Effect Of Blur Research Articles

Standardization of histogram- and gray-level co-occurrence matrices-based radiomics in the presence of blur and noise

Radiomics have been extensively investigated as quantitative biomarkers that can enhance the utility of imaging studies and aid the clinical decision making process. A major challenge to the clinical translation of radiomics is their variability as a result of different imaging and reconstruction protocols. In this work, we present a novel radiomics standardization framework capable of modeling and recovering the underlying radiomic feature in images that have been corrupted by the effects of spatial resolution and noise. We focus on two classes of radiomics based on pixel value distributions—i.e. histograms and gray-level co-occurrence matrices (GLCMs). We developed a model that predicts these distributions in the presence of system blur and noise, and used that model to invert these physical effects and recover the underlying distributions. Specifically, the effect of blur on histogram and GLCM is highly image-dependent, while additive noise convolves the histogram/GLCM of the noiseless image with those of the noise. The recovery method therefore consists of two deconvolution operations: the first in the image domain to remove the effect of system blur, the second in the histogram/GLCM domain to remove the effect of noise. The performance of the proposed recovery strategy was investigated using a set of texture phantoms and an emulated computed tomography imaging chain with a range of realistic blur and noise levels. The proposed method was able to obtain histogram and GLCM estimates that closely resemble the ground truth. The method performed well across imaging conditions and significantly lowered the variability associated with different imaging protocols. This improvement also translated to better classification accuracy, where recovered radiomic values result in greater separation of radiomic clusters for two different texture phantoms as compared to values derived from the original blurred and noisy images. In summary, the novel radiomics standardization framework demonstrates high potential for mitigating radiomic variability as a result of the imaging system and can potentially be integrated as a preprocessing step towards more robust and reproducible radiomic models.

The effect of induced monocular blur by bangerter filters on measures of visual acuity and stereoacuity

ABSTRACT To evaluate the effect of monocular blur induced by Bangerter filters (BF) on visual acuity (VA) and stereoacuity. Monocular blur was induced by a range of Trusetal BF strengths (0.1, 0.2, 0.4, 0.6) to 23 visually normal adult volunteers (aged 18-25, mean age 20.33 ± 1.79 years). The right monocular and binocular logMAR VA and distance stereoacuity using the FD2 were measured at 6 m with no filter (baseline) and with each filter strength. The order of testing the filters was randomized. Results were analyzed using one factor repeated measures ANOVAs, t–tests using Bonferroni correction and Pearson’s product moment correlation. All filters degraded right monocular distance VA from baseline, but less significantly between the 0.4 and 0.6 filters than the other filters (p < .05 and p < .001, respectively). Degradation was in concordance with the labeled filter density, excluding filter strength 0.4. For all filter strengths, binocular VA was significantly but not clinically reduced from baseline (p < .05). Filters statistically and clinically significantly degraded distance stereoacuity from baseline (p < .001). A significant negative correlation existed between the mean degraded right monocular VA and stereoacuity (r = −0.998, p < .02). Trusetal BF significantly reduced monocular VA and stereoacuity but binocular VA remained within normal levels. The impact on VA and stereoacuity in a normal population can be of clinical relevance when applied to the use of BF for management of amblyopia and intractable diplopia. It is advised that clinicians ensure the desired level of degradation has been achieved in clinic before prescribing a particular filter strength.

본태적 난시도 그룹간 난시성 흐림이 대비감도에 미치는 영향

본태적 난시도 그룹간 난시성 흐림이 대비감도에 미치는 영향

A reference-free underwater image quality assessment metric in frequency domain

Owing to the complexity of the underwater environment and the limitations of imaging devices, the quality of underwater images varies differently, which may affect the practical applications in modern military, scientific research, and other fields. Thus, achieving subjective quality assessment to distinguish different qualities of underwater images has an important guiding role for subsequent tasks. In this paper, considering the underwater image degradation effect and human visual perception scheme, an effective reference-free underwater image quality assessment metric is designed by combining the colorfulness, contrast, and sharpness cues. Specifically, inspired by the different sensibility of humans to high-frequency and low-frequency information, we design a more comprehensive color measurement in spatial domain and frequency domain. In addition, for the low contrast caused by the backward scattering, we propose a dark channel prior weighted contrast measure to enhance the discrimination ability of the original contrast measurement. The sharpness measurement is used to evaluate the blur effect caused by the forward scattering of the underwater image. Finally, these three measurements are combined by the weighted summation, where the weighed coefficients are obtained by multiple linear regression. Moreover, we collect a large dataset for underwater image quality assessment for testing and evaluating different methods. Experiments on this dataset demonstrate the superior performance both qualitatively and quantitatively.

