1,720,998 research outputs found
Speckle suppression in SAR images employing modified anisotropic diffusion filtering in wavelet domain for environment monitoring
No full textSynthetic Aperture Radar (SAR) is a tool of coherent imagery utilized for meteorological and astronomical purposes. But, these images are contaminated with speckle noise which degrades the image quality and automatic information extraction becomes difficult. This paper presents an improved filtering technique which combines the Wavelets and proposed Anisotropic Diffusion (AD) filter for despeckling SAR images. The speckled image is initially decomposed into sub-bands using 2D-Discrete Wavelet Transform (2D-DWT) followed by application of modified AD filter. The diffusion coefficient presented in this
modified AD filter consists of a combination of gradient and Laplacian operators. The spatial variation of this diffusion coefficient occurs in such a way that it prefers forward diffusion to backward diffusion resulting in effective reconstruction of structural content and detection of weak edges. The filtered sub-bands are then reconstructed after soft thresholding. Based on the simulation results as well as the values of image quality metrics; filtered SAR images obtained by the proposed speckle suppression methodology can be claimed better in comparison to other recent works
Bilateral despeckling filter in homogeneity domain for breast ultrasound images
No full textBreast sonograms are more effective towards differentiation of cysts from solid tumours; if they could be post-processed for minimization of speckle content without blurring of edges. The approach presented in this paper consists of a bilateral filtering in homogeneity domain so that the despeckling process do not compromises the texture and features of masses. The proposed despeckled approach decomposes the input image into homogeneous and non-homogeneous regions; which are then selectively processed using the bilateral filter. The domain filtering component is made dominant when applied to homogeneous pixels providing smoothening while the range filter dominates on the non-homogeneous pixels leading to edge preservation. Simulations carried out on breast ultrasound images depict satisfactory speckle filtering supported with improvement in values of performance parameters (PSNR, SSIM & SSI)
A Reduced Reference Distortion Measure for Performance Improvement of Smart Cameras
No full textApplication specific information processing (ASIP)
unit in smart cameras requires sophisticated image processing
algorithms for image quality improvement and extraction of
relevant features for image understanding and machine vision.
The improvement in performance as well as robustness can be
achieved by intelligent moderation of the parameters both at algorithm
(image resolution, contrast, compression, and so on) as well
as hardware levels (camera orientation, field of view, and so on).
This paper discusses the employment of ISO/IEC/IEEE 21451
smart transducer standards for performance improvement of
smart cameras. The standardized transducer electronic data
sheets (TEDS-by IEEE 21450) provide the self description of
sensors, of which the calibration details are of vital importance to
yield a smart and reconfigurable imaging system. This is possible
by exercising intelligent control over the TEDS (smart camera)
calibration details as well as automated tuning of algorithm
parameters (in ASIP) based on decisions by perceptually efficient
image quality assessment (IQA) tool. Estimation of distortion
based on reduced reference IQA has been highlighted as a reliable
methodology for this purpose. The proposed IQA approach
uses wavelets for features extraction followed by estimation of
luminance, contrast, and divergence parameters to obtain the
proposed distortion measure (Q). The computational complexity
in the process has been catalyzed using integral image and
gradient magnitude approaches. The validation of Q metric
is carried out by evaluating the image quality for various
types of distortions on images from Content-based Strategies of
Image Quality assessment (CSIQ) and Information Visualization
CyberInfrastructure (IVC) databases. Simulation results yield a
healthy correlation of Q and the subjective human opinions
EEG Signal Processing and Acquisition for Detecting Abnormalities via Bio-implantable Devices
No full textThe proposed research illustrates an innovating implantable micro-apparatus to be encompassed under the scalp for
monitoring and retrieving electrical cerebral activities. The illustrated system considers its theoretical realization including, design of
circuital electronic components and energy harvesting, 3D package, chemical aspects concerning the utilization of UHMWPE (Ultra
High Molecular Weight Polyethylene) polymeric materials for packaging including mechanical simulations and comparison with
titanium material, and electromagnetic aspects regarding the Wi-Fi radiation. A full description of necessary circuitry is included. Moreover, for chemical viewpoint, requirements of polymeric nanomaterials, embedding silver or copper nanoparticles to be used for its fabrication, are discussed illustrating antibacterial and electromagnetic wall barrier properties. The study of the proposed work concerns the whole design of the system
Multimodal Medical Image Sensor Fusion Framework Using Cascade of Wavelet and Contourlet Transform Domains
No full textMultimodal medical image fusion is effectuated to minimize the redundancy while augmenting the necessary information from the input images acquired using different medical imaging sensors. The sole aim is to yield a single fused image, which could be more informative for an efficient clinical analysis.
This paper presents a two-stage multimodal fusion framework using the cascaded combination of stationary wavelet transform (SWT) and non sub-sampled Contourlet transform (NSCT) domains for images acquired using two distinct medical imaging sensor modalities (i.e., magnetic resonance imaging and computed
tomography scan). The major advantage of using a cascaded combination of SWT and NSCT is to improve upon the shift variance, directionality, and phase information in the finally fused image. The first stage employs a principal component analysis algorithm in SWT domain to minimize the redundancy.
