Yapay Zeka Hızlı Bakış & Çıkarımlar
AI InsightThe results obtained in this study have shown that the modified ANFIS algorithm is the convenient choice in terms of speed, accuracy.
📌 3 Maddede Bu Makale
Özet
Applications related with WSNs may include thousands of separate sensor nodes, production and control data for different industrial sectors. It is important to manage these applications, monitor the network and reprogram the nodes to avoid operational problems. In this study, we propose a smart wireless sensor network using a reconfigurable embedded system of Field-Programmable Gate Arrays (FPGAs) with a soft-core processor. This processor can be programmed dynamically and synthesized to implement the preprocessing of sensed data by ensemble Hybrid Neuro-Fuzzy algorithms such as Adaptive Neuro-Fuzzy Inference System (ANFIS). The first part of the proposed work is based on Matlab software to develop and train the ANFIS algorithm. Two different types of data sets (temperature and humidity) downloaded from Internet have been used in order to make a comparison between the Matlab Toolbox and modified ANFIS algorithm with momentum factor. The results obtained in this study have shown that the modified ANFIS algorithm is the convenient choice in terms of speed, accuracy.
Abstract
In this study, digital image steganography, a branch of steganography, and DCT (Discrete Cosine Transform) and DWT (Discrete Wavelet Transform, frequency-based steganography methods that are a sub-branch of it, are mentioned. Methods such as MSE (Mean Squared Error), PSNR (Peak Signal Noise Ratio) which are performance calculation parameters of steganographic methods are explained and the methods of calculating image capacity like KL-Divergence, JS-Divergence and QTS (Quard Tree Segmentation) for increasing the values of these parameters are mentioned. This study explains the OPAP (Optimum Pixel Adjustment Process) method, which allows the existing capacity in the pictures to be further increased, in detail and provides an ideal steganography scenario. Here, we made use of the ability of the DWT to extract low frequency and bands suitable for data hiding and the use of the LSB method by obtaining the feature coefficients of DCT in these bands.. In addition, this scenario has been tried and consequently reached the result that the images with higher data concealment capacity than QTS have higher PSNR values.