No abstract

With the technology of free space optical communication, information can be transmitted from the transmitter to receiver wirelessly without the necessity of fiber optic cables. This technology offers system security, extended bandwidth, high data rate, and simple installation. This work aims to improve the optical channel based on the optimization of different optical amplifiers and filters. Performance analysis is carried out using a rectangular optical filter (ROF) and two electrical amplifiers named automatic gain control (AGC) and transimpedance amplifier (TIA). The results are presented in terms of maximum quality factor as a function of link range. The proposed systems (represented by ROF and AGC) brought better performance than traditional one (represented by TIA) via the same link range and data rates used. The findings displayed the progress of the AGC which has better quality factor than TIA and ROF. For instance at 5 m length, the AGC achieves a maximum Q-factor of 12.29, while the ROF and ATI reveal a Q-factor in the range of 9.8 and 7.01, respectively.

No abstract

Tumor segmentation is one of the most significant tasks in brain image analysis due to the significant information obtained by the tumor region. Therefore, many methods have been proposed during the last two decades for segmenting the tumor in MRI images. In this paper, an automated method is proposed using an active contour model with an initial contour creation using edge sharpening, thresholding, and morphological operations. Four methods of edge detection are utilized in the edge sharpening process (Sobel, Roberts, Prewitt, and Canny) and their performance was investigated in terms of Dice, Jaccard, and F1 score. The experiments were implemented on BRATS datasets with both HGG and LGG images. The study indicates that sharpening the edges using edge detection is essential to improve the segmentation of the tumor region especially when it is used with an active contour model. The achieved results show the effectiveness of the proposed method and it outperformed some recent existing methods.

No abstract