「蛇の回転」錯視における刺激位置・画像ぼけ・移動方法の効果

「蛇の回転」錯視における刺激位置・画像ぼけ・移動方法の効果

Saliency-Aware Subtle Augmentation Improves Human Visual Search Performance in VR.

Visual search becomes challenging when the time to find the target is limited. Here we focus on how performance in visual search can be improved via a subtle saliency-aware modulation of the scene. Specifically, we investigate whether blurring salient regions of the scene can improve participant’s ability to find the target faster when the target is located in non-salient areas. A set of real-world omnidirectional images were displayed in virtual reality with a search target overlaid on the visual scene at a pseudorandom location. Participants performed a visual search task in three conditions defined by blur strength, where the task was to find the target as fast as possible. The mean search time, and the proportion of trials where participants failed to find the target, were compared across different conditions. Furthermore, the number and duration of fixations were evaluated. A significant effect of blur on behavioral and fixation metrics was found using linear mixed models. This study shows that it is possible to improve the performance by a saliency-aware subtle scene modulation in a challenging realistic visual search scenario. The current work provides an insight into potential visual augmentation designs aiming to improve user’s performance in everyday visual search tasks.

Experimental Comparison between Event and Global Shutter Cameras.

We compare event-cameras with fast (global shutter) frame-cameras experimentally, asking: “What is the application domain, in which an event-camera surpasses a fast frame-camera?” Surprisingly, finding the answer has been difficult. Our methodology was to test event- and frame-cameras on generic computer vision tasks where event-camera advantages should manifest. We used two methods: (1) a controlled, cheap, and easily reproducible experiment (observing a marker on a rotating disk at varying speeds); (2) selecting one challenging practical ballistic experiment (observing a flying bullet having a ground truth provided by an ultra-high-speed expensive frame-camera). The experimental results include sampling/detection rates and position estimation errors as functions of illuminance and motion speed; and the minimum pixel latency of two commercial state-of-the-art event-cameras (ATIS, DVS240). Event-cameras respond more slowly to positive than to negative large and sudden contrast changes. They outperformed a frame-camera in bandwidth efficiency in all our experiments. Both camera types provide comparable position estimation accuracy. The better event-camera was limited by pixel latency when tracking small objects, resulting in motion blur effects. Sensor bandwidth limited the event-camera in object recognition. However, future generations of event-cameras might alleviate bandwidth limitations.

DIGITAL IMAGE PROCESSING AND RECOGNITION USING PYTHON

In this project, the primary aim will be the conversion of images into Grayscale in which conversion of pixels to array takes place and apply Blur effect using The Gaussian blur which is a type of image-blurring filter that uses a Gaussian function which also expresses the normal distribution in statistics for calculating the transformation to apply to each pixel in the image. The above two processesare applied to the input images. These two above mentioned processes can be achieved by utilizing the most relevant python libraries and functions, followed by conversion of the digital image to numerical data and then, applying the effects to the image to get back the image with applied effects in it. Face recognition refers to matching a face present in an input image from the training/pre-saved dataset and by applying Deep Learning Concept. This will be achieved by defining a function to read and convert images to data, apply the python function, and then, recreating the image with results.

Iterative multi‐scale residual network for deblurring

In dynamic scene deblurring, recent neural network–based methods have been very successful. But with the improvement of deep deblurring performance, network structure and learning become more complicated. Compared with large-scale network parameters and complex network structures, an iterative multi-scale residual network to achieve a more effective parameter sharing scheme is proposed. In each iterative unit, fast multi-scale residual blocks to replace superimposed convolutional layers or classic residual blocks are used. On the basis of preventing model overfitting, the receptive field of the network is increased. At the same time, the gated recurrent unit is introduced to connect modules of different stages. The model does not rely on the estimation of the blur kernel and directly generates sharp images in an end-to-end manner. The experimental structure on the benchmark dataset and real-world images showed that this method has better quality than the existing methods in terms of large-scale blur and subjective perception effects, both in quantitative and qualitative terms.

