Browsing by Author "Kisangiri, Michael"

Now showing 1 - 5 of 5
  • Thumbnail Image
    Item
    Diffusion-steered denoising framework for suppressing multiplicative noise in ultrasonograms
    (African Journal of Applied Research, 2017-10-07) Kessy, Suzan; Msuya, Hubert; Kisangiri, Michael; Maiseli, Baraka J.
    Ultrasound imaging, a non-invasive and cost-effective imaging modality, is probably the most preferred diagnostic tool in medicine. Despite its merits, ultrasonograms are usually corrupted by multiplicative noise, a consequence that limits doctors to provide more accurate treatments and decisions. Attempts to address the problem have been made, but we have found little works that adopt the diffusion framework, which scholars have reported that it produces promising results in additive noise cases. In the current work, we have modified the classical Perona-Malik (PM) diffusion model to deal with multiplicative noise. Inspired by the ability of PM to restore semantically critical features, we have embedded a log-based regularization term, statistically modeled to mitigate multiplicative effects in the ultrasound images, into the modified PM. Additionally, the diffusivity kernel of PM has been re-designed to ensure that the diffusion process is properly steered. Modification of the PM kernel was achieved through integration of the half-quadratic diffusivity, which has a corresponding energy functional that is strictly convex, a promising mathematical property that encourages unique solutions and guarantees stability of the evolutionary system. Our interest is to emphasize regularization in flat image regions while maintaining plausible edges and contours. Subjective and quantitative evaluations demonstrate that the proposed model produces better results compared with some state-of-the-art methods. Even more importantly, our approach guarantees convergence, stability, and robustness when tested for a range of ultrasonograms. Probably the intriguing property of our framework is its ability to evolve a denoising image over a longer period without smudging or destroying its sensitive features. The proposed approach may further be extended and actualized in practical imaging devices.
  • Thumbnail Image
    Item
    A Grey Level Fitting Mechanism based on Gompertz Function for Two Phase Flow Imaging using Electrical Capacitance Tomography Measurement Systems
    (2014) Nombo, Josiah; Mwambela, Alfred; Kisangiri, Michael
    Electrical Capacitance Tomography (ECT) is an image generating system based on soft field sensory system. The preferred Linear Back Projection (LPB) reconstruction algorithm for multi-phase measurement has blurring effect on the image generated. These two inherent factors, among others, affect the quality of image generated from ECT systems. Introducing fitting in the image generation process is one the solutions to improving its quality. In this article an alternative fitting mechanism based on the Gompertz function has been developed and evaluated. Comparative analysis results shows improvement on the spatial quality of images generated, in terms of minimum relative image and distribution errors, maximum correlation coefficient, and at relatively no additional computational cost. The mechanism is more effective for annular than stratified flow data hence complimenting the weakness of Xie method for annular flow.
  • Thumbnail Image
    Item
    Performance Analysis of Grey Level Fitting Mechanism based Gompertz Function for Image Reconstruction Algorithms in Electrical Capacitance Tomography Measurement System
    (2015) Nombo, Josiah; Mwambela, Alfred; Kisangiri, Michael
    This paper analyses the performance of grey level fitting mechanism based on Gompertz function used in Electrical Capacitance Tomography measurement system. In order to evaluate its performance, the data fitting mechanism has been applied to common image reconstruction algorithms which include; Linear Back Projection, Singular Value Decomposition, Tikhonov Regularization, Iterative Tikhonov Regularization, Landweber iteration and Projected Landweber iteration. Images were reconstructed using measured capacitance data for annular and stratified flows, and qualitative and quantitative evaluation were done on the reconstructed images in comparison with respective reference images. Results show that this grey level fitting mechanism is better in terms of improving image spatial resolution, minimizing relative image error and distribution error and maximizing correlation coefficient.
  • Thumbnail Image
    Item
    Perona–Malik model with self-adjusting shape-defining constant
    (Information Processing Letters, 2018-09-01) Maiseli, Baraka; Msuya, Hubert; Kessy, Suzan; Kisangiri, Michael
    For decades, the Perona–Malik (PM) diffusion model has been receiving a considerable attention of scholars for its ability to restore detailed scenes. The model, despite its promising results, demands manual tuning of the shape-defining constant—a process that consumes time, prompts for human intervention, and limits flexibility of the model in real-time systems. Most works have tried to address other weaknesses of the PM model (non-convexity and non-monotonicity, which produce chances for instability and multiple solutions), but automating PM remains an open-ended question. In this work, we have introduced a new implementation approach that fully automates the PM model. In particular, the tuning parameters have been conditioned to ensure that the model guarantees convergence and is entirely convex over the scale-space domain. Experiments show that our implementation strategy is flexible, automatic, and achieves convincing results.
  • Thumbnail Image
    Item
    Vehicle Plate Number Detection and Recognition Using Improved Algorithm
    (IISTE, 2014) Munuo, Cosmo H.; Kisangiri, Michael; Mvungi, Nerey H.
    The growing Tanzanian population currently estimated to be 48 Million people and their use of vehicles as means of transport has kept increasing making enforcing traffic rules and regulations among road users a major challenge. This calls for a need to have an automated system that monitors the motorists with a pre-defined sense of intelligence. A Vehicle Detection and Recognition Algorithm which can provide automated access to relevant information to a number plate from information systems containing and managing databases on vehicle and their movements is required. This paper presents work on developed algorithm that localizes plate area, extract and segment character, and finally recognizes and interprets registration number from vehicle image. MATLAB R2012b Simulation software with Image Processing toolbox is employed. HSV color space image, morphological and statistical analysis operations were integrated and employed to a vehicle image to compute plate number area. In segmentation the properties like aspect ratio, extent, and area ratio were important measurement parameters. Finally, the template matching database and statistical character extracted from car image was correlated to recognize alphanumeric character to deduce car registration number.