Blur Filter Research Articles

Post-processing techniques using 3D Slicer for T1-weighted MRI analysis of radiochromic gel dosimeters

Radiochromic FXG gel dosimeters were investigated for end-to-end validation of TG-119 plans delivered in the pre-clinical Elekta 1.5 T-7 MV FFF MR-Linac. Due to the noise levels in the T1-weighted MR images, this preliminary study investigated post-processing techniques available in 3D Slicer and their impact on gamma passing rates. Binomial blur and discrete Gaussian image filters were identified as the most impactful on increasing gamma passing rates for T1-weighted MR images; however, any application of image filters should be implemented with caution on the interpretation of the results.

875 Using Stress Testing to Identify Vulnerabilities in Artificial Intelligence Models for the Identification of Culprit Carotid Lesions in Cerebrovascular Events

Abstract Introduction Carotid atherosclerosis is a major risk factor for ischaemic stroke, a leading cause of death. Carotid CT angiography (CTA) is commonly performed following a stroke or transient ischaemic attack (TIA) to help guide patient management in secondary prevention of stroke. Deep learning algorithms can help extract greater information from scans. Method The dataset comprised CTA scans from 40 culprit and 40 non-culprit carotid arteries of patients with recent stroke/TIA, and 40 carotid arteries of asymptomatic patients without previous stroke/TIA. A 3D convolutional neural network was trained to classify carotid artery type. Each input comprised 14 axial CTA carotid patches (centred around the carotid artery) concatenated together to form a 3D volume (capturing ∼3cm of artery). 75% of the dataset was used for training and 25% for internal validation. Following training, computer vision operations were applied to input images to assess their impact on the model’s classification decisions. Results The model achieved 100% accuracy on the training set and 67% on the internal validation set. However, after subjecting input images to image operations, vulnerabilities in the deep learning model were revealed, even when using input images from the training set. For example, using a Gaussian blur filter with sigma 1.0 was sufficient to change classification decisions, as was horizontally flipping the image. Conclusions Deep learning has exceptional capabilities for learning, however the risk with such high-capacity models is failure to learn relevant features from the data. Stress testing provides a viable method to further evaluate deep learning models before clinical deployment.

Automatic Blob Detection for Dental Caries

Dental Caries are one of the most prevalent chronic diseases around the globe. Detecting carious lesions is a challenging task. Conventional computer aided diagnosis and detection methods in the past have heavily relied on the visual inspection of teeth. These methods are only effective on large and clearly visible caries on affected teeth. Conventional methods have been limited in performance due to the complex visual characteristics of dental caries images, which consist of hidden or inaccessible lesions. The early detection of dental caries is an important determinant for treatment and benefits much from the introduction of new tools, such as dental radiography. In this paper, we propose a deep learning-based technique for dental caries detection namely: blob detection. The proposed technique automatically detects hidden and inaccessible dental caries lesions in bitewing radio-graphs. The approach employs data augmentation to increase the number of images in the data set to have a total of 11,114 dental images. Image pre-processing on the data set was through the use of Gaussian blur filters. Image segmentation was handled through thresholding, erosion and dilation morphology, while image boundary detection was achieved through active contours method. Furthermore, the deep learning based network through the sequential model in Keras extracts features from the images through blob detection. Finally, a convexity threshold value of 0.9 is introduced to aid in the classification of caries as either present or not present. The process of detection and classifying dental caries achieved the results of 97% and 96% for the precision and recall values, respectively.

