No abstract

A novel double‐layer chessboard‐like configuration is proposed to reduce radar cross‐section (RCS) in microwave and mm‐Wave bands. This structure is composed of two artificial magnetic conductor cells. One of these cells consists of four square patches on substrate and the other one is formed only by substrate without any copper on it. Despite the simple design of this structure, it has a big influence on significantly expanding the bandwidth of RCS reduction (RCSR). In comparison with the other works, the measured bandwidth of 10 dB RCSR is broadened to 113.33% (16.43–59.4 GHz). This wideband behaviour is achieved by effectively controlling the ratio between two dielectric constants and the ratio between their thicknesses and so the effective dielectric constant which adjusts the positions of all resonances. Although this structure contains two layers, its overall thickness is 1.4 mm, which is less than the ones that the recent works have. Measurements and simulations are performed for both monostatic and bistatic RCSs with the best number of the unit cells in the chessboard‐like structure. The design, simulations, semi‐analytical expressions and fabrication are presented and the measurements show an excellent agreement with the simulations.

A novel wideband three‐layer chessboard‐like structure is proposed to reduce the radar cross section (RCS) of the radar target. This configuration is composed of two artificial magnetic conductor (AMC) cells formed by two crossed ellipses with different sizes in two cells. A desired 180° ± 37° phase difference is achieved by combining these unit cells and the measured 10 dB RCS‐reduction bandwidth is extensively broadened to more than 96% (from 8.11 to 23.32 GHz, covering X, Ku, and K bands for different radars) in comparison with the other works. This characteristic is obtained by carefully adjusting the positions of all resonances using the proper sizes for the ellipses and the proper dielectric constants and thicknesses for the three layers. Although, the proposed design has three layers with the overall thickness of 2 mm, it is still thinner than most of the recent related works. This low‐profile structure is also cost‐effective due to the fact that 60% of the overall thickness is formed by an air substrate. The proposed cells are designed, simulated, and fabricated in a chessboard‐like configuration for both monostatic and bistatic RCSs. Simulations and measurements are in a good agreement, which shows the capabilities of the design.

Cataract is one of the most common eye disorders that causes vision distortion. Accurate and timely detection of cataracts is the best way to control the risk and avoid blindness. Recently, artificial intelligence-based cataract detection systems have been received research attention. In this paper, a novel deep neural network, namely CataractNet, is proposed for automatic cataract detection in fundus images. The loss and activation functions are tuned to train the network with small kernels, fewer training parameters, and layers. Thus, the computational cost and average running time of CataractNet are significantly reduced compared to other pre-trained Convolutional Neural Network (CNN) models. The proposed network is optimized with the Adam optimizer. A total of 1130 cataract and non-cataract fundus images were collected and augmented to 4746 images to train the model. For avoiding the over-fitting problem, the dataset is extended through augmentation before model training. Experimental results prove that the proposed method outperforms the state-of-the-art cataract detection approaches with an average accuracy of 99.13%.

No abstract