Gaussian Smoothing Research Articles

VSTNet: Robust watermarking scheme based on voxel space transformation for diffusion tensor imaging images

Deep watermarking has been successfully applied to real-value domain (e.g., CT, MRI, DWI) medical images for copyright protection; however, manifold-value domain images, such as diffusion tensor imaging (DTI), have received less academic attention. The non-Euclidean nature of DTI images hinders deep models from generating plausible DTI images because it is difficult to guarantee that each generated DTI image voxel lies on a 3 × 3 symmetric positive definite manifold. In this paper, we propose a robust watermarking scheme based on voxel space transformation for DTI images. First, a voxel space transformation is proposed, which effectively solves the problem of fiber orientation estimation in Riemann networks for DTI images and obtains the eigenvalues and eigenvectors of each voxel using singular value decomposition (SVD) to transform the DTI voxels from the 3 × 3 symmetric positive definite manifold space to Euclidean space. Second, SVD is combined with a deep neural network to extract the high-level features of each voxel eigenvalue in the DTI images and embedding watermarking messages, avoiding the false positive errors problem in traditional watermarking algorithms based on an SVD transform. Finally, using a deep neural network combining multiscale dilated convolution, dense residual connection and channel attention, the algorithm demonstrates high robustness against various intentional or unintentional attacks on DTI images while ensuring the good visual quality and diffusion characteristics of the DTI images embedded with watermarking messages. Experiments show that the watermarking message extraction accuracy reaches 97.8% for Crop (p = 0.7) attacks and 95.4% for Gaussian Blur (k = 7) attacks.

Image processing methods based on physical models

In order to achieve better image processing results, the author proposes research on image processing methods based on physical models. The author first developed a simple and user-friendly GUIDE interactive interface based on Matlab software and image processing technology. Calculate and process the interference fringe patterns captured in static and dynamic experimental measurements in optical experiments separately to achieve experimental measurement objectives. Secondly, the image processing program was analyzed to highlight important information using methods such as Gaussian blur and binarization, and algorithms were used to construct surface 3D images or detect stripe motion features, establish 3D models, and accurately calculate corresponding experimental results. The final experimental results indicate that the phase transition value of piezoelectric devices increases linearly with the increase of voltage, which is consistent with the working characteristics of piezoelectric devices in the linear region. This program can analyze the image sets of horizontal, vertical, and oblique stripes, and calculate the piezoelectric moduli of piezoelectric devices to be 3.355 nm/V, 3.120 nm/V, and 3.557 nm/V, respectively, with relative errors of E1 = 4.9 %, E2 = 2.9 %, and E3 = 12 %. This method is suitable for various complex experimental measurement scenarios and provides an intuitive method for innovative design of optical interference experimental measurement teaching.

Effect of Matching Algorithm and Profile Shape on Pulsar Pulse Time of Arrival Uncertainties

For high-precision pulsar timing analysis and low-frequency gravitational wave detection, it is essential to accurately determine pulsar pulse times of arrival (ToAs) and associated uncertainties. To measure the ToAs and their uncertainties, various cross-correlation-based techniques can be employed. We develop methodologies to investigate the impact of the template-matching method, profile shape, signal-to-noise ratio of both template and observation on ToA uncertainties. These methodologies are then applied to data from the International Pulsar Timing Array. We demonstrate that the Fourier domain Markov chain Monte Carlo method is generally superior to other methods, while the Gaussian interpolation shift method outperforms other methods in certain cases, such as profiles with large duty cycles or smooth profiles without sharp features. However, it is important to note that our study focuses solely on ToA uncertainty, and the optimal method for determining both ToA and ToA uncertainty may differ.

An Image Denoising Technique Using Wavelet-Anisotropic Gaussian Filter-Based Denoising Convolutional Neural Network for CT Images

