Repository logo
 

Faculty of Engineering and Built Environment

Permanent URI for this communityhttp://ir-dev.dut.ac.za/handle/10321/9

Browse

Subject: search.filters.subject.Image processing

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Comparative study of binary classifiers for reducing false negative detection of melanoma in skin lesions
    (IEEE, 2022-10-27) Jooravan, Amith; Reddy, Serendra; Pillay, Nelendran
    Reliable and accurate classification of a skin lesion is essential to the early diagnosis of skin cancer, especially melanoma. Traditional classification methods require performing a biopsy on the lesion. The overlap of benign and malignant clinical features may lead to incorrect melanoma diagnosis and/or excising an excessive number of benign lesions. This paper focuses on the use of machine learning to aid physicians with the non-invasive classification methodology of skin lesions, whilst prioritising the minimization of false negative classification. The clinical features used are based on the ABCD rule, representing the asymmetry, border, colour and diameter of the lesion. The dermoscopic images chosen are of melanoma lesions less than 0,76mm in thickness which corresponds to the early stages of cancer. The investigated classification methods include K-Nearest neighbours (KNN), Naïve Bayes and linear support vector machine. (LSVM). This research proposes the use of a LSVM machine learning algorithm to classify a skin lesion as being either melanoma or non-melanoma with the lowest false negative rate of the investigated classification. Classification accuracy of 85% and a false negative rate of 5% is achieved.
  • Thumbnail Image
    Item
    Learning rate optimisation of an image processing deep convolutional neural network
    (2021) Buthelezi, Sibusiso Blessing; Reddy, Seren; Twala, Bhekisipho
    The major contribution of this dissertation is the proposal of the use of mathematical models to identify an optimal learning rate for an image processing deep convolutional neural network (DCNN). This model is derived from a nonlinear regression relationship between the learning rate and the accuracy of a test DCNN model. This relationship is meant to (A) resolve the problem of arbitrarily selecting the initial learning rate (B) reduce computational resource requirement and (C) reduce training instabilities. An algorithm is developed to analyse an inputted DCNN model and subsequently render output parameters that may be used to aid in the selection of an OLR. The benefit of an OLR includes improved training stability and reduced computational resources. The results rendered by the OLR algorithm proposes that an optimal learning rate improves model performance; this is described by the test model average accuracy of 91%. Furthermore, a model validation graph is also extrapolated. which will illustrate the mathematical model accuracy and the region of interest (ROI). The ROI defines the region in the learning rate spectrum with a positive effect on model performance.