennal mechanosensory feedback severely compromises their ability to control flight. Similarly, when the head movements of freely flying moths are restricted, their flight ability is also severely impaired. The role of compensatory head movements as well as multimodal feedback in insect flight raises an interesting question: in insects that lack halteres, what sensory cues are required for head stabilization? Here, we show that in the nocturnal hawkmoth Daphnis nerii, compensatory head movements are mediated by combined visual and antennal mechanosensory feedback. We subjected tethered moths to open-loop body roll rotations under different lighting conditions, and measured their ability to maintain head angle in the presence or absence of antennal mechanosensory feedback. Our study suggests that head stabilization in moths is mediated primarily by visual feedback during roll movements at lower frequencies, whereas antennal mechanosensory feedback is required when roll occurs at higher frequency. These findings are consistent with the hypothesis that control of head angle results from a multimodal feedback loop that integrates both visual and antennal mechanosensory feedback, albeit at different latencies. At adequate light levels, visual feedback is sufficient for head stabilization primarily at low frequencies of body roll. However, under dark conditions, antennal mechanosensory feedback is essential for the control of head movements at high frequencies of body roll.

Comparison of super-resolution and noise reduction for passive single-photon imaging

Single-photon sensitive image sensors have recently gained popularity in passive imaging applications where the goal is to capture photon flux (brightness) values of different scene points in the presence of challenging lighting conditions and scene motion. Recent work has shown that high-speed bursts of single-photon timestamp information captured using a single-photon avalanche diode camera can be used to estimate and correct for scene motion thereby improving signal-to-noise ratio and reducing motion blur artifacts. We perform a comparison of various design choices in the processing pipeline used for noise reduction, motion compensation, and upsampling of single-photon timestamp frames. We consider various pixelwise noise reduction techniques in combination with state-of-the-art deep neural network upscaling algorithms to super-resolve intensity images formed with single-photon timestamp data. We explore the trade space of motion blur and signal noise in various scenes with different motion content. Using real data captured with a hardware prototype, we achieved super-resolution reconstruction at frame rates up to 65.8 kHz (native sampling rate of the sensor) and captured videos of fast-moving objects. The best reconstruction is obtained with the motion compensation approach, which achieves a structural similarity (SSIM) of about 0.67 for fast-moving rigid objects. We are able to reconstruct subpixel resolution. These results show the relative superiority of our motion compensation compared to other approaches that do not exceed an SSIM of 0.5.

Anti-motion blur single-pixel imaging with calibrated radon spectrum.

Single-pixel imaging (SPI), a novel computational imaging technique that has emerged in the past decades, can effectively capture the image of a static object by consecutively measuring light intensities from it. However, when SPI is applied to imaging the dynamic object, severe motion blur in the restored image tends to appear. In this Letter, a new SPI scheme is proposed to largely alleviate such a problem by leveraging a calibrated radon spectrum. Such a spectrum is obtained by translating the acquired one-dimensional projection functions (1DPFs) according to the positional relationship among the 1DPFs. Simulation and experimental results demonstrate that, without prior knowledge, our approach can effectively reduce motion blur and restore high-quality images of the fast-moving object. In addition, the proposed scheme can also be used for fast object tracking.

GOOD VIBRATIONS? HOW IMAGE STABILISATION INFLUENCES PHOTOGRAMMETRY

Abstract. Image stabilisation (IS) is a family of approaches whose aim is to reduce motion blur in still images and shaking effect in video frames. A variety of techniques are currently implemented in cameras and camcorders: some involve hardware solutions, other are software approaches. In general, IS for still photography entails hardware in-camera or in-lens solutions. Video stabilisation, on the other hand, can be accomplished with software algorithms, either in real-time within the camera or in post-processing.Whereas IS aids photography and video making, its influence on the photogrammetric 3D modelling process has not been investigated. This article addresses this aspect. To this purpose, several laboratory and real-world tests were carried out, whose results showed that IS must be disabled when accuracy matters in photogrammetric projects. Details are provided in the manuscript.

Improved Coding Landmark-Based Visual Sensor Position Measurement and Planning Strategy for Multiwarehouse Automated Guided Vehicle

The application of vision-guided multiwarehouse automated guided vehicles (AGVs) is becoming more extensive. This article mainly studies the position measurement and corresponding path planning of AGV guided by visual sensors. An improved vehicle-mounted visual sensor model is proposed, which effectively reduces the cost of the sensor and achieves position prediction. The proposed manual coding landmark significantly improves the decoding speed and accuracy while ensuring the information coding capacity. An improved heuristic path planning algorithm combined with the visual sensor system is implemented, considering the planning effectiveness and directionality. Finally, simulations and experiments verify the reliability of the visual guidance method in position measurement and the advancement of the planning strategy. Compared with the traditional 2-D code and bottom visual model, the improved visual sensor position measurement effectively reduces motion blur and corresponding pose deviation. The decoding speed and accuracy are improved by more than 41.21% and 53.85%, respectively. Planning performance comparison is achieved between the same type of heuristic algorithms (A*, D*, and algorithm deployment). The significant improvement in core system indicators includes task efficiency (up to 70.02% improvement), parking time (up to 72% reduction), deadlock (up to 85.05% reduction), and planning time (up to 89.39% reduction).