Denoising computed tomography (CT) medical images is crucial in preserving information and restoring images contaminated with noise. Standard filters have extensively been used for noise removal and fine details’ preservation. During the transmission of medical images, noise degrades the visibility of anatomical structures and subtle abnormalities, making it difficult for radiologists to accurately diagnose and interpret medical conditions. In recent studies, an optimum denoising filter using the wavelet threshold and deep-CNN was used to eliminate Gaussian noise in CT images using the image quality index (IQI) and peak signal-to-noise ratio (PSNR). Although the results were better than those with traditional techniques, the performance resulted in a loss of clarity and fine details’ preservation that rendered the CT images unsuitable. To address these challenges, this paper focuses on eliminating noise in CT scan images corrupted with additive Gaussian blur noise (AGBN) using an ensemble approach that integrates anisotropic Gaussian filter (AGF) and wavelet transform with a deep learning denoising convolutional neural network (DnCNN). First, the noisy image is denoised by AGF and Haar wavelet transform as preprocessing operations to eliminate AGBN. The DnCNN is then combined with AGF and wavelet for post-processing operation to eliminate the rest of the noises. Specifically, we used AGF due to its adaptability to edge orientation and directional information, which prevents blurring along edges for non-uniform noise distribution. Denoised images are evaluated using PSNR, mean squared error (MSE), and the structural similarity index measure (SSIM). Results revealed that the average PSNR value of the proposed ensemble approach is 28.28, and the average computational time is 0.01666 s. The implication is that the MSE between the original and reconstructed images is very low, implying that the image is restored correctly. Since the SSIM values are between 0 and 1.0, 1.0 perfectly matches the reconstructed image with the original image. In addition, the SSIM values at 1.0 or near 1.0 implicitly reveal a remarkable structural similarity between the denoised CT image and the original image. Compared to other techniques, the proposed ensemble approach has demonstrated exceptional performance in maintaining the quality of the image and fine details’ preservation.

Analysis of Optical Character Recognition using EasyOCR under Image Degradation

This project explores EasyOCR’s performance with Latin characters under image degradation. Variables like character-background intensity difference, Gaussian blur, and relative character size were tested. EasyOCR excels in distinguishing unique lowercase and uppercase characters but tends to favor uppercase for similar shapes like C, S, U, or Z. Results showed that high character-background intensity differences affected OCR output, with confidence scores ranging from 3 % to 80%. Higher differences caused confusion between characters like o and 0, or i and 1. Increased Gaussian blur hindered recognition but improved it for certain letters like v. Image size had a significant impact, with character detection failing as sizes decreased to 40% to 30% of the original. These findings provide insights into EasyOCR’s capabilities and limitations with Latin characters under image degradation.

Recent Advancement in Prediction and Analyzation of Brain Tumour using the Artificial Intelligence Method

Brain tumour identification and categorization are critical for diagnosis and treatment. This work uses preprocessing and classification algorithms to discover and categories brain tumours. Gaussian smoothing reduces noise and improves image quality, Genetic Algorithms select and optimize features, Deep Learning-Based Segmentation accurately segments tumours, Local Binary Patterns (LBP) extract texture features, and K-Nearest Neighbors classify tumours. Gaussian smoothing reduces noise and improves brain imaging data. Genetic Algorithms extract the most relevant and discriminative features from preprocessed photos. A Deep Learning-Based Segmentation model accurately segments brain tumour regions using these features. After segmentation, Local Binary Patterns (LBP) extract tumour texture features. The K-Nearest Neighbors (KNN) method classifies tumours using these texture features to capture tumour spatial patterns. Our suggested brain tumour detection and classification method combines several techniques to increase accuracy and reliability. Gaussian smoothing and LBP feature extraction improve feature discrimination. Deep Learning-Based Segmentation and the KNN classifier ensure precise tumour location and robust classification. The proposed method will be tested on brain scans including tumour areas. Classification performance measures include accuracy, sensitivity, specificity, and AUC. This work will improve brain tumour detection and classification methods for more accurate diagnoses and treatment planning. The primary goal of the work is to enhance brain tumour identification and categorization using pre processing, classification algorithms, and advanced techniques, ultimately improving diagnosis and treatment outcomes.

Integrated Approach for Cobb Angle Estimation in X-Ray C-Curve Scoliosis Using Image Processing and Inverse Cosine Law

Cobb angle is the angle subtended by the most tilted vertebrae. It represents the degree of the spinal curvature and has been used to diagnose scoliosis. The severity of the scoliosis based on the cobb angle will determine the guide treatment and surgical planning. Manual measurement of Cobb angle using protractor or semi-manual using ONIS software is quite time consuming, and it is subjected to variations in interobserver and intraobserver measurements. This research proposed an algorithm to determine the Cobb angle using image processing and inverse cosine methods. The raw samples of X-ray images were obtained from Pusat Kesihatan Universiti (PKU), UTHM. Several interviews were conducted with radiologist at Pantai Hospital for verification purpose. Gaussian blur and unsharp mask were initially applied for noise removal in the preprocessing step. Three points were marked at the vertebrae image to assist for centroid estimation using Haar Cascade. The Cobb angle was estimated by applying the triangle formula derived from the inverse cosine law. In comparison to semi-manual method (Onis Software), the proposed technique has an error of less than 5 % and computes the Cobb angle 3.28 times quicker. In the future, it is suggested to explore other techniques such as deep learning for alternative data analysis for Cobb angle estimation.