Vehicles Detection System at Different Weather Conditions

The importance of efficient vehicle detection (VD) is increased with the expansion of road networks and the number of vehicles in the Intelligent Transportation Systems (ITS). This paper proposes a system for detecting vehicles at different weather conditions such as sunny, rainy, cloudy and foggy days. The first step to the proposed system implementation is to determine whether the video’s weather condition is normal or abnormal. The Random Forest (RF) weather condition classification was performed in the video while the features were extracted for the first two frames by using the Gray Level Co-occurrence Matrix (GLCM). In this system, the background subtraction was applied by the mixture of Gaussian 2 (MOG 2) then applying a number of pre-processing operations, such as Gaussian blur filter, dilation, erosion, and threshold. The main contribution of this paper is to propose a histogram equalization technique for complex weather conditions instead of a Gaussian blur filter for improving the video clarity, which leads to increase detection accuracy. Based on the previous steps, the system defines each region in the frame expected to contain vehicles. Finally, Support Vector Machine (SVM) classifies the defined regions into a vehicle or not. As compared to the previous methods, the proposed system showed high efficiency of about 96.4% and ability to detect vehicles at different weather conditions.

Analysis of Crop disease detection with SVM, KNN and Random forest classification

Due to an uneven climatic condition crops are being affected which leads to decrease in agriculture yield. It greatly affects global agricultural economy. However, the condition becomes more worse when diseases are identified in crops. Agriculture plays a vital role in every country’s economy. Thus, there is a need to identify the crop disease before it is visible on a crop so that disease can be avoided by using appropriate measures. The traditional way of identifying a crop disease is through observation by naked eyes. But as it requires large number of experts and continuous monitoring of crop it will be costly for large fields. Hence, an automatic system is required which can not only examine the crops to detect disease but also can classify the type of disease on crops. The proposed system determines disease from an input image. The input image has to go through following stages: Image Acquisition, Image pre-processing, Image segmentation, Feature Extraction, and Classification in order to determine diseased crop and accordingly provides remedy for that disease. Infected crop image is taken as input in Image Acquisition stage. In Image pre-processing stage noise is removed from the input image by applying gaussian blur filter and non-local means denoising algorithm. Also, the background of image is eliminated using Thresholding algorithm. To extract Region of Interest (ROI) i.e. infected region from input image K-means Clustering algorithm is used. Then color, texture and shape features are extracted from ROI and features are further send to the classification stage. Three different classification algorithms namely Support Vector Machine (SVM), K-Nearest Neighbors (KNN) and Random Forest are implemented for classification out of which Support Vector Machine Algorithm is found to be best in terms of accuracy. Hence, classification is carried out by using Multivariate Support Vector Machine algorithm which detect disease present in crop accurately. In this way, the proposed system detects a disease from the given input image.

Mobile‐aided screening system for proliferative diabetic retinopathy

AbstractNeoVascularization (NV) occurs in the Proliferative Diabetic Retinopathy (PDR) stage, where the development progress of new vessels presents a high risk for severe vision loss and blindness. Therefore, early NV detection is primordial to preserve patient's vision. Several automated methods have been proposed to detect the NV on retinograph‐captured fundus images. However, their employment is constrained by the reduced ophthalmologist‐per‐person ratio and the expensive equipment for image capturing.This paper presents a novel method for NV detection in smartphone‐captured fundus images. The implementation of the method on a smartphone having an optical lens for fundus capturing leads to a Mobiles‐Aided‐Screening system of PDR (MAS‐PDR). The challenge is to ensure accurate and robust detection even with moderate quality of fundus image, on reduced execution time. Within this objective, we identify the major criteria of neovascularized vessels which are tortuosity, width, bifurcation, and density. Our main contribution consists in proposing a sharp feature to reflect each criterion on reduced computational complexity processing. Therefore, the features are provided to a random forest classifier to deduce the PDR stage.A dataset raised from publicly databases is used on a 10‐cross validation process where 98.69% accuracy, 97.73% sensitivity, and 99.12% specificity are achieved. To evaluate the robustness, the same experimentation is repeated after applying motion blur filters to the fundus image dataset, where 98.91% accuracy, 96.75% sensitivity, and 100% specificity are deduced. Moreover, NV screening is performed under 3 s when executed in smartphone devices demonstrating the appropriateness of our method to MAS‐PDR.

