Search

Palm oil is an important commodity for Indonesia. Therefore, oil palm cultivation is a supporter of the success of oil palm commodities in producing products. The harvesting process is an activity that must be considered because it affects the production and quality of oil palm fresh fruit bunches. For this reason, the sorting of oil palm fresh fruit bunches is very important. This paper tries to classify the maturity level of oil palm fresh fruit bunches through image management using the Linear Discriminant Analysis (LDA) algorithm. LDA algorithm is an approach used to classify data into several classes. The steps taken in this paper are starting with the transformation of the color space from RGB images to L*a*b images, then image segmentation with thresholding and feature extraction with HSV. Image transformation to HSV aims to make it easier to classify image characteristics based on hue and saturation. Furthermore, the LDA algorithm will perform a dispersion matrix analysis in order to obtain optimal projections to enter into spaces to look for patterns that can be separated. This paper finds that the classification of oil palm fresh fruit bunches image processing using the LDA algorithm produces an accuracy rate of 85%. [ABSTRACT FROM AUTHOR]

Maximal clique enumeration is a graph clustering method for finding all vertices that have the most influence in a graph. The Bron-Kerbosch algorithm is one of the fastest algorithms to find all maximal cliques. Hence, this paper will focus on that algorithm to find all maximal cliques. Counting all maximal cliques of a graph usually uses an adjacency matrix of the graph to find all vertices in the graph that have the most influence. But, in this paper, a complement matrix of a graph will be used in counting all maximal cliques. This research uses a graph that represents a domestic flight route of one of the airlines in Indonesia. By using Bron-Kerbosch algorithm, all maximal cliques of the graph will be listed, so that it will come up with the vertices which are the most influential in this graph. The graph will be represented in complement matrix. The result of applying the Bron-Kerbosch algorithm with the complement matrix of graph will determine vertices that have the most influence in the graph. Besides that, by using a complement matrix, the result gives more information on the vertices which are only connected to the vertices that have the most influence. [ABSTRACT FROM AUTHOR]

Multispectral optical remote sensing images are often contaminated by haze and cirrus, especially for tropical regions such as Indonesia. This will affect the image quality and reduce the accuracy of the interpretation of the existing data. A number of methods for performing haze detection and haze removal have been developed by experts. However, haze detection and removal still remain one of the most important challenges for the correction of multispectral optical data. Haze detection and removal are also needed, especially in producing cloud-free mosaic images. A method to perform haze removal on SPOT 6/7 images using the haze index algorithm is presented in this paper. The haze removal method is developed based on the reflectance values of the blue and the red bands of SPOT 6/7 imagery. A new algorithm with new coefficient is proposed to remove the haze on SPOT 6/7 imagery. Haze removal is done by using haze index algorithm and haze threshold limit value. The experimental result demonstrate that the new coefficient used in the haze index algorithm give better result in performing haze removal on SPOT 6/7 imagery. [ABSTRACT FROM AUTHOR]

Small-Scale Hydropower (SHP) had been important electric energy power source in Indonesia. Indonesia is vast countries, consists of more than 17.000 islands. It has large fresh water resource about 3 m of rainfall and 2 m of runoff. Much of its topography is mountainous, remote but abundant with potential energy. Millions of people do not have sufficient access to electricity, some live in the remote places. Recently, SHP development was encouraged for energy supply of the places. Development of global hydrology data provides opportunity to predict distribution of hydropower potential. In this paper, we demonstrate run-of-river type SHP spot prediction tool using SWAT and a river diversion algorithm. The use of Soil and Water Assessment Tool (SWAT) with input of CFSR (Climate Forecast System Re-analysis) of 10 years period had been implemented to predict spatially distributed flow cumulative distribution function (CDF). A simple algorithm to maximize potential head of a location by a river diversion expressing head race and penstock had been applied. Firm flow and power of the SHP were estimated from the CDF and the algorithm. The tool applied to Upper Citarum River Basin and three out of four existing hydropower locations had been well predicted. The result implies that this tool is able to support acceleration of SHP development at earlier phase. [ABSTRACT FROM AUTHOR]

Maximum Standard Deviation Reduction (MSDR) is a graph clustering algorithm to minimize the distance variation within a cluster. In this paper we propose a modified MSDR by replacing one technical step in MSDR which uses polynomial regression, with a new and simpler step. This leads to our new algorithm called Modified MSDR (MMSDR). We implement the new algorithm to separate a domestic flight network of an Indonesian airline into two large clusters. Further analysis allows us to discover a weak link in the network, which should be improved by adding more flights. [ABSTRACT FROM AUTHOR]

Despite the slow growth of EOR development in Indonesia, CO2 EOR has recently gained its momentum due to its versatility to diverse reservoir systems in Indonesia. Optimizing CO2 Injection in EOR activity is a must, since no incentive or tax holiday is given in Indonesia for this activity, contrary to the majority of policies around the globe. This study offers an innovative approach of CO2 flooding injection rate determination using closed loop optimization in project economic evaluation. Particle Swarm Optimization (PSO) algorithm was implemented as a method of optimizing the injection rate of CO2. These novel algorithms are known and proven to be able to work with massive number of datasets, as well as identifying and separating bad dataset(s). The result is optimum injection rate that brings maximum economic value to the project. Utilization of this method increases the NPV of the project by 10.4% and 12% increase in RF. [ABSTRACT FROM AUTHOR]

Nowadays, nearly all things sold worldwide are transported through sea using ships. It's mean that maritime has a big role impact on human life in the world. However, the increase of maritime mobilities comes also with various issues from transshipment, oil spilling, until hijacking. Initially, a device such as Automatic Identification System (AIS) has been widely deployed and used to provide a cooperative position report for marine communication but nowadays it is also deployed to overcome the issues. In recent work, AIS is also installed on the satellite as space-based AIS to get worldwide coverage. The data have shown that falsification of AIS messages is possible, and therefore could mask or favor illegal actions, lead to disturbance of monitoring systems and new maritime risks. This paper presents these new threats and introduces a novel algorithm for detecting spoofing. The algorithm is quite straight forward by comparing vessel displacement and distance. Some variances are introduced to produce the best algorithm. The area of interest for this research is Indonesia with longitude from 95° to 141° and latitude -11° to 6°. AIS data from LAPAN-A2 and LAPAN-A3 satellites are used with ranged from January 1, 2019 until June 30, 2019. From this research, the best algorithm was achieved by using an average speed with constant tolerance of 61.596 km. [ABSTRACT FROM AUTHOR]

In the Bayesian mixture modeling requires stages the identification number of the most appropriate mixture components thus obtained mixture models fit the data through data driven concept. Reversible Jump Markov Chain Monte Carlo (RJMCMC) is a combination of the reversible jump (RJ) concept and the Markov Chain Monte Carlo (MCMC) concept used by some researchers to solve the problem of identifying the number of mixture components which are not known with certainty number. In its application, RJMCMC using the concept of the birth/death and the split-merge with six types of movement, that are w updating, θ updating, z updating, hyperparameter β updating, split-merge for components and birth/death from blank components. The development of the RJMCMC algorithm needs to be done according to the observed case. The purpose of this study is to know the performance of RJMCMC algorithm development in identifying the number of mixture components which are not known with certainty number in the Bayesian mixture modeling for microarray data in Indonesia. The results of this study represent that the concept RJMCMC algorithm development able to properly identify the number of mixture components in the Bayesian normal mixture model wherein the component mixture in the case of microarray data in Indonesia is not known for certain number. [ABSTRACT FROM AUTHOR]