A Hybrid Approach of Using Particle Swarm Optimization and Volumetric Active Contour without Edge for Segmenting Brain Tumors in MRI Scan

One of the main reservoir development plans is to find optimal locations for drilling new wells in order to optimize cumulative oil recovery. Reservoir simulation is a necessary tool to study different configurations of well locations to investigate the future of the reservoir and determine the optimal places for well drilling. Conventional well-known numerical methods require modern hardware for the simulation and optimization of large reservoirs. Simulation of such heterogeneous reservoirs with complex geological structures with the streamline-based simulation method is more efficient than the common simulation techniques. Also, this method by calculation of a new parameter called “time-of-flight” (TOF) offers a very useful tool to engineers. In the present study, TOF and distribution of streamlines are used to define a novel function which can be used as the objective function in an optimization problem to determine the optimal locations of injectors and producers in waterflooding projects. This new function which is called “well location assessment based on TOF” (WATOF) has this advantage that can be computed without full time simulation, in contrast with the cumulative oil production (COP) function. WATOF is employed for optimal well placement using the particle swarm optimization (PSO) approach, and its results are compared with those of the same problem with COP function, which leads to satisfactory outcomes. Then, WATOF function is used in a hybrid approach to initialize PSO algorithm to maximize COP in order to find optimal locations of water injectors and oil producers. This method is tested and validated in different 2D problems, and finally, the 3D heterogeneous SPE-10 reservoir model is considered to search locations of wells. By using the new objective function and employing the hybrid method with the streamline simulator, optimal well placement projects can be improved remarkably.

Download Full-text

A Hybrid Approach Using Particle Swarm Optimization and Parallel Simulated Annealing: A Case Study of n-Queens Problem

International Journal of Computer and Electrical Engineering ◽

10.7763/ijcee.2014.v6.828 ◽

2014 ◽

Vol 6 (3) ◽

pp. 231-235 ◽

Cited By ~ 1

Author(s):

Pourya Jafarzadeh ◽

Vahid Mohammadi Saffarzadeh

Keyword(s):

Particle Swarm Optimization ◽

Simulated Annealing ◽

Hybrid Approach ◽

Particle Swarm ◽

Swarm Optimization ◽

Parallel Simulated Annealing

Download Full-text

Multiple Active Contours Driven by Particle Swarm Optimization for Cardiac Medical Image Segmentation

Computational and Mathematical Methods in Medicine ◽

10.1155/2013/132953 ◽

2013 ◽

Vol 2013 ◽

pp. 1-13 ◽

Cited By ~ 8

Author(s):

I. Cruz-Aceves ◽

J. G. Aviña-Cervantes ◽

J. M. López-Hernández ◽

S. E. González-Reyna

Keyword(s):

Particle Swarm Optimization ◽

Image Segmentation ◽

Active Contour ◽

Medical Image ◽

Active Contours ◽

Active Contour Model ◽

Particle Swarm ◽

Magnetic Resonance Images ◽

Swarm Optimization ◽

Contour Model

This paper presents a novel image segmentation method based on multiple active contours driven by particle swarm optimization (MACPSO). The proposed method uses particle swarm optimization over a polar coordinate system to increase the energy-minimizing capability with respect to the traditional active contour model. In the first stage, to evaluate the robustness of the proposed method, a set of synthetic images containing objects with several concavities and Gaussian noise is presented. Subsequently, MACPSO is used to segment the human heart and the human left ventricle from datasets of sequential computed tomography and magnetic resonance images, respectively. Finally, to assess the performance of the medical image segmentations with respect to regions outlined by experts and by the graph cut method objectively and quantifiably, a set of distance and similarity metrics has been adopted. The experimental results demonstrate that MACPSO outperforms the traditional active contour model in terms of segmentation accuracy and stability.

Download Full-text