Dental Images' Segmentation Using Threshold Connected Component Analysis.

Recent advances in medical imaging analysis, especially the use of deep learning, are helping to identify, detect, classify, and quantify patterns in radiographs. At the center of these advances is the ability to explore hierarchical feature representations learned from data. Deep learning is invaluably becoming the most sought out technique, leading to enhanced performance in analysis of medical applications and systems. Deep learning techniques have achieved great performance results in dental image segmentation. Segmentation of dental radiographs is a crucial step that helps the dentist to diagnose dental caries. The performance of these deep networks is however restrained by various challenging features of dental carious lesions. Segmentation of dental images becomes difficult due to a vast variety in topologies, intricacies of medical structures, and poor image qualities caused by conditions such as low contrast, noise, irregular, and fuzzy edges borders, which result in unsuccessful segmentation. The dental segmentation method used is based on thresholding and connected component analysis. Images are preprocessed using the Gaussian blur filter to remove noise and corrupted pixels. Images are then enhanced using erosion and dilation morphology operations. Finally, segmentation is done through thresholding, and connected components are identified to extract the Region of Interest (ROI) of the teeth. The method was evaluated on an augmented dataset of 11,114 dental images. It was trained with 10 090 training set images and tested on 1024 testing set images. The proposed method gave results of 93% for both precision and recall values, respectively.

Neighbour Local Variability for Multi-Focus Images Fusion

The goal of multi-focus image fusion is to integrate images with different focus objects in order to obtain a single image with all focus objects. In this paper, we give a new method based on neighbour local variability (NLV) to fuse multi-focus images. At each pixel, the method uses the local variability calculated from the quadratic difference between the value of the pixel and the value of all pixels in its neighbourhood. It expresses the behaviour of the pixel with respect to its neighbours. The variability preserves the edge function because it detects the sharp intensity of the image. The proposed fusion of each pixel consists of weighting each pixel by the exponential of its local variability. The quality of this fusion depends on the size of the neighbourhood region considered. The size depends on the variance and the size of the blur filter. We start by modelling the value of the neighbourhood region size as a function of the variance and the size of the blur filter. We compare our method to other methods given in the literature. We show that our method gives a better result.

GESTURE RECOGNITION SYSTEM FOR NIGERIAN TRIBAL GREETING POSTURES USING SUPPORT VECTOR MACHINE

This paper presents the application of Gaussian blur filters and Support Vector Machine (SVM) techniques for greeting recognition among the Yoruba tribe of Nigeria. Existing efforts have considered different recognition gestures. However, tribal greeting postures or gestures recognition for the Nigerian geographical space has not been studied before. Some cultural gestures are not correctly identified by people of the same tribe, not to mention other people from different tribes, thereby posing a challenge of misinterpretation of meaning. Also, some cultural gestures are unknown to most people outside a tribe, which could also hinder human interaction; hence there is a need to automate the recognition of Nigerian tribal greeting gestures. This work hence develops a Gaussian Blur – SVM based system capable of recognizing the Yoruba tribe greeting postures for men and women. Videos of individuals performing various greeting gestures were collected and processed into image frames. The images were resized and a Gaussian blur filter was used to remove noise from them. This research used a moment-based feature extraction algorithm to extract shape features that were passed as input to SVM. SVM is exploited and trained to perform the greeting gesture recognition task to recognize two Nigerian tribe greeting postures. To confirm the robustness of the system, 20%, 25% and 30% of the dataset acquired from the preprocessed images were used to test the system. A recognition rate of 94% could be achieved when SVM is used, as shown by the result which invariably proves that the proposed method is efficient.

