EMITTER - International Journal of Engineering Technology
Not a member yet
261 research outputs found
Sort by
Comparison of Adaptive Ant Colony Optimization for Image Edge Detection of Leaves Bone Structure
No full textLeaf bone structure has a characteristic that can be used as a reference in digital image processing. One form of digital image processing is image edge detection. Edge detection is the process of extracting edge information from an image. In this research, Adaptive Ant Colony Optimization algorithm is proposed for edge image detection of leaf bone structure. The Adaptive Ant Colony Optimization method is a modification of Ant Colony Optimization, in which the initial an ant dissemination process is no longer random, but it is done by a pixel placement process that allows for an edge based on the value of the image gradient. As a comparison also performed edge detection using Robert and Sobel method. Based on the experiments performed, Adaptive Ant Colony Optimization algorithm is capable of producing more detailed image edge detection and has thicker borders than others. Keywords: edge detection, ant colony optimization, robert, sobe
Power Generation Forecasting of Dual-Axis Solar Tracked PV System Based on Averaging and Simple Weighting Ensemble Neural Networks
Get PDFSolar power is a renewable energy interest many researchers around the world to be explored for human life beneficial especially for electric power generation. Photovoltaic (PV) is one of technology developed massively to exploit the solar power for this purpose. However, its performance is very sensitive to environmental condition such as solar irradiance, weather, and climatic behavior. Thus, the hybrid power generation systems are developed to solve this output uncertainty problem. To support this such hybrid system, this paper proposes an ensemble neural network based forecaster of the power output of PV systems which will lead an efficient power management. The object of this research is the PV systems equipped with two axes automated solar tracking with peak power 10Wp. The proposed ensemble forecaster model employs four multi-layer perceptron neural networks with two hidden layers as base forecasters while the input number of historical data is varied in order to exploit the forecaster diversity. The final prediction is calculated both by conventional averaging and simple weighting optimized by the least square fitting technique. According to the research results, the both proposed approaches provide low error rate. Moreover, in term of comparison, the ensemble model with averaging combining technique gives the highest accuracy comparing to the other ensemble and conventional neural network structures
Botnet Detection Using On-line Clustering with Pursuit Reinforcement Competitive Learning (PRCL)
Get PDFBotnet is a malicious software that often occurs at this time, and can perform malicious activities, such as DDoS, spamming, phishing, keylogging, clickfraud, steal personal information and important data. Botnets can replicate themselves without user consent. Several systems of botnet detection has been done by using classification methods. Classification methods have high precision, but it needs more effort to determine appropiate classification model. In this paper, we propose reinforced approach to detect botnet with On-line Clustering using Reinforcement Learning. Reinforcement Learning involving interaction with the environment and became new paradigm in machine learning. The reinforcement learning will be implemented with some rule detection, because botnet ISCX dataset is categorized as unbalanced dataset which have high range of each number of class. Therefore we implemented Reinforcement Learning to Detect Botnet using Pursuit Reinforcement Competitive Learning (PRCL) with additional rule detection which has reward and punisment rules to achieve the solution. Based on the experimental result, PRCL can detect botnet in real time with high accuracy (100% for Neris, 99.9% for Rbot, 78% for SMTP_Spam, 80.9% for Nsis, 80.7% for Virut, and 96.0% for Zeus) and fast processing time up to 176 ms. Meanwhile the step of CPU and memory usage which are 78 % and 4.3 GB for pre-processing, 34% and 3.18 GB for online clustering with PRCL, and  23% and 3.11 GB evaluation. The proposed method is one solution for network administrators to detect botnet which has unpredictable behavior in network traffic
Rule-based Sentiment Degree Measurement of Opinion Mining of Community Participatory in the Government of Surabaya
Get PDFDiskominfo Surabaya, as a government agency, received much community participatory for improvement of governmental services, with increasing number of 698, 2717, 4176 and 4298 participatory data respectively in 2011, 2012, 2013 and 2014. It is challenging for Diskominfo Surabaya to set a target by giving the response back within 24 hours. Due to task complexity to address the degree of participatory and to categorize the group of participatory, they faced difficulty to fulfill the target. In this research, we present a new system for measuring the sentiment degree of community participatory. We provide 5 functions in our system, which are: (1) Data Collection, (2) Data Preprocessing, (3) Text Mining, (4) Sentiment Analysis and (5) Validation. We propose our rule-based technique for the sentiment analysis of opinion mining with detection of 8 important parts, which are (1) Verb, (2) Adjective, (3) Preposition, (4) Noun, (5) Adverb, (6) Symbol, (7) Phrase, and (8) Complimentary. For applicability of our proposed system, we made a series of experiment with 410 data of community participatory in Twitter for Diskominfo Surabaya and compared with other sentiment classification algorithms which are SVM and Naive Bayes Classifier. Our system performed 77.32% rate of accuracy and outperformed to other comparing algorithms
Real Performance Evaluation On MQTT and COAP Protocol in Ubiquitous Network Robot Platform (UNRPF) for Disaster Multi-robot Communication