Motion blur treatment utilizing deep learning for time-resolved particle image velocimetry

A new method is hereby presented to reduce motion blur induced error of time-resolved particle image velocimetry. The Monte-Carlo method (MCM) was applied to synthetic images to quantify the error due to blurred particle images. As the size of the streaks grew, it caused large errors in estimating displacements and increased the frequency of outliers beyond 20% for some cases. The mean displacement error was also about 0.2 – 0.55 px, which is larger than the nominally accepted PIV uncertainty of 0.1 px. A novel deblur filter (i.e., the generator) using a generative adversarial network (GAN) was developed, using 1 million synthetic images. The generator was verified using unlearned data from the MCM. The frequency of outliers, which was originally higher than 20% for the worst case, decreased to about 6%, and the displacement error was reduced to less than 0.3 px. The generator was applied to actual experimental images of a synthetic jet that had image blur and resulted in a substantial reduction of outliers. We also checked the performance of the generator in a uniform channel flow, and found that the deblurred images resulted in less PIV velocity error, and was closer to the results from the sharp images than those from the blurry images.Graphic abstract

Review of Methods for Animal Videography Using Camera Systems that Automatically Move to Follow the Animal.

Digital photography and videography provide rich data for the study of animal behavior and are consequently widely used techniques. For fixed, unmoving cameras there is a resolution versus field-of-view tradeoff and motion blur smears the subject on the sensor during exposure. While these fundamental tradeoffs with stationary cameras can be sidestepped by employing multiple cameras and providing additional illumination, this may not always be desirable. An alternative that overcomes these issues of stationary cameras is to direct a high-magnification camera at an animal continually as it moves. Here, we review systems in which automatic tracking is used to maintain an animal in the working volume of a moving optical path. Such methods provide an opportunity to escape the tradeoff between resolution and field of view and also to reduce motion blur while still enabling automated image acquisition. We argue that further development will be useful and outline potential innovations that may improve the technology and lead to more widespread use.

Infrared Temperature Measurements of the Blade Tip for a Turbine Operating at Corrected Engine Conditions

AbstractA high-speed infrared camera is used to measure the temperature of blade tips in a cooled high-pressure turbine operating at corrected engine conditions in The Ohio State University short duration Turbine Test Facility. These experiments create a challenging problem for infrared imaging since the rotor turns at over 13,000 rpm with tip speeds on the order of 300 m/s, and the surface temperature of the airfoils is on the order of 350 K. This means that the camera needs to capture a low-intensity signal in a very short time period. This article reviews the design and operation of a measurement procedure to accomplish this difficult task along with the postprocessing steps necessary to extract useful data. Raw infrared images are processed by deblurring the images using a nonblind Wiener filter and mapping the two-dimensional data onto the three-dimensional blade. This article also describes experiments covering a range of cooling flowrates and main flow temperatures. In addition, several tests with no main flow and only cooling flow were performed at lower speeds to reduce motion blur and enable the separation of internal and external heat transfer information. Results show that the infrared data are consistent and can provide quantitative comparisons of cooling performance even at the high rotation speed. This article presents the lessons learned for high-speed infrared measurement along with the representative data to illustrate the repeatability and capability of the measurement scheme as well as suggested improvements to guide further development.

Deep Learning-Based Denoising in High-Speed Portable Reflectance Confocal Microscopy.

Portable confocal microscopy (PCM) is a low-cost reflectance confocal microscopy technique that can visualize cellular details of human skin in vivo. When PCM images are acquired with a short exposure time to reduce motion blur and enable real-time 3D imaging, the signal-to-noise ratio (SNR) is decreased significantly, which poses challenges in reliably analyzing cellular features. In this paper, we evaluated deep learning (DL)-based approach for reducing noise in PCM images acquired with a short exposure time. Content-aware image restoration (CARE) network was trained with pairs of low-SNR input and high-SNR ground truth PCM images obtained from 309 distinctive regions of interest (ROIs). Low-SNR input images were acquired from human skin in vivo at the imaging speed of 180 frames/second. The high-SNR ground truth images were generated by registering 30 low-SNR input images obtained from the same ROI and summing them. The CARE network was trained using the Google Colaboratory Pro platform. The denoising performance of the trained CARE network was quantitatively and qualitatively evaluated by using image pairs from 45 unseen ROIs. CARE denoising improved the image quality significantly, increasing similarity with the ground truth image by 1.9 times, reducing noise by 2.35 times, and increasing SNR by 7.4 dB. Banding noise, prominent in input images, was significantly reduced in CARE denoised images. CARE denoising provided quantitatively and qualitatively better noise reduction than non-DL filtering methods. Qualitative image assessment by three confocal readers showed that CARE denoised images exhibited negligible noise more often than input images and non-DL filtered images. Results showed the potential of using a DL-based method for denoising PCM images obtained at a high imaging speed. The DL-based denoising method needs to be further trained and tested for PCM images obtained from disease-suspicious skin lesions.

Reducing Motion Blur in Ghost Imaging Via the Hessian Matrix

Different from conventional imaging, ghost imaging (GI) is an indirect modality of imaging that needs multiple measurements of the second-order correlation of data collected from two detectors. In some particular cases, the exposure time of two detectors or the rotation speed of the ground glass may not meet the need of experimental condition, resulting in motion blur that reduces the quality of the reconstructed image. In this paper, we propose a method to solve this problem. By convolving the data from the reference arm with the Hessian matrix, the intensity of the light in the data is replaced by the gradient of intensity and the influence of the motion blur in the reconstructed image can be reduced.