A convolutional neural network model for marble quality classification

The fundamental policy of marble industries is to establish sustainable high-quality products in a standardized manner. Identification and classification of different types of marbles is a critical task that is usually carried out by human experts. However, marble quality classification by humans can be time-consuming, error-prone, inconsistent, and subjective. Automated and computerized approaches are required to obtain faster, more reliable, and less subjective results. In this study, a deep learning model is developed to perform multi-classification of marble slab images with six different quality types. Blur filter, 5 ✕ 5 low-pass 2D linear separable convolution filter using Gaussian kernel, and erosion filter were applied to the images for data augmentation, and a special convolutional neural network (CNN) architecture was designed and implemented. It has been observed that the data augmentation approach for marble image samples has significantly improved the accuracy of the CNN model ranging between 0.922 and 0.961.

Measurement of multifractal character of bubble paths using image analysis

The trajectory of the virtual mass centers of sequentially departing gas bubbles in water flow has been analyzed in this study. The bubbles were generated from a brass nozzle of 1 mm inner diameter in a glass tank (300 × 150 × 500 mm3), filled with distilled water maintained at 21 ± 1 °C. A three-mirror system was used to obtain two frames of the workspace in a single shot of a Phantom v1610 high-speed camera. A three-dimensional computer reconstruction algorithm was developed, which consisted of a blur filter, canny edge filter, region of interest (ROI), and Kalman filter. This algorithm was used to obtain the coordinates of the center of mass of each individual gas bubble. We analyzed the bubble trajectories obtained for three different air volumetric flow rates. A multifractal analysis was used to evaluate the gas bubble dynamics. The wavelet transform modulus maxima method was used to obtain the singularity spectra of the bubble paths. The local Hölder exponent value proved that the gas bubble motion trajectories tended to behave in a deterministic and chaotic manner. In addition, the sensitivity of the chaotic bubble paths to small- and large-scale fluctuations of the lateral bubble displacement was investigated. In conclusion, the proposed approach allows us to distinguish between deterministic chaotic dynamics and the stochastic evolution of bubble paths.

Image-obfuscation as a means for privacy-conscious visual data acquisition from building systems

In the last two decades, numerous studies have demonstrated the viability of using High Dynamic Range Imaging (HDRI) to quantify lighting conditions in the built environment. Several human factor studies have demonstrated correlation between visual comfort perceived by occupants and glare metrics calculated by analysing HDR images. However, the use of HDRI in real-world applications has been severely limited owing to privacy concerns. This research investigates the feasibility of employing obfuscated (i.e. deliberately distorted) HDR images for analysing glare. The authors present a pilot study where visual conditions inside an office-space were simulated and captured as HDR images using a validated, physically-based renderer. The images were then obfuscated to various degrees by application of blur filters. Glare metrics calculated for the obfuscated images, when compared with the metrics generated for the original HDR images, were found to be within 2%-12% relative error. The proof-of-concept demonstrated through this study provides the framework for field-testing of an HDR-based lighting control system in a real office space.

Qualitative assessment of visual impairment in patients with age-related macular degeneration using standardized, image-based questionnaires

The purpose of the study is to test a simple approach for characterizing subjective perception in patients with choroidal neovascularization (CNV) secondary to age-related macular degeneration (AMD). This multi-center study was performed in four private ophthalmology practices including 20 patients with CNV. Subjects were presented the original image on the affected eye and afterwards modified candidate images to the contralateral eye. Then they chose one image that most closely matched the perception on the affected eye. This was repeated with three different images representing three everyday life scenes manipulated to mimic visual perception with reduced color perception (image 1), central scotomas (image 2), and blurring with four varying sizes (image 3). The frequency of response to the forced choice questionnaire experiment for each image was recorded. Our results show that images manipulated using a Gaussian blur filter, a brightness filter, and a sinusoidal distortion filter were selected by 9, 10, and 1 patients, respectively. Size of modification was not associated with visual acuity in image 1 or image 3. In image 2, however, the size of the modification was dependent on visual acuity ( p = .01). In conclusion, subjective perception in patients with AMD may significantly differ between individual patients. A better understanding of the visual disturbance will facilitate communication and involvement of patients in treatment decisions.

