Volume 1 Issue 1
Authors: E. A. Zanaty; Ashraf Afifi
Abstract: In this paper, a new modified fuzzy c-means algorithm is presented which could improve the medical image segmentation. The proposed algorithm is realized by modifying the objective function of the conventional FCM algorithm with a flexible penalty. This penalty is based on a data shape and data size used for the generation of fuzzy terms. The complexity of the proposed algorithm is reduced using initial seed information into the objective function instead of whole data set. The proposed algorithm is applied to magnetic resonance image (MRI) datasets. Compared with the existing approaches, the proposed method can achieve the best accurate results. The results of the conducted experiments show that the efficiency of the proposed method in preserving the regions homogeneity and its robustness in segmenting noisy images is better than other FCM-based methods.
Keywords: Fuzzy Clustering; Modified Fuzzy C-Means; Medical Image Segmentation
Authors: Zhongguo Li; Jie Hou; Kai Wang; Qinghua Liu
Abstract: Improvement of Particle Swarm Optimization (PSO) algorithm is a significant work. In this paper, a practical method is proposed to instruct this task with recording all particle search positions and tracking the best particle shift process. Firstly, appropriate velocity bounds are obtained with tracking particle velocity components during iterations. Then particle initialization method is modified. Uniform Probability Random Value (UPRV) is substituted with Uniform Distributed Fixed Value (UDFV) to initiate particles. And it concludes a significant performance improvement. Stochasticity of results initialized with UDFV apparently decreases. It also makes PSO better cover with the search space which causes greater probability to obtain the global best. At least 3 particles can be competent for task with UDFV initialization method after analyzing the best particle shift process among all particles. That greatly enhances the algorithm speed. This paper can be a reference for application and improvement of PSO algorithm used in Support Vector Machine (SVM) parameter optimization.
Keywords: Support Vector Machine; Vertical load; Road type recognition; Initialization method
Authors: Michael F. Dossis; Dimitrios E. Amanatidis
Abstract: An integrated, formal high-Level synthesis (HLS) framework is used in this work for hardware implementation of cellular neural networks, which are used in real time image processing. The Custom Coprocessors Compilation (CCC) HLS behavioral synthesiser generates correct-by-construction register transfer – level (RTL) VHDL hardware models of computation-intensive applications. Thus, time-consuming RTL and gate-level simulations are avoided and verification time is cut down to a fraction of the usual time that takes to achieve the same goal with traditional approaches. Such applications include image processing with cellular neural networks (CNNs). The synthesizer utilizes formal compiler-compiler and logic programming techniques, to transform algorithmic ADA into RTL VHDL or Verilog which are directly implementable into hardware using any available RTL synthesizer. The CNNs were rapidly coded, compiled and verified along with all the necessary testbenches in GNU ADA. The applications targeted here are edge-detection, halftoning and morphological processing, which are used to evaluate the CCC HLS framework. The contribution of this work is hardware implementation of CNNs using the CCC HLS tools to formally, and rapidly develop, verify and prototype advanced image processing applications.
Keywords: Image morphological processing; Formal methods; Cellular Neural Networks; High-level Synthesis; VHDL; ADA; High-level verification