Bubble Plume Tracking Using a Backseat Driver on an Autonomous Underwater Vehicle

Autonomous underwater vehicles (AUVs) have been applied in various scientific missions including oceanographic research, bathymetry studies, sea mine detection, and marine pollution tracking. We have designed and field-tested in the ocean a backseat driver autonomous system for a 5.5 m survey-class Explorer AUV to detect and track a mixed-phase oil plume. While the first driver is responsible for controlling and safely operating the vehicle; the second driver processes real-time data surrounding the vehicle based on in situ sensor measurements and adaptively modifies the mission details. This adaptive sensing and tracking method uses the Gaussian blur and occupancy grid method. Using a large bubble plume as a proxy, our approach enables real-time adaptive modifications to the AUV’s mission details, and field tests show successful plume detection and tracking. Our results provide for remote detection of underwater oil plumes and enhanced autonomy with these large AUVs.

Development of Seeding Rate Monitoring System Applicable to a Mechanical Pot-Seeding Machine

In this study, we developed a monitoring system to accurately track the seeding rate and to identify the locations where the mechanical pot-seeding machine failed to sow seeds correctly. The monitoring system employs diverse image processing techniques, including the Hough transform, hue–saturation–value color space conversion, image morphology techniques, and Gaussian blur, to accurately pinpoint the seeding rate and the locations where seeds are missing. To determine the optimal operating conditions for the seeding rate monitoring system, a factorial experiment was conducted by varying the brightness and saturation values of the image data. When the derived optimal operating conditions were applied, the system consistently achieved a 100% seed recognition rate across various seeding conditions. The monitoring system developed in this study has the potential to significantly reduce the labor required for supplementary planting by enabling the real-time identification of locations where seeds were not sown during pot-seeding operations.

Threshold-Based BRISQUE-Assisted Deep Learning for Enhancing Crack Detection in Concrete Structures.

Automated visual inspection has made significant advancements in the detection of cracks on the surfaces of concrete structures. However, low-quality images significantly affect the classification performance of convolutional neural networks (CNNs). Therefore, it is essential to evaluate the suitability of image datasets used in deep learning models, like Visual Geometry Group 16 (VGG16), for accurate crack detection. This study explores the sensitivity of the BRISQUE method to different types of image degradations, such as Gaussian noise and Gaussian blur. By evaluating the performance of the VGG16 model on these degraded datasets with varying levels of noise and blur, a correlation is established between image degradation and BRISQUE scores. The results demonstrate that images with lower BRISQUE scores achieve higher accuracy, F1 score, and Matthew's correlation coefficient (MCC) in crack classification. The study proposes the implementation of a BRISQUE score threshold (BT) to optimise training and testing times, leading to reduced computational costs. These findings have significant implications for enhancing accuracy and reliability in automated visual inspection systems for crack detection and structural health monitoring (SHM).

A statistical deformation model-based data augmentation method for volumetric medical image segmentation

The accurate delineation of organs-at-risk (OARs) is a crucial step in treatment planning during radiotherapy, as it minimizes the potential adverse effects of radiation on surrounding healthy organs. However, manual contouring of OARs in computed tomography (CT) images is labor-intensive and susceptible to errors, particularly for low-contrast soft tissue. Deep learning-based artificial intelligence algorithms surpass traditional methods but require large datasets. Obtaining annotated medical images is both time-consuming and expensive, hindering the collection of extensive training sets. To enhance the performance of medical image segmentation, augmentation strategies such as rotation and Gaussian smoothing are employed during preprocessing. However, these conventional data augmentation techniques cannot generate more realistic deformations, limiting improvements in accuracy. To address this issue, this study introduces a statistical deformation model-based data augmentation method for volumetric medical image segmentation. By applying diverse and realistic data augmentation to CT images from a limited patient cohort, our method significantly improves the fully automated segmentation of OARs across various body parts. We evaluate our framework on three datasets containing tumor OARs from the head, neck, chest, and abdomen. Test results demonstrate that the proposed method achieves state-of-the-art performance in numerous OARs segmentation challenges. This innovative approach holds considerable potential as a powerful tool for various medical imaging-related sub-fields, effectively addressing the challenge of limited data access.

