Search

Morphological operations have been an integral part of the enhancement of digital imaging programs, especially for filtering noise to improve the quality of images by utilizing the two most basic morphological operations: erosion and dilation. The main role of dilation is to fill the defined region in an image with pixels, whereas erosion removes pixels from the region. As we know, the methods of erosion followed by dilation, or dilation followed by erosion, are indeed attractive approaches amongst researchers who deal with filtering noise problems. However, these approaches need more computational time and have a high-percentage chance of losing essential pixel area. To cover these issues, this paper introduces a new approach called an adaptive morphological operation to boost the performance of image enhancement. Based on 2011, 2013, 2015, and 2016 weather image datasets collected from the WITH radar installed on the rooftop of the Information Engineering building, University of the Ryukyus, the experimental results confirm that the proposed approach is more efficient than the conventional approaches.

Mango’s grade is determined by its color and sugar content. However currently it is done by human visually. Visual inspection of grade of mango is human power consuming operation. To this end, the system to classify mangoes by their images has been developed. Traditional system is based on the color histograms of mangoes and thresholding their feature. However traditional research did not validate the assumption that there is correlation between mango’s color, And its sugar content and the assumption is essential for image-based classification approach. To this end, authors showed the evidence that there is some correlation between mango’s color and its sugar content.

Estimation and analysis of traffic jams plays a vital role in an intelligent transportation system and advances safety in the transportation system as well as mobility and optimization of environmental impact. For these reasons, many researchers currently mainly focus on the brilliant machine learning-based prediction approaches for traffic prediction systems. This paper primarily addresses the analysis and comparison of prediction accuracy between two machine learning algorithms: Naive Bayes and K-Nearest Neighbor (K-NN). Based on the fact that optimized estimation accuracy of these methods mainly depends on a large amount of recounted data and that they require much time to compute the same function heuristically for each action, we propose an approach that applies multi-threading to these heuristic methods. It is obvious that the greater the amount of historical data, the more processing time is necessary. For a real-time system, operational response time is vital, and the proposed system also focuses on the time complexity cost as well as computational complexity. It is experimentally confirmed that K-NN does much better than Naive Bayes, not only in prediction accuracy but also in processing time. Multithreading- based K-NN could compute four times faster than classical K-NN, whereas multithreading - based Naive Bayes could process only twice as fast as classical Bayes.

Unexpected heavy rainfall has been seriously occurred in most parts of the world, especially during monsoon season. As a serious consequence of heavy rainfall, the people in those areas battered by heavy rainfall faced many hardship lives. Without exception, prevention is the best way of minimizing these negative effects. In spite of all, we developed a rainfall series prediction system for different series patterns by applying the dynamic programming approach aiming to acquire the rainfall level of the whole rainfall cycle. The simple idea behind the proposed dynamic programming approach is to find the similarity of two rainfall sequences upon the maximum match of the rainfall level of those sequences. Based on 2011 and 2013 real data sets collected from WITH radar, which is installed on the rooftop of Information Engineering, University of the Ryukyus, the comparison between the conventional approach (Polynomial Regression) and the proposed approach is investigated. These correlation experiments confirm that the dynamic programming approach is more efficient for predicting different rainfall series.

The k-Nearest Neighbor (k-NN) is very simple and powerful approach to conceptually approximate real-valued or discrete-valued target function. Many researchers have recently approved that K-NN is a high-prediction accuracy algorithm for a variety of real-world systems using many different types of datasets. However, as we know, k-NN is a type of lazy learning algorithms as it has to compare to each of stored training examples for each observed instance. Besides, the prediction accuracy of k-NN is under the influence of K values. Mostly, the higher K values make the algorithm yield lower prediction accuracy according to our experiments. For these issues, this paper focuses on two properties that are to upgrade the classification accuracy by introducing Regional Distance-based k-NN (RD-kNN) and to speed up the processing time performance of k-NN by applying multi-threading approach. For the experiments, we used the real data sets (wine, iris, heart stalog, breast cancer, and breast tissue) from UCI machine learning repository. According to our test cases and simulations carried out, it was also experimentally confirmed that the new approach, RD-kNN, has a better performance than classical kNN. Keywords: k-NN, RD-kNN, Journal of Information and Communication Engineering (JICE)