Accuracy Evaluation of Videogrammetry Using A Low-Cost Spherical Camera for Narrow Architectural Heritage: An Observational Study with Variable Baselines and Blur Filters

Three-dimensional (3D) reconstruction using video frames extracted from spherical cameras introduces an innovative measurement method in narrow scenes of architectural heritage, but the accuracy of 3D models and their correlations with frame extraction ratios and blur filters are yet to be evaluated. This article addresses these issues for two narrow scenes of architectural heritage that are distinctive in layout, surface material, and lighting conditions. The videos captured with a hand-held spherical camera (30 frames per second) are extracted to frames with various ratios starting from 10 and increasing every 10 frames (10, 20, …, n). Two different blur assessment methods are employed for comparative analyses. Ground truth models obtained from terrestrial laser scanning and photogrammetry are employed for assessing the accuracy of 3D models from different groups. The results show that the relative accuracy (median absolute errors/object dimensions) of spherical-camera videogrammetry range from 1/500 to 1/2000, catering to the surveying and mapping of architectural heritage with medium accuracy and resolution. Sparser baselines (the length between neighboring image pairs) do not necessarily generate higher accuracy than those from denser baselines, and an optimal frame network should consider the essential completeness of complex components and potential degeneracy cases. Substituting blur frames with adjacent sharp frames could reduce global errors by 5–15%.

Blind deblurring of natural stochastic textures using an anisotropic fractal model and phase retrieval algorithm.

The challenging inverse problem of blind deblurring has been investigated thoroughly for natural images. Existing algorithms exploit edge-type structures, or similarity to smaller patches within the image, to estimate the correct blurring kernel. However, these methods do not perform well enough on natural stochastic textures (NST), which are mostly random and in general are not characterized by distinct edges and contours. In NST even small kernels cause severe degradation to images. Restoration poses therefore an outstanding challenge. In this work, we refine an existing method by implementing an anisotropic fractal model to estimate the blur kernel's power spectral density. The final kernel is then estimated via an adaptation of a phase retrieval algorithm, originally proposed for sparse signals. We further incorporate additional constraints that are specific to blur filters, to yield even better results. The latter are compared with results obtained by recently published blind deblurring methods.

An approach to a robust watermark extraction from images containing text

This paper presents an approach to a robust watermark extraction from images containing text. Data extraction based on developed approach to robust watermark embedding into text data, characterizing by conversion invariance of text data into an image format. The comparative analysis of existing approaches of steganographic data embedding into text data is carried out, their advantages and disadvantages are determined. The choice of groups to steganographic data embedding methods based on text formatting is justified. As an embedding algorithm is determined approach based on interline space shifting. The block diagram and the description of the developed algorithm of data embedding into text data are given. An experimental estimation of the embedding capacity and perceptual invisibility of the developed data embedding approach was carried out. An approach to extract embedded information from images containing a robust watermark, based on the existing limitations, has been developed. The Radon transform is chosen as the basic extraction procedure of embedded information, allowing to extract values of the interline spacing. An approach based on Gaussian mixture model separating to isolate the values of the bits was chosen. The limits of the retrieval of embedded data have been experimentally established, and the robustness of the developed embedding approach to the implementation of various transformations has been estimated. The following parameters of robustness developed approach are defined: rotation of an image containing embedded data at any angle; scaling an image with a scaling factor not exceeding 1.5; conversion to any bitmap format; the application of a median filter to an image with a convolution core limit of not more than 9, a Gaussian blur filter with a blurring limit not exceeding 8 and an average filter with a convolution kernel limit of not more than 5.