Noise insensitive volumetric fusion method for enhanced photoacoustic microscopy.

Photoacoustic imaging is an emerging imaging modality that combines the high contrast of optical imaging and the high penetration of acoustic imaging. However, the strong focusing of the laser beam in optical-resolution photoacoustic microscopy (OR-PAM) leads to a limited depth-of-field (DoF). Here, a volumetric photoacoustic information fusion method was proposed to achieve large volumetric photoacoustic imaging at low cost. First, the initial decision map was built through the focus detection based on the proposed three-dimensional Laplacian operator. Majority filter-based consistency verification and Gaussian filter-based map smoothing were then utilized to generate the final decision map for the construction of photoacoustic imaging with extended DoF. The performance of the proposed method was tested to show that our method can expand the limited DoF by a factor of 1.7 without the sacrifice of lateral resolution. Four sets of multi-focus vessel data at different noise levels were fused to verify the effectiveness and robustness of the proposed method. The proposed method can efficiently extend the DoF of OR-PAM under different noise levels.

An Efficient Approach of Assessing Quality of Blurred Image

This paper presents a method for evaluating the quality of images altered by Gaussian blur. The method is based on the observation of bokeh mode images where the region of interest (foreground) is sharp, while the remaining parts (background) are intentionally blurred to enhance the perceptual quality of the image. The blurriness of the background increases attention towards the foreground part of the image. The proposed quality metric is obtained by combining the attention factor and the sharpness of the region of interest. The accuracy, in terms of Spearman’s-rank-order correlation-coefficient (SROCC), for popular and publicly available databases such as LIVE, VCL, TID2008, CSIQ, and TID2013, is 0.963, 0.925, 0.900, 0.930, and 0.930, respectively. The proposed method achieves high and consistent Spearman’s rank-order correlation coefficient (SROCC) values compared to the majority of state-of-the-art algorithms. Furthermore, in terms of speed, the proposed method surpasses other state-of-the-art methods. The MATLAB code of the proposed metric is publicly available at https://drive.google.com/drive/folders/1SRmUp0N157Ati9l3kV13uoCxw5PhMgQn?usp=sharing .

Digital enhancement of OCT images toward better diagnostic of precursor cancer lesions in the uterine colon

We present a combination of image processing operations, which improve the quality of OCT images. The sequence is: (1) Gaussian blur, (2) Enhance Local Contrast (CLAHE), (3) FFT based band-pass filter, (4) Background Subtraction, (5) adjustment of the contrast of the image with light saturation (Enhance Contrast) and finally (6) reassignment of the palette such as to enhance the bright part of the images. The multi-step treatment reduces pixel-to-pixel noise, improves the homogeneity of the contrast, supresses vertical strips, reduces cosmetic distractions, equalizes contrast across a collection of images, and uses the pallet to emphasise areas of greater intensity. The enhanced images show a significant improvement in their quality, with details that do not appear in the original OCT image. The processing of these images and the comparison with untreated OCT images are shown throughout the paper.

Analysis of under-ice ambient noise characteristics of Gakkel Ridge in the Arctic.

This paper presents an analysis of the under-ice acoustic data and environmental parameters measured over a three-month period from August 31 to November 28, 2021, within the area of the Gakkel Ridge in the Arctic. After "spikes" caused by micro-level events are removed, the distribution of the retained under-ice noise related to macro-level events can be described satisfactorily by a Gaussian distribution, as verified by Q-Q plots and kurtosis/skewness analysis. We use sliding window analysis to deal with the features of under-ice ambient noise and model the data by Gaussian interpolation. This shows that the ambient noise level over the low-frequency range (10-100 Hz) is comparatively flat at about 60 dB; with the frequency increases from 100 to 2560 Hz, the ANL decreased to about 40 dB. We then introduce canonical correlation analysis (CCA) to analyze the potential relation between environmental forcing and the under-ice noise level. The results of CCA indicate that the seawater parameters (including temperature, salinity, and sound velocity) close to the ice-water interface have the greatest influence on the under-ice noise level among all environmental parameters recorded in the air, sea-ice, and seawater. Additionally, the under-ice noise level forced by the environment does not exhibit any particular frequency dependence.