Exploring the Effects of Blur and Deblurring to Visual Object Tracking

The existence of motion blur can inevitably influence the performance of visual object tracking. However, in contrast to the rapid development of visual trackers, the quantitative effects of increasing levels of motion blur on the performance of visual trackers still remain unstudied. Meanwhile, although image-deblurring can produce visually sharp videos for pleasant visual perception, it is also unknown whether visual object tracking can benefit from image deblurring or not. In this paper, we present a Blurred Video Tracking (BVT) benchmark to address these two problems, which contains a large variety of videos with different levels of motion blurs, as well as ground-truth tracking results. To explore the effects of blur and deblurring to visual object tracking, we extensively evaluate 25 trackers on the proposed BVT benchmark and obtain several new interesting findings. Specifically, we find that light motion blur may improve the accuracy of many trackers, but heavy blur usually hurts the tracking performance. We also observe that image deblurring is helpful to improve tracking accuracy on heavily-blurred videos but hurts the performance of lightly-blurred videos. According to these observations, we then propose a new general GAN-based scheme to improve a tracker's robustness to motion blur. In this scheme, a fine-tuned discriminator can effectively serve as an adaptive blur assessor to enable selective frames deblurring during the tracking process. We use this scheme to successfully improve the accuracy of 6 state-of-the-art trackers on motion-blurred videos.

Retracted] High‐Precision 3D Reconstruction of Cooperative Markers under Motion Blur

The application of artificial intelligence and deep learning in the fields of wireless communication, image and speech recognition, and 3D reconstruction has successfully solved some difficult modeling problems. This paper focuses on the high‐precision 3D reconstruction of the motion‐blurred cooperative markers, including the Chinese character coded targets (CCTs) and the noncoded circular markers. A simulation‐based motion‐blurred image generation model is constructed to provide sufficient samples for training the convolutional neural network to identify and match the motion‐blurred CCTs on the moving object. The blurred noncoded marker matching is performed through homography. The 3D reconstruction of the markers is realized via the optimization of the spatial moving path within the exposure period. The midpoint of the moving path of the markers is taken as the final reconstruction result. The experimental results show that the 3D reconstruction accuracy of the markers with a certain motion blur effect is about 0.08 mm.

An Evaluation of VGG16 Binary Classifier Deep Neural Network for Noise and Blur Corrupted Images

Deep learning networks has become an important tool for image classification applications. Distortions on images may cause the performance of a classifier to decrease significantly. In the present paper, a comparative investigation for binary classification performance of VGG16 network under corrupted inputs has been presented. For this purpose, images corrupted at various levels and fixed levels with Gaussian noise, Salt and Pepper noise and blur effect were used for testing. Convolutional layers of the VGG16 were frozen except the last three convolutional layers and a dense layer for binary classification was added. According to experimental results, as the effect of distortion is increased, performance of the deep learning classifier drops significantly. In the case of augmented training with distortion effects, the results were improved significantly.

Using super-resolution generative adversarial network models and transfer learning to obtain high resolution digital periapical radiographs

Periapical Radiographs are commonly used to detect several anomalies, like caries, periodontal, and periapical diseases. Even considering that digital imaging systems used nowadays tend to provide high-quality images, external factors, or even system limitations can result in a vast amount of radiographic images with low quality and resolution. Commercial solutions offer tools based on interpolation methods to increase image resolution. However, previous literature shows that these methods may create undesirable effects in the images affecting the diagnosis accuracy. One alternative is using deep learning-based super-resolution methods to achieve better high-resolution images. Nevertheless, the amount of data for training such models is limited, demanding transfer learning approaches. In this work, we propose the use of super-resolution generative adversarial network (SRGAN) models and transfer learning to achieve periapical images with higher quality and resolution. Moreover, we evaluate the influence of using the transfer learning approach and the datasets selected for it in the final generated images. For that, we performed an experiment comparing the performance of the SRGAN models (with and without transfer learning) with other super-resolution methods. Considering Mean Square Error (MSE), Peak Signal to Noise Ratio (PSNR), Structural Similarity Index (SSIM), and Mean Opinion Score (MOS), the results of SRGAN models using transfer learning were better on average. This superiority was also verified statistically using the Wilcoxon paired test. In the visual analysis, the high quality achieved by the SRGAN models, in general, is visible, resulting in more defined edges details and fewer blur effects.