Significance-Driven Logic Compression for Energy-Efficient Multiplier Design

Approximate arithmetic has recently emerged as a promising paradigm for many imprecision-tolerant applications. It can offer substantial reductions in circuit complexity, delay, and energy consumption by relaxing accuracy requirements. In this paper, we propose a novel energy-efficient approximate multiplier design using a significance-driven logic compression (SDLC) approach. Fundamental to this approach is an algorithmic and configurable lossy compression of the partial product rows based on their progressive bit significance. This is followed by the commutative remapping of the resulting product terms to reduce the number of product rows. As such, the complexity of the multiplier in terms of logic cell counts and lengths of critical paths is drastically reduced. A number of multipliers with different bit-widths (4-bit to 128-bit) are designed in SystemVerilog and synthesized using Synopsys Design Compiler. Post-synthesis experiments showed that up to an order of magnitude energy savings, and reductions of 65% in critical delay, and almost 45% in silicon area can be achieved for an 128-bit multiplier, compared with an accurate equivalent. These gains are achieved with low accuracy losses estimated at less than 0.0028 mean relative error. Additionally, we demonstrate the performance-energy-quality tradeoffs for different degrees of compression, achieved through configurable logic clustering. While evaluating the effectiveness of the proposed approach three case studies were set up. First, a Gaussian blur filter was designed, which demonstrated up to 80% energy reduction with a meagre loss of image quality. Second, we evaluate our approach in machine learning application using perceptron classifier, showed up to 74% energy reduction with negligible error rate. Third, the proposed multiplier designs were used in a power-constrained image processing application. We showed that SDLC can achieve $60\times $ improvement in computation capability, with potential to be employed in ubiquitous systems.

Enhancement of Principal Component Analysis using Gaussian Blur Filter

Characteristic evolving is most serious move that deal with image discrimination. It makes the content of images as ideal as possible. Gaussian blur filter used to eliminate noise and add purity to images. Principal component analysis algorithm is a straightforward and active method to evolve feature vector and to minimize the dimensionality of data set, this paper proposed using the Gaussian blur filter to eliminate noise of images and improve the PCA for feature extraction. The traditional PCA result as total average of recall and precision are (93% ,97%) and for the improved PCA average recall and precision are (98% ,100%), this show that the improved PCA is more effective in recall and precision.

Optimal digital design of steerable differentiators with the flatness of polynomial filters and the isotropy of Gaussian filters

Shape- and scale-selective digital-filters, with steerable finite/infinite impulse responses (FIR/IIRs) and non-recursive/recursive realizations, that are separable in both spatial dimensions and adequately isotropic, are derived. The filters are conveniently designed in the frequency domain via derivative constraints at dc, which guarantees orthogonality and monomial selectivity in the pixel domain (i.e. vanishing moments), unlike more commonly used FIR filters derived from Gaussian functions. A two-stage low-pass/high-pass architecture, for blur/derivative operations, is recommended. Expressions for the coefficients of a low-order IIR blur filter with repeated poles are provided, as a function of scale; discrete Butterworth (IIR), and colored Savitzky-Golay (FIR), blurs are also examined. Parallel software implementations on central processing units (CPUs) and graphics processing units (GPUs), for scale-selective blob-detection in aerial surveillance imagery, are analyzed. It is shown that recursive IIR filters are significantly faster than non-recursive FIR filters when detecting large objects at coarse scales, i.e. using filters with long impulse responses; however, the margin of outperformance decreases as the degree of parallelization increases.

A Solution to Image Processing with Parallel MPI I/O and Distributed NVRAM Cache

The paper presents a new approach to parallel image processing using byte addressable, non-volatile memory (NVRAM). We show that our custom built MPI I/O implementation of selected functions that use a distributed cache that incorporates NVRAMs located in cluster nodes can be used for efficient processing of large images. We demonstrate performance benefits of such a solution compared to a traditional implementation without NVRAM for various sizes of buffers used to read image parts, process and write back to storage. We also show that our implementation benefits from overlapping reading subsequent images while processing already loaded ones. We present results obtained in a cluster environment for three parallel implementation of blur, multipass blur and Sobel filters, for various NVRAM parameters such as latencies and bandwidth values.