Yoshiko Hanada

Multi-objective optimization of diesel engine emissions and fuel economy using SPEA2+

The diesel engine has excellent fuel economy and is widely used especially in commercial vehicles. However, there are ever increasing concerns with regard to environmental problems, drawing attention to air pollution caused by the NOx and Soot exhaust from diesel engines. In this paper, diesel engines that have the small amounts of NOx and SOOT with the high fuel economy are tried to design. To design such diesel engines, one of multi-objective genetic algorithms, SPEA2+[3], is applied. SPEA2+ is an enhancement of SPEA2 and has not only the high searching ability but also the mechanism of maintain the diversity in the design variable space. Through the numerical examples, the effectiveness of SPEA2+ was examined by applying SPEA2+, SPEA2[1], and NSGA-II[2] to the fuel emission scheduling problem. Our results indicated that multi-objective optimization is effective for designing diesel engines.

Island model genetic programming based on frequent trees

The Island Model encourages genetic diversity, and often displays better search performance than single population models. In order to enhance the Island Model in the framework of genetic programming (GP), we propose a novel migration strategy based on frequent trees, where the frequent trees in an island mean the sub-trees appearing frequently among the individuals in the island. The proposed method evaluates each island by measuring its activation level in terms of not only how high the best fitness value is but also how many types of frequent trees are newly created, and then makes several individuals migrate from an island with high activation level to an island with low activation level, and vice versa. Using three benchmark problems widely adopted in the literature, we demonstrate that performance improvement can be achieved through incorporating the information of frequent trees into a migration strategy, and the proposed method significantly outperforms a typical method of the Island Model GP.

A New Population Initialization Approach Based on Bordered Hessian for Portfolio Optimization Problems

In the portfolio optimization problems, the proportion-weighted combination in a portfolio is represented as a real-valued array between 0 and 1. While applying any evolutionary algorithm, however, the algorithm hardly takes the ends of a given real value. It means that the evolutionary algorithms have a problem that they cannot give the not-selected asset whose weight is represented as 0. In order to avoid this problem, we propose a new population initialization approach using the extreme point of the bordered Hessian and then apply our approach to the initial population of GA for the portfolio optimization problems in this paper. In the numerical experiments, we show that our method employing the population initialization approach and GA works very well for the portfolio optimizations even if the portfolio consists of the large number of assets.

Depth-dependent crossover in genetic programming with frequent trees

One of the most well studied issues in genetic programming is how to make building blocks efficiently. To make building blocks, it is important to find the substructures that appear in the individuals with higher fitness. Recently, a method based on frequent substructures has been proposed, and it has shown good performance; however, the depth of trees is not considered in the method. In this paper, we propose a hybrid crossover that involves the consideration of a combination of frequent trees and the depth of trees and apply the proposed method to symbolic regression problems. We experimentally demonstrate the effectiveness of the proposed method.

Automatic information retrieval from data embedded printing images using correction of rotational angles based on reference marks

A data extraction method from data embedding printing images by capturing devices like scanners or mobile cameras has attracted much attention. This method has several problems to be solved, which is different with the digital image watermarking. In the method, consideration for geometrical deformation, especially rotations, in the data extraction process is essential. For the correction, addition of a black frame to an image as a marker has been considered. However it may not be preferable in some case, since the frame may change the impression of the image. This paper proposes new correction method without the remarkable marker. This method can adopt the relationship between the mark in the DFT domain and the rotation of images. By embedding the mark in the DFT domain, the rotational angle can be easily obtained. Related to this new embedding marker, the area in the DCT domain for embedding the information is investigated. Also an image resize technique suitable for this method is developed. From the experimental results, the effectiveness of the proposed method is shown.

Implementation of data embedding to printing images for information retrieving by cellular phones considering lens distortion

Recently, several data extraction methods from data embedding printing images by using cellular phones with a camera have been proposed. In these techniques, consideration for geometrical deformation and lens distortion in the data extraction process is essential. This paper proposes new correction algorithm based on line based correction (LBR) and projective transformation to the scanning of the printing images for reduction of the geometrical deformation by the camera. This method can be applied to both of barrel distortion and pincushion one. To increase the amount of embedding data, a new data embedding method in which the codes for data embedding are changed according to the embedding data is also proposed. From experimental results, effectiveness of the proposed method for the color image is shown.

An improvement of unsupervised design method for weighted median filters using GA

Design of both a suitable window shape and appropriate weights in weighted median filters is one of important problems. Hitherto, unsupervised design methods of the filters by using Simulated Annealing (SA) or Genetic Algorithm (GA) have been proposed for texture images corrupted by impulse noise. These techniques estimate the optimal window shape and the optimal filter weights separately, and they have been shown to perform as well as a supervised method. In this paper, we propose a new approach which optimize both the window shape and the weight at the same time to design more sophisticated filters. We apply GA and adopt Rank-Ordered Logarithmic Difference (ROLD) statistics as objective function to design a weighted median filter. Through experiments, it was shown that our new approach outperformed compared to conventional methods.

Data Embedding and Extraction Method for Printed Images by Log Polar Mapping

Data extraction methods from data embedding printed images by capturing devices like scanners or mobile cameras have attracted much attention. In this technique, consideration for geometrical deformation, especially rotation and scaling, in the data extraction process is essential. This paper proposes a new correction algorithm without remarkable markers. This method exploits the log polar mapping (LPM) to detect and correct the distortion. The proposed method estimates the rotational angle and scaling factor from the amount of shift in the LPM domain. Therefore no data embedding or adding for correcting the deformation is required and the data embedded image with high quality is obtained. An image clipping technique suitable for the proposed method is also developed. From the experimental results, the effectiveness of the proposed method is shown.

Mega process genetic algorithm using grid MP

In this study, a new Genetic Algorithm (GA) using the Tabu · Local Search mechanism is proposed. The GA described in this paper is considered a Mega Process GA, which has an effective mechanism to use massive processors, i.e., Mega Processors, in large-scale computing systems. Our proposed method has a GA-specific database that possesses information of searched space and performs a local search for the space that is not searched. Such mechanisms enable us to comprehend the quantitative rate of a searched region during the search. Using this information, the searched space can be expanded linearly as the number of computing resources increases and the exhaustive search is guaranteed under infinite computations. The proposed GA was applied to numerical test functions and the energy minimization problems of protein tertiary structures. The latter problem was performed under a heterogeneous distributed computing environment, which was built up with Grid MP produced by United Devices Inc.

Tabu: local search mechanism for mega process genetic algorithm

In this study a new genetic algorithm (GA) using tabu: local search mechanism for large-scale computer systems is proposed. We call the GA that uses huge computing resources a mega process GA. The GA described in this paper is considered a mega process GA which has the effective mechanism to solve the problems quickly and to use massive processors, namely mega processors, comprised in large-scale computing systems such as super PC clusters and grid computation environments. Our proposed method has a GA-specific database that possesses information of space that has been already searched. At the same time, the proposed GA performs a local search for the space that is not searched. Such mechanisms enable us to comprehend the quantitative rate of a searched region during the search. Using this information, the searched space can be expanded linearly as the number of computing resources increase and the exhaustive search is guaranteed under infinite computations. Using and describing different experiments, the features of the introduced GA are discussed and examined. At first, this method was applied on one max problem and 3-deceptive problem; the former is one of primitive functions and the latter is one of trap functions. Through this experiment, it is shown that the method ensures an effective exhaustive search. This method was then applied to the test functions of continuous optimization problems under restricted computing costs. Using such an experiment, it is clear that this method has the same performance as a conventional GA.