Get PDFDisaster multi-robot has a significant role in a disaster area to do many tasks like detection of fire, search and rescue of victims, etc. It needs to build good communication between the operator and multi-robot and among multi-robot themselves to perform their tasks quickly and efficiently. This relates with the queue message protocol system. In this research, we implemented the queue message protocol on mesh topology and integrated it on the robot platform. Recently, development of IoT (Internet of Things) Technology causes development of communication protocol. MQTT and CoAP are among the communication protocols used for IoT needs. Both protocols performance were compared when  used and implemented into disaster multi-robot. We also integrated MQTT protocol and robot platform python based (UNR-PF). The result shows that MQTT protocol is easier to be implemented on to disaster multi-robot platform (UNR-PF) on mesh topology than CoAP, and that data transfer rate of MQTT protocol has data transfer rate higher than CoAP
Application of Sliding Mode Control in Indirect Field Oriented Control (IFOC) for Model Based Controller
Get PDFIndirect Field Oriented Control (IFOC) is one of the vector control methods that can be applied to induction motor in the industrial world rather than Direct Field Oriented Control (DFOC) because of the flux is obtained from the formulation. However, IFOC can not guarantee the robustness and stability of the systems. Stability analysis such as Lyapunov Stability Theory can be used to make the system stable but not the robustness. Model based controller that can guarantee the stability and robustness such as sliding mode control (SMC) and fuzzy needs to be added in IFOC system to achieve proportional response system. Robust current regulator using sliding mode control was designed in this paper from state space model for model based controller. In transient response and under disturbance SMC shows better performance than PID in rising time and robustness at rotor speed and stator current
Trusted Data Transmission Using Data Scrambling Security Method with Asymmetric Key Algorithm for Synchronization
Get PDFSecurity is a major concern of the internet world because the development of the Internet requires the security of data transmission. The security method helps us to store valuable information and send it over an insecure network so that it can not be read by anyone except the intended recipient. Security algorithm uses data randomization method. This method of data information randomization has a low computation time with a large number of bits when compared to other encryption algorithms. In general, the encryption algorithm is used to encrypt data information, but in this research the encryption algorithm is used for synchronization between the sender and the intended recipient. Number of bits on asymmetric key algorithm for synchronization are the 64-bits, 512-bits and 1024-bits. We will prove that security methods can secure data sent with low computational time with large number of bits. In the result will be shown the value of computing time with variable number of bits sent. When data are sent by 50 bytes, encryption time required 2 ms using 1024 bits for synchronization technique asymmetric key algorithm.Â
Technique of Standing Up From Prone Position of a Soccer Robot
Get PDFOne of the humanoid robots being developed in the field of sports is a soccer robot. A soccer robot is a humanoid robot that can perform activities such as playing football. And a variety method fall down of robot soccer such: falling down toward the front direction, side direction, and rear direction. This paper describes the most stands up methods of a soccer robot from its prone position. The proposed method requires only limited movement with degrees of freedom. The movement standing-up of soccer robot has been implemented on the real robot. Tests we performed showed that reliable standing-up from prone position is possible after a fall and such recovery procedures greatly improve the overall robustness of a Soccer Robot
Automatic Samples Selection Using Histogram of Oriented Gradients (HOG) Feature Distance
Get PDFFinding victims at a disaster site is the primary goal of Search-and-Rescue (SAR) operations. Many technologies created from research for searching disaster victims through aerial imaging. but, most of them are difficult to detect victims at tsunami disaster sites with victims and backgrounds which are look similar. This research collects post-tsunami aerial imaging data from the internet to builds dataset and model for detecting tsunami disaster victims. Datasets are built based on distance differences from features every sample using Histogram-of-Oriented-Gradient (HOG) method. We use the longest distance to collect samples from photo to generate victim and non-victim samples. We claim steps to collect samples by measuring HOG feature distance from all samples. the longest distance between samples will take as a candidate to build the dataset, then classify victim (positives) and non-victim (negatives) samples manually. The dataset of tsunami disaster victims was re-analyzed using cross-validation Leave-One-Out (LOO) with Support-Vector-Machine (SVM) method. The experimental results show the performance of two test photos with 61.70% precision, 77.60% accuracy, 74.36% recall and f-measure 67.44% to distinguish victim (positives) and non-victim (negatives)
Influence of Logistic Regression Models For Prediction and Analysis of Diabetes Risk Factors
Get PDFDiabetes is a very serious chronic. Diabetes can occurs when the pancreas doesn't produce enough insulin (a hormone used to regulate blood sugar), cause glucose in the blood to be high. The purpose of this study is to provide a different approach in dealing with cases of diabetes, that's with data mining techniques mengguanakan logistic regression algorithm to predict and analyze the risk of diabetes that is implemented in the mobile framework. The dataset used for data modeling using logistic regression algorithm was taken from Soewandhie Hospital on August 1 until September 30, 2017. Attributes obtained from the Hospital Laboratory have 11 attribute, with remove 1 attribute that is the medical record number so it becomes 10 attributes. In the data preparation dataset done preprocessing process using replace missing value, normalization, and feature extraction to produce a good accuracy. The result of this research is performance measure with ROC Curve, and also the attribute analysis that influence to diabetes using p-value. From these results it is known that by using modeling logistic regression algorithm and validation test using leave one out obtained accuracy of 94.77%. And for attributes that affect diabetes is 9 attributes, age, hemoglobin, sex, blood sugar pressure, creatin serum, white cell count, urea, total cholesterol, and bmi. And for attributes triglycerides have no effect on diabetes