Geospatial patterns of progress towards UNAIDS ‘95-95-95’ targets and community vulnerability in Zambia: insights from population-based HIV impact assessments

IntroductionIn sub-Saharan Africa, HIV/AIDS remains a leading cause of death. The UNAIDS established the ‘95-95-95’ targets to improve HIV care continuum outcomes. Using geospatial data from the Zambia Population-based HIV...

Handwritten Kannada Character Recognition Using Convolutional Neural Networks and Transfer Learning

Handwritten Kannada Character Recognition is special field of pattern recognition as different human being will have their own handwritten styles, machine considers even noise as a pattern. In order to classify each character, we present Convolutional Neural Networks and Transfer Learning techniques. Here we use 74k dataset and labelling individual character as one class totally we had made 25 classes. Initially we pre-process each image of character to remove noise, convert to gray-scale, applied gaussian blur and canny edge detection and cropped the images and at last resizing it. These methods help in training neural networks to recognize the characters with good accuracy. Then we apply transfer learning to same dataset, here we resize image into three-dimensional and convert images into gray-scale and using xception we train the model. Even though a few of the kannada language’s characters have similarities, this model can accurately categorise each character to the appropriate class and identify it with better accuracy than CNN.

Adaptive Optimized Discriminative Learning based Image Deblurring using Deep CNN

Image degradation plays a major problem in many image processing applications. Due to blurring, the quality of an image is degraded and there will be a reduction in bandwidth. Blur in an image is due to variations in atmospheric turbulence, focal length, camera settings, etc. Various types of blurs include Gaussian blur, Motion blur, Out-of-focus blur. The effect of noise along with blur further corrupts the captured image. Many techniques have evolved to deblur the degraded image. The leading approach to solve various degraded images are either based on discriminative learning models or on optimization models. Each method has its own advantages and disadvantages. Learning by discriminative methods is faster but restricted to a specific task whereas optimization models handle flexibly but consume more time. Integrating optimization models suitably by learning with discriminative manner results in effective image restoration. In this paper, a set of effective and fast Convolutional Neural Networks (CNNs) are employed to deblur the Gaussian, motion and out-of-focus blurred images that integrate with optimization models to further avoid noise effects. The proposed methods work more efficiently for applications with low-level vision.

Analysis and comparison of automatic image focusing algorithms in digital image processing

Optical imaging equipment is widely used in many fields such as industry, medicine and military, etc. The autofocus, as the main component affecting image clarity, still needs to be optimized and improved. A simpler passive autofocus system is selected, and a grayscale covariance matrix is introduced to use its output of inter-pixel grayscale relationship data to attenuate the influence of noise and realize the visual measurement of image quality. Next, the focus window locking is optimized, and a first-order moment window method incorporating image difference shadow is proposed, in which the Gaussian blur can enhance the smoothness of the image, and the difference shadow map is obtained according to the difference between the original image and the smoothed image, and the method keyed out the background of the image to enhance the detail portrayal of the target region, so as to locate the center of focus, and finally the window locking is realized in accordance with the proposed size. The study verifies the performance of the focusing model through simulation experiments, which show that the mean value of the accuracy of the model is the best value of 0.96 among the control models in the face of the Gaussian noise of (0.1,0.004). Therefore, this autofocus model can effectively improve the autofocus efficiency.

Hyperspectral image classification Based on weakened Laplacian pyramid and guided filtering

ABSTRACT In recent years, methods of spatial-spectral feature extraction have been widely used in hyperspectral image (HSI) classification and have achieved good performance. However, the ways are often unable to distinguish effectively the boundary of the adjacent ground objects. Multi-scale feature fusion alleviates the problems to a certain extent, but in the case of limited samples, the previous methods often fail to achieve the desired effect. In this paper, we propose a multi-scale hyperspectral image classification method based on weakened Laplacian pyramid and guided filtering to nicely distinguish the boundary of the different ground objects. This structure first obtains multi-scale information by constructing a weakened Laplacian pyramid, and weakens the magnification of Gaussian blur in the upsampling process. While obtaining multi-scale images, it can not only retain enough image information to expand feature samples but also emphasize edge information in the edge extraction of the Laplacian pyramid. Then, the weakened Laplacian pyramid feature map is used as the guiding image of the guiding filter, which can better preserve the edge features and further improve the classification performance of HSI. Extensive experiments are carried out on four HSI data sets. The results show that our method is superior in classification accuracy and computational cost.