Maximum fusion rule is then applied in NSCT domain at second stage to enhance the contrast of the diagnostic features. A quantitative analysis of fused images is carried out using dedicated fusion metrics. The fusion responses of the proposed approach are also compared with other state-of-the-art fusion
approaches; depicting the superiority of the obtained fusion results
Multispectral medical image fusion in Contourlet domain for computer based diagnosis of Alzheimer's disease
No full textComputer based diagnosis of Alzheimer's disease can be performed by dint of the analysis of the functional and structural changes in the brain. Multispectral image fusion deliberates upon fusion of the complementary information while discarding the surplus information to achieve a solitary image which encloses both spatial and spectral details. This paper presents a Non-Sub-sampled Contourlet Transform (NSCT) based multispectral image fusion model for computer-aided diagnosis of Alzheimer's disease. The proposed fusion methodology involves color transformation of the input multispectral image. The multispectral image in YIQ color space is decomposed using NSCT followed by dimensionality reduction using modified Principal Component Analysis algorithm on the low frequency coefficients. Further, the high frequency coefficients are enhanced using non-linear enhancement function. Two different fusion rules are then applied to the low-pass and high-pass sub-bands: Phase congruency is applied to low frequency coefficients and a combination of directive contrast and normalized Shannon entropy is applied to high frequency coefficients. The superiority of the fusion response is depicted by the comparisons made with the other state-of-the-art fusion approaches (in terms of various fusion metrics)
Exploring the Diversity of Evolved Cognitive Models with Cluster Analysis
No full textCognitive scientists often represent theories of cognitive behavior in the form of computer programs which simulate and model the performance of humans in experimental settings. Earlier work has demonstrated that evolutionary techniques, specifically genetic programming (GP), can be used to generate a pool of candidate models in the form of executable computer programs. However, previous work has not considered the impact of changes to hyper-parameter values, such as those controlling the behavior and timing of operators or those controlling the operation of the GP process. In this paper, we develop and use a cluster analysis technique based around the Silhouette index to investigate the impact of hyper-parameter changes on the composition of evolved populations of programs. Our metrics support visualizations and enable a user to assess both qualitatively and quantitatively the diversity of candidates from different populations. In this way, a cognitive scientist can analyze the output of the evolutionary system in order to uncover or inspire potentially novel theories of human behavior
Artificial Intelligence Supported Site Mapping for Building Pop-Up Habitats
No full textBuilding pop-up habitats in extreme weather conditions such as deserts requires preliminary contextual, i.e., site studies. Since the site’s condition is constantly changing due to sand relocation induced by wind, a rapid mapping solution is proposed. This is implemented by generating a 3D mesh model of the site with the help of a visual workflow and advanced computational design methods to implement in-situ 3D printing of habitats. This paper presents an integrated approach utilizing Computer Vision (CV), Deep Learning (DL), and generative design tools like Grasshopper. By harnessing the potential of Convolutional Neural Networks (CNNs), a robust framework is developed to recognize complex desert terrain features, independent of solar orientation and camera positioning. The methodology employs a state-of-the-art CNN customized for detecting features in desert settings. This is further enhanced by using Grasshopper to systematically generate a diverse dataset that enriches the model’s learning process. The resulting model efficiently extracts precise 3D meshes from 2D images, optimizing site mapping and integrating habitat printing workflows. This automated approach offers an effective solution for habitat construction in challenging environments, showcasing real-time processing.</p
An IoT-Based Health Monitoring System for Elderly Patients
No full textInternet of Things in health care significantly improves the requirement of medical facilities to patients and assists healthcare professionals. Since many elderly people require to be checked up on frequently due to high risk of life-threatening diseases, a health monitoring system using a cloud platform becomes effective in real-time observation. This paper focuses on a remote health monitoring system that tracks and updates the patient’s caretaker with the data collected from the patient. This enables the patient’s overseer to analyze the data being fed to them and subsequently provide treatment for the patient. Any sudden changes in the patient’s vitals will ensure that quick assistance be provided to the patient. The data stored in the cloud platform allows the process rate to be even quicker, thus reducing the time required to go to a hospital or waiting for treatment.</p
A Hybrid CNN-LSTM Deep Learning Framework for Multi-DOF Control of Upper-Limb Prosthetic Using EMG Signals
No full textLosing a limb is a life-altering event that significantly affects an individual’s independence and quality of life. Among prosthetic advancements, upper-limb prosthetics have gained greater attention due to their essential role in restoring functionality and autonomy. However, current upper-limb prosthetics are constrained by two primary limitations: restricted ranges of movement and limited simultaneous control. In this paper, we propose a modified CNN-LSTM hybrid architecture designed for continuous computation of four degrees of freedom (DOF), enabling control of the elbow angle (θ), the horizontal (X) and vertical (Y) positions of the wrist joint, and velocity (v) of arm movement. We specifically examine the effect of incorporating historical timesteps on enhancing the decoding performance of these parameters. Our results demonstrate significant enhancements in decoding accuracy when historical timesteps are incorporated surpassing state-of-the-art methods applied on the same dataset and other studies utilizing similar hybrid approaches on different datasets.</p
- …
