EMITTER - International Journal of Engineering Technology
Not a member yet
261 research outputs found
Sort by
Observation of Fish Dissemination Pattern on Madura Coastal Using Segmentation of Satellite Images
Full text linkAlmost traditional fishermen still use manual methods to catch fish that rely on experience in fishing and information among fellow fishermen. This method is not effective for maximizing fish production. A good pattern or strategy is needed to increase fish production. In determining dissemination pattern of fish, it can be predicted from physical parameters such as temperature, salinity, chlorophyll, turbidity, total suspended solids, and colored dissolved organic matter using the Landsat 8 images. Â This research area is on the Island of Madura Coast. The pattern is determined by using Lagrange Interpolation and clustering using K-Means. The results of the study of the pattern of fish dissemination were then validated with data from the Dinas Kelautan dan Perikanan Jawa Timur. The results between fish patterns and validation data in 2015 showed similarities in January, February, March, May, June, July, August, September. In 2016, results between fish patterns and validation data showed that similarities in July, August, September, and December. In 2017, results between fish patterns and validation data showed similarities in November. 2015 has the most similarities between the patterns and validation data and the least similarity are 2017
Augmented Reality as Learning Medium for Preservation of Traditional Musical Instruments in Bangka
Full text linkNowadays the use of technology is something that can be found anywhere. This condition has an impact on the loss of awareness of the Indonesian cultural treasures value in the form of traditional tools. No exception to traditional musical instruments on Bangka Island which began to lose its popularity. At present, most teenagers on Bangka Island cannot play traditional musical instruments. Likewise with the children who do not know and not even know their own regional musical instruments. With the continuation of this condition, it is feared that the existence of traditional Bangka musical instruments will disappear, as well as human resources that can play it. Augmented Reality (AR) is a visual technology that can display objects in 3D. The advantage of this technology is being able to give a display of real-time musical instrument in the form of dynamic 3D visualization of objects and it is in accordance with the movements of the user's smartphone camera. AR has also been applied in various cases as a solution to problem-solving. Therefore, to overcome this problem, an application to preserve the traditional Bangka musical instruments using Augmented Reality (AR) is forming. The 3D objects of musical instruments are made using Maya. Unity is also used as an engine for the application of 3D modeling on the Android system, and Vuforia SDK as it’s augmented reality engine. The results of performance testing obtained 100% running well. From the results of testing the user experience with the HARUS method, it is proven that the system has comprehensibility aspects of 75.98% and manipulability aspects of 80.74% so that the total value HARUS be 78.36%
Spatio-Temporal Associative Mining for Earthquake Data Distribution in Indonesia
Full text linkIndonesia is a country that has the highest seismically activity in the world. This country has really high earthquake frequency because of it traversed by three plate meeting plate and located in Ring of Fire area. The shaking events from an earthquake are very strong and propagate in all directions, capable of destroying even the strongest civilian buildings, so there is no doubt that there are many victims of human lives. The other facts, earthquake in Indonesia have seismic relation between the provinces. In this paper, we present a new earthquake Spatio-temporal mapping system based on the association confidence value from the result of associative mining process on earthquake data distribution in Indonesia. The system proposed three main functions which are (1) Data Acquisition which taken from four data provider, then preprocess and combine it become one, (2) Associative Mining process to get the rule of association earthquake between provinces in Indonesia, and (3) Earthquake Association Spatio-Temporal Model from the highest confidence value and Visualization. We use data from several earthquake data providers from 1900 until 2018. To perform our proposed Spatio-temporal earthquake association mapping system, we divided the data to become a 5-year discrete partition. After that, we mining the rule and get the highest confidence value from each period. This confidence value is used for modeling and visualization of our Spatio-temporal mapping system. As a result of this study, we manage to generate earthquake association risk mapping from 13 provinces that had earthquake connectivity between each other. The provinces are Aceh, Sumatera Utara, Bengkulu, East Java, Bali, NTB, NTT, Maluku, North Maluku, Gorontalo, North Sulawesi, Papua dan West Papua
Focused Time Delay Neural Network For Tuning Automatic Voltage Regulator
Full text linkThis paper proposes a novel controller for automatic voltage regulator (AVR) system. The controller is used Focused Time Delay Neural Network (FTDNN). It does not require dynamic backpropagation to compute the network gradient. FTDNN AVR can train network faster than other dynamic networks. Simulation was performed to compare load angle (load angle) and Speed. The performance of the system with FTDNN-AVR has compared with a Conventional AVR (C-AVR) and RNN AVR. Simulations in Matlab/Simulink show the effectiveness of FTDNN-AVR design, and superior robust performance with different cases
Content-Dependent Image Search System for Aggregation of Color, Shape and Texture Features
Full text linkThe existing image search system often faces difficulty to find a appropriate retrieved image corresponding to an image query. The difficulty is commonly caused by that the users’ intention for searching image is different with dominant information of the image collected from feature extraction. In this paper we present a new approach for content-dependent image search system. The system utilizes information of color distribution inside an image and detects a cloud of clustered colors as something - supposed as an object. We applies segmentation of image as content-dependent process before feature extraction in order to identify is there any object or not inside an image. The system extracts 3 features, which are color, shape, and texture features and aggregates these features for similarity measurement between an image query and image database. HSV histogram color is used to extract color feature of image. While the shape feature extraction used Connected Component Labeling (CCL) which is calculated the area value, equivalent diameter, extent, convex hull, solidity, eccentricity, and perimeter of each object. The texture feature extraction used Leung Malik (LM)’s approach with 15 kernels. For applicability of our proposed system, we applied the system with benchmark 1000 image SIMPLIcity dataset consisting of 10 categories namely Africans, beaches, buildings historians, buses, dinosaurs, elephants, roses, horses, mountains, and food. The experimental results performed 62% accuracy rate to detect objects by color feature, 71% by texture feature, 60% by shape feature, 72% by combined color-texture feature, 67% by combined color-shape feature, 72 % combined texture-shape features and 73% combined all features