Vision-based micro-manipulations in simulation

Manipulating micro-objects is needed in various applications such as micro-assembly, minimally invasive surgery and micro-organism manipulation, which refers to micro-manipulation in this paper. During such operations, controlling the gripping force is mandatory, in order to prevent damage or the breaking of the micro-parts being grasped and handled, given that they are often very fragile. The force sensing is frequently carried out by sensors embedded in the micro-gripper, but this produces strong design and manufacturing constraints consequently. In this paper, vision-based manipulation of micro-parts is investigated. Moreover, instead of working with real setups, this paper investigates micro-manipulations carried out in simulation. Working on simulators instead of real setups allows to easily and quickly implement different vision-based force measurement methods, different force control strategies and even different micro-gripper systems. Using the simulators, we are also able to compare the performance and limitations achieved in each manipulation, and to study the impact of certain parameters on manipulation performance. MATLAB-based simulators were designed in this paper. A vision-based force measurement method was then implemented and integrated into a closed-loop force control. In the end, successful grasp-hold-release manipulations of micro-parts (400–600 µm size) were achieved with a force-sensing resolution of less than 6 µN. Moreover, the manipulation was observed to be successful when softness variation is introduced on the micro-object (± 20% around the average value), and when pictures are degraded because of noise and blur effects.

Back-of-the-envelope image resolution estimation using an aberrated Rayleigh criterion.

In an optical imaging sensor (i.e., lens plus focal plane array), the critical output is the image itself, and quality of that image is a measure of the sensor's performance. Image quality is dependent on several terms such as the sensor-level point-spread function (PSF), sampling period, and signal-to-noise ratio. The sensor resolution is mainly dependent on the PSF, which is comprised of many aspects. Often, the resolution is determined by the diffraction-limited Rayleigh criterion; however, due to the blur contributions of the collecting optics wavefront errors, pixel size, etc., the sensor PSF is not often diffraction-limited. Here, we develop two simple back-of-the-envelope calculations that provide an estimate of a nondiffraction-limited imaging sensor PSF size based on the sensor-level required resolution. Moreover, an allocation based on this PSF size can be used to dole out an allotment to the various blur effects, e.g., wavefront error, during sensor-level design.

The Mysterious Green Streaks Below STEVE

Strong thermal emission velocity enhancement (STEVE) is an optical phenomenon of the subauroral ionosphere arising from extreme ion drift speeds. STEVE consists of two distinct components in true‐color imagery: a mauve or whitish arc extended in the magnetic east–west direction and a region of green emission adjacent to the arc, often structured into quasiperiodic columns aligned with the geomagnetic field (the “picket fence”). This work employs high‐resolution imagery by citizen scientists in a critical examination of fine‐scale features within the green emission region. Of particular interest are narrow “streaks” of emission forming underneath field‐aligned picket fence elements in the 100‐ to 110‐km altitude range. The streaks propagate in curved trajectories with dominant direction toward STEVE from the poleward side. The elongation is along the direction of motion, suggesting a drifting point‐like excitation source, with the apparent elongation due to a combination of motion blur and radiative lifetime effects. The cross‐sectional dimension is &lt;1 km, and the cases observed have a duration of ∼20–30 s. The uniform coloration of all STEVE green features in these events suggests a common optical spectrum dominated by the oxygen 557.7‐nm emission line. The source is most likely direct excitation of ambient oxygen by superthermal electrons generated by ionospheric turbulence induced by the extreme electric fields driving STEVE. Some conjectures about causal connections with overlying field‐aligned structures are presented, based on coupling of thermal and gradient‐drift instabilities, with analogues to similar dynamics observed from chemical release and ionospheric heating experiments.

