Wavelet-based Image Segmentation Research Articles

DEEP NEURAL MULTI WAVELET SEGMENTATION AND RESNET-50 IMAGE CLASSIFICATION (DNMWS-RESNET-50IC) BASED EARLY GLAUCOMA DETECTION

Glaucoma, a leading cause of irreversible blindness worldwide, results from progressive damage to the optic nerve, often linked to elevated Intra Ocular Pressure (IOP). Early detection is critical for preventing vision loss, yet traditional diagnostic methods can be limited in accessibility and effectiveness, particularly in the early stages of the disease. Addressing the need for early detection, we propose the Deep Neural Multi-Wavelet Segmentation and ResNet-50 Image Classification (DNMWS-ResNet-50IC) method, specifically designed to detect glaucoma at an early stage using retinal fundus images. In this work, we apply Anisotropic Gaussian Filtering for the preprocessing of retinal fundus images, effectively reducing image noise while preserving the original quality of the image pixels by creating an adaptive window size and scale space. We then utilize multi- wavelet-based image segmentation, leveraging wavelet transforms to analyze and decompose the image into its various frequency components. This technique is particularly advantageous for managing images with complex structures and textures. Subsequently, the segmented features are classified using the ResNet-50 model, which categorizes the images as normal, abnormal, or indicative of early- stage glaucoma. The effectiveness of the proposed method is assessed by measuring three key performance indicators sensitivity, specificity, and accuracy on digital retinal images from the HRF image database. Additionally, the model's performance is further evaluated on a separate test set, considering metrics such as accuracy, precision, recall, and prediction time.

Detection of Diabetic Retinopathy from Retinal Fundus Image using Wavelet based Image Segmentation

Detection of Diabetic Retinopathy from Retinal Fundus Image using Wavelet based Image Segmentation

Kinematics of the jet in M 87 on scales of 100–1000 Schwarzschild radii

Very long baseline interferometry (VLBI) imaging of radio emission from extragalactic jets provides a unique probe of physical mechanisms governing the launching, acceleration, and collimation of relativistic outflows. The two-dimensional structure and kinematics of the jet in M\,87 (NGC\,4486) have been studied by applying the Wavelet-based Image Segmentation and Evaluation (WISE) method to 11 images obtained from multi-epoch Very Long Baseline Array (VLBA) observations made in January-August 2007 at 43 GHz ($\lambda = 7$ mm). The WISE analysis recovers a detailed two-dimensional velocity field in the jet in M\,87 at sub-parsec scales. The observed evolution of the flow velocity with distance from the jet base can be explained in the framework of MHD jet acceleration and Poynting flux conversion. A linear acceleration regime is observed up to $z_{obs} \sim 2$\,mas. The acceleration is reduced at larger scales, which is consistent with saturation of Poynting flux conversion. Stacked cross correlation analysis of the images reveals a pronounced stratification of the flow. The flow consists of a slow, mildly relativistic layer (moving at $\beta \sim 0.5\,c$), associated either with instability pattern speed or an outer wind, and a fast, accelerating stream line (with $\beta \sim 0.92$, corresponding to a bulk Lorentz factor $\gamma \sim 2.5$). A systematic difference of the apparent speeds in the northern and southern limbs of the jet is detected, providing evidence for jet rotation. The angular velocity of the magnetic field line associated with this rotation suggests that the jet in M87 is launched in the inner part of the disk, at a distance $r_0 \sim 5\, R_\mathrm{s}$ from the central engine. The combined results of the analysis imply that MHD acceleration and conversion of Poynting flux to kinetic energy play the dominant roles in collimation and acceleration of the flow in M\,87.

