Yazar "Ghoushchi, Saeid Jafarzadeh" seçeneğine göre listele

Listeleniyor 1 - 2 / 2
  • Küçük Resim Yok
    Öğe
    ETACM: an encoded-texture active contour model for image segmentation with fuzzy boundaries
    (Springer, 2023) Ranjbarzadeh, Ramin; Sadeghi, Soroush; Fadaeian, Aida; Ghoushchi, Saeid Jafarzadeh; Tirkolaee, Erfan Babaee; Caputo, Annalina; Bendechache, Malika
    Active contour models (ACMs) have been widely used in image segmentation to segment objects. However, when it comes to segmenting images with severe intensity inhomogeneity, most current frameworks do not perform well, which can make it difficult to achieve the desired results. To address this issue, a decision-making model is proposed, which involves using enhanced local direction pattern (ELDP) and local directional number pattern (LDNP) texture descriptors to create an encoded-texture ACM. The principal component analysis (PCA) is then used to optimize the two encoded images and reduce the correlations before they are fused. To further improve the performance of the encoded-texture ACM, a function of minimizing energy globally (FMEG) is suggested by applying the vector-valued exploration technique from a non-convex surface to region-based ACMs. This approach enables the development of a model capable of directly building complex decision boundaries. The experimental results show that the proposed encoded-texture ACM outperforms many recent frameworks in terms of robustness and accuracy for segmenting images with intensity inhomogeneity, fuzzy boundaries, and noise. Therefore, the suggested approach provides a more effective and efficient solution to the problem of image segmentation, particularly for challenging images.
  • Küçük Resim
    Öğe
    An extended approach to the diagnosis of tumour location in breast cancer using deep learning
    (SPRINGER HEIDELBERG, 2021) Ghoushchi, Saeid Jafarzadeh; Ranjbarzadeh, Ramin; Najafabadi, Saeed Aghasoleimani; Osgooei, Elnaz; Tirkolaee, Erfan Babaee
    Breast cancer is one of the most common and deadly cancers in women. However, early detection increases the likelihood of survival by 100%. Radiologists use mammograms to take X-ray images of the breast to look for signs of tumour formation, such as breast masses. The purpose of detecting these signs using convolutional networks is a modern machine learning (ML) model that performs image segmentation in one learning step. Therefore, this study develops a new machine learning approach based on modified deep learning (DL) to diagnose the tumour location in breast cancer. In this study, the data obtained from the databases (BCDRD01) are developed and resized and divided into data sets. A simple architecture is used for the first group of experiments, one of which utilizes a weighted function to counter the class imbalance. At first, after visualizing the images and using the Gabor filter, the exact location of the breast tissue is determined. In the following, two other complicated network-based architectures (VGG) (9 layers and 2.9 million parameters) and remaining networks (10 layers and 0.9 million parameters) are employed for the following experiments. The results indicate that convolutional neural networks (CNNs) are an appropriate option for the separation of breast cancer lesions.