Yazar "Nahar, Ali" seçeneğine göre listele

Listeleniyor 1 - 2 / 2
  • [ X ]
    Öğe
    Magnetic resonance imaging (MRI) for brain tumor and seizures classification using recurrent neural network
    (Institute of Electrical and Electronics Engineers Inc., 2020) Aljaleeli, Marwah; Nahar, Ali; Mahmood, Mohammed; Bayat, Oğuz
    The aim of this paper is to use develop and evaluate a Magnetic Resonance Imaging (MRI) for brain tumor and seizures classification using Recurrent Neural Network (RNN). Medical Science in Image Processing is an emerging field which has proposed a lot of advanced techniques in detection and analysis of a particular disease. Treatment of brain tumors in recent years is getting more and more challenging due to complex structure, shape and texture of the tumor. Therefore, by advancing in image processing, various methodologies have been proposed to identify the tumors in the brain. The advancement in this field created an urge in me to research more on the techniques and methodologies developed for tumor extraction. Hence, we propose a system to extract Tumor from the brain using MRI images. This technique involves different image processing methodologies such as noise removal, filtering, segmentation and morphological operations. Extraction of Brain tumor can be accomplished successfully by performing these operations on MATLAB software. Cross-correlation is computed between the target variable vector and the tumor region to determine how pixels' values of the tumor region are closely related using image processing and RNN technique achieving an accuracy of 99.71% © 2020 IEEE.
  • Küçük Resim
    Öğe
    Polygon number algorithm for peak-to-average ratio reduction of massive 5G systems using modified partial transmit sequence scheme
    (Wiley, 2021) Alkatrani, Hayder; İlyas, Muhammad; Alyassri, Salam; Nahar, Ali; Al-Turjman, Fadi; Rasheed, Jawad; Alshahrani, Ali; Al-Kasasbeh, Basil
    The high peak-to-average power ratio (PAPR) of the transmit signal is a major shortcoming of OFDM systems, which results in band radiation and distortion due to the nonlinearity of the high-power amplifier (H.P). To resolve the traditional OFDM highPAPR issue, where the transmit sequence is designed to avoid similar data from being sent in the same order to reduce PAPR, there are numerous conventional ways for lowering the PAPR for OFDM system, such as selective mapping, tone reservation, block coding, filtering, clipping, and partial transmit sequence (PTS). This study proposes a new method called polygon number algorithm (PN) with conventional partial transmit sequence (C-PTS). This method (PN-PTS) processes the entered data before sending it, taking advantage of the number nonsimilarity according to the geometry of the number to prevent direct sending of similar data via PTS, and thus, this improved the level of PAPR rise in the proposed system. The amount of reduction that can be achieved in PAPR is up to 8 dB by different techniques. The best result obtained was the amount of reduction between the conventional method and the proposed method is 4.5683 where N = 64. Besides this, there is no transmission of side information (SI), which improves transmission efficiency. Finally, this method is easy in the calculation process and the ciphering and deciphering of data, which adds a few calculations.