Detection of multiple velocity components in partially overlapping emitting regions

Velocity measurements made from multiple-epoch astronomical images of evolving objects with optically thin continuum emission (e.g. as relativistic jets or expanding supernova shells) may be confused as a result of the overlap of semi-transparent features moving at different speeds. Multi-scale wavelet decomposition can be effectively applied to identify and track such overlapping features, provided that their respective structural responses can be separated over the spatial scales used for the decomposition. We developed a new method that combines the stacked cross-correlation with the wavelet-based image segmentation and evaluation (WISE) technique of decomposition of two-dimensional structures, to separate and track dominant spatial responses of overlapping evolving features. The method is tested on a set of simulated images of a stratified relativistic jet, demonstrating the robust detection of both the faster spine and the slower sheath speeds. The method is applied to mutliple-epoch images from the MOJAVE survey, revealing two different superluminal streams inside the jet in 3C 273 and the acceleration of the flow in 3C 120. The method can be applied to densely monitored objects with composite structural evolution such as the parsec-scale jet in M87 or heavily resolved expanding supernova shells.

Wavelet-based decomposition and analysis of structural patterns in astronomical images

Context. Images of spatially resolved astrophysical objects contain a wealth of morphological and dynamical information, and effective extraction of this information is of paramount importance for understanding the physics and evolution of these objects. Algorithms and methods employed presently for this purpose (such as, for instance, Gaussian model fitting) often use simplified approaches for describing the structure of resolved objects. Aims. Automated (unsupervised) methods for structure decomposition and tracking of structural patterns are needed for this purpose, in order to be able to deal with the complexity of structure and large amount of data involved. Methods. A new Wavelet-based Image Segmentation and Evaluation (WISE) method is developed for multiscale decomposition, segmentation, and tracking of structural patterns in astronomical images. Results. The method is tested against simulated images of relativistic jets and applied to data from long-term monitoring of parsec- scale radio jets in 3C 273 and 3C 120. Working at its coarsest resolution, WISE reproduces exceptionally well the previous results of model fitting evaluation of the structure and kinematics in these jets. Extending the WISE structure analysis to the finer scales provides the first robust measurements of two-dimensional velocity fields in these jets and indicates that the velocity fields are likely to reflect the evolution of Kelvin-Hemlholtz instability developing the flow.

MR image-guided portal verification for brain treatment field

Purpose : To investigate a method for the generation of digitally reconstruction radiographs directly from MR images (DRR-MRI) to guide a computerized portal verification procedure. Methods and Materials : Several major steps were developed to perform an MR image-guided portal verification procedure. Initially, a wavelet-based multiresolution adaptive thresholding method was used to segment the skin slice-by-slice in MR brain axial images. Some selected anatomical structures, such as target volume and critical orgsns, were then manually identified and were reassigned to relatively higher intensities. Interslice information was interpolated with a directional method to achieve comparable display resolution in three dimensions. Next, a ray-tracing method was used to generate a DRR-MRI image at the planned treatment position, and the ray tracing was simply performed on summation of voxels along the ray. The skin and its relative positions were also projected to the DRR-MRI and were used to guide the search of similar features in the portal image. A Canny edge detector was used to enhanced the brain contour in both portal and simulation images. The skin in the brain portal image was then extracted using a knowledge-based searching technique. Finally, a Chamfer matching technique was used to correlate features between DRR-MRi and portal image. Results : The MIR image-guided portal verification method was evaluated using a brain phantom case and a clinical patient case. Both DRR-CT and DRR-MRI were generated using CT and MR phantom images with th same beam orientation and then compared. The matching results indicated that the maximum deviation of internal structures was less than 1 mm. The segmented results for brain MR slice images indicated that a wavelet-based image segmentation technique provided a reasonable estimation for the brain skin. For the clinical patient case with a given portal field, the MR image-guided verification method provided an excellent match between features in both DRR-MRI and portal image. Moreover, target volume could be accurately visualized in the DRR-MRI and mapped over the corresponding portal image for treatment verification. The accuracy of DRR-MRI was also examined by comparing it to the corresponding simulation image. The matching results indicated that the maximum deviation of anatomical features was less than 2.5 mm. Conclusion : A method for MR image-guided portal verification of brain treatment field was developed. Although The radiographic appearance in the DRR-MRI is different from that in the portal image, DRR-MRI provides essential anatomical features (landmarks and target volume) as well as their relative locations to be used as references for computerized portal verification.