Multi-Distance Veins Projection Based on Single Axis Camera and Projector System
Full text linkEvery person has different location of veins, some veins are easily detected because it is visible due to thin tissue, and the other are invisible. This different location of veins causes intravenous access procedures and the procreas of intravenous therapy become longer. Multi-distance vein projections aim to simplify the measurement process where the device and object do not have to be at a certain distance. Some research that has been done especially for real-time vein projection does not conduct how the characteristics of projection at different distances. In this paper, we propose a method for performing multi-distance real-time back-projection by using the intersection between camera and projector. This method equiped with an ultrasonic distance sensor to identify the projection characteristic in any distance. In its implementation, this method is able to project at a distance of 20-40 cm with a maximum projection error of 0.6 mm. The measurement angle tolerance between the object and the device is ±5 degrees with a maximum error of 0.7 mm
Automatic Detection of Wrecked Airplanes from UAV Images
Full text linkSearching the accident site of a missing airplane is the primary step taken by the search and rescue team before rescuing the victims. However, due to the vast exploration area, lack of technology, no access road, and rough terrain make the search process nontrivial and thus causing much delay in handling the victims. Therefore, this paper aims to develop an automatic wrecked airplane detection system using visual information taken from aerial images such as from a camera. A new deep network is proposed to distinguish robustly the wrecked airplane that has high pose, scale, color variation, and high deformable object. The network leverages the last layers to capture more abstract and semantics information for robust wrecked airplane detection. The network is intertwined by adding more extra layers connected at the end of the layers. To reduce missing detection which is crucial for wrecked airplane detection, an image is then composed into five patches going feed-forwarded to the net in a convolutional manner. Experiments show very well that the proposed method successfully reaches AP=91.87%, and we believe it could bring many benefits for the search and rescue team for accelerating the searching of wrecked airplanes and thus reducing the number of victims
Implementation Fuzzy C-Means on Decision Support System BPNT (Bantuan Pangan Non-Tunai) Ministry of Social Affairs Indonesia
Full text linkDecision Support System can be an alternative solution to determine the candidate's decision. Bantuan Pangan Non-Tunai (BPNT) are selected based on criteria determined by the Ministry of Social Affairs of the Republic of Indonesia. BPNT recipients are conducted by the government to help someone who is less able to meet their daily needs. The occurrence of errors in determining the eligibility of prospective beneficiaries is a major problem, based on these problems there needs to be an information system that can provide a valid BPNT recommendation and one of which uses a grouping method with the Fuzzy C-Means (FCM) algorithm. System development using the waterfall method. The results of system implementation and testing showed that 90% of the system was following what was expected according to the results of the test with the system being built
Nuclei Detection and Classification System Based On Speeded Up Robust Feature (SURF)
Full text linkTumors contain a high degree of cellular heterogeneity. Various type of cells infiltrate the organs rapidly due to uncontrollable cell division and the evolution of those cells. The heterogeneous cell type and its quantity in infiltrated organs determine the level maglinancy of the tumor. Therefore, the analysis of those cells through their nuclei is needed for better understanding of tumor and also specify its proper treatment. In this paper, Speeded Up Robust Feature (SURF) is implemented to build a system that can detect the centroid position of nuclei on histopathology image of colon cancer. Feature extraction of each nuclei is also generated by system to classify the nuclei into two types, inflammatory nuclei and non-inflammatory nuclei. There are three classifiers that are used to classify the nuclei as performance comparison, those are k-Nearest Neighbor (k-NN), Random Forest (RF), and State Vector Machine (SVM). Based on the experimental result, the highest F1 score for nuclei detection is 0.722 with Determinant of Hessian (DoH) thresholding = 50 as parameter. For classification of nuclei, Random Forest classifier produces F1 score of 0.527, it is the highest score as compared to the other classifier
AGC of a multi sources power system with natural choice of power plants
Full text linkThis paper presents an application of optimal control theory in multi sources power system by considering natural choice of power plants participating in automatic generation control (AGC) scheme. However, for successful operation of large power system, the natural choices of generation suitable for AGC system are hydro and thermal power plants since gas and nuclear power plants are rarely participates in the AGC scheme. Therefore, this work presents design and implementation of proportional integral (PI) structured optimal AGC controller in the presence of hydro and thermal power plants by using state vector feedback control theory. Moreover, various case studies are identified to obtain: (i) Cost aspects of physical realization of optimal AGC controller, (ii) Closed loop system stability margin through patterns of eigenvalues and (iii) System dynamic performance. Further, results have shown that when optimal AGC scheme is implemented in power system, the dynamic performance of power system is outstanding over those obtained with genetic algorithms (GAs) tuned PI structured AGC controller. Besides, with optimal AGC controller, cheaper cost of control structure, increased in system closed loop stability margin and outstanding dynamic performance of power system have been found when lessening in hydro generation is replaced by generation from thermal power plants for various case studies under investigation