Image enhancement comparison to improve underwater cultural heritage survey

This work aims at presenting an underwater image application to obtain an improved 3D model of cultural assets. In 2016, more than 500 images were acquired by a GoPro Camera with a low resolution of 72 dpi and focal length of 3 mm, without flash and are now used to reconstruct the 3D model of some amphoras of a Roman shipwreck found in Albenga (Italy). We have applied state-of-art image enhancement techniques, such as ACE, CLAHE, LAB and SP algorithms, to improve the quality of underwater images affected by low contrast, poor visibility conditions, not uniform lighting, colour variations, noise and blur effect. The visual quality has been evaluated through quantitative metrics, like average luminance, information entropy, average gradient of image, UCIQE and UIQM. Then, our efforts have been devoted to the dense 3D point cloud generation using a SfM-MVS software. In particular, the 3D reconstruction results are in line with the metric evaluation: in fact, the more accurate 3D objects are obtained from that enhanced dataset with the highest measured image quality.

How the Presence and Size of Static Peripheral Blur Affects Cybersickness in Virtual Reality

Cybersickness (CS) is one of the challenges that has hindered the widespread adoption of Virtual Reality and its applications. Consequently, a number of studies have focused on extensively understanding and reducing CS. Inspired by previous work that has sought to reduce CS using foveated rendering and Field of View (FOV) restrictions, we investigated how the presence and size of a static central window in peripheral FOV blurring affects CS. To facilitate this peripheral FOV blur, we applied a Gaussian blur effect in the display peripheral region, provisioning a full-resolution central window. Thirty participants took part in a three-session, within-subjects experiment, performing search and spatial updating tasks in a first-person, slow-walking, maze-traveling scenario. Two different central window sizes (small and large) were tested against a baseline condition that didn’t feature display peripheral blurring. Results revealed that the baseline condition produced higher levels of CS than both conditions with a central window. While there were no significant differences between the small and large windows, we observed interaction effects suggesting an influence of window size on “adaptation to CS.” When the central window is small, adaptation to CS seems to take more time but is more pronounced. The interventions had no effect on spatial updating and presence, but were detectable when the blurred area was larger (small central window). Lower sickness levels observed in both window conditions supports the use of peripheral FOV blurring to reduce CS, reducing our dependence on eye tracking. This being said, researchers must strive to find the right balance between window size and detectability to ensure seamless virtual experiences.

A Deconvolutional Reconstruction Method Based on Lucy–Richardson Algorithm for Joint Scanning Laser Thermography

Joint scanning laser thermography (JLST) is well-known for its efficiency to overcome the field of view (FOV) limitation of thermal imagers. However, JSLT requires a data reconstruction to reveal the location of the defective area straightforwardly. Moreover, its detection capacity is limited by the lack of a deconvolution algorithm adaptive to the reconstructed data. In this study, a deconvolutional reconstruction method based on the Lucy–Richardson (LR) algorithm has been developed for JST, which is effective in suppressing random noise and the blur effect caused by the thermal diffusion. A JSLT inspection is carried out on a functional coating material with cylinder-like defects to test the performance of the proposed method. In comparison to the directly processed method on the original data, the proposed method is processed on the reconstructed data and then compared with principal component analysis (PCA), restored pseudo heat flux (RPHF), fast Fourier transform (FFT) methods and non-negative matrix factorization (NMF). The experimental results indicated that our proposed LR method exhibited a higher signal-to-noise ratio. Besides, it can detect the cylinder-mocked debonding defects with a diameter of 1.5 mm and a depth of 2.0 mm buried under the 1.0-mm coating. In addition, the defect detection diameter-to-depth ratio reached 1.5, while the defect detection rate of the test specimens can approach 90%.

Recognition of rust grade and rust ratio of steel structures based on ensembled convolutional neural network

AbstractEnsembled convolutional neural network (ECNN) was utilized to recognize the rust grade and rust ratio of steel structure to partially replace traditional visual inspection. The performance of ECNN was demonstrated by theoretical analysis and experimental verification, and the application scenarios of ECNN in the task of rust grade recognition and rust ratio recognition were discussed. The accuracy of ECNN classifier reached 93%, which improves upon the highest accuracy of 90% achieved by using a single classifier. By visualizing the misclassified images, it was found that the rust grade of misclassified image is indistinguishable and the classifiers show strong fault tolerance. The ensembled model is more robust than the single model in the task of rust ratio recognition. Gaussian blur was applied to the test images to study the effect of image blur on model performance, and the results show that the rust segmentation model was not susceptible to image blur.