Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control
Not a member yet
424 research outputs found
Sort by
Object Detection and Monitor System for Building Security Based on Internet of Things (IoT) Using Illumination Invariant Face Recognition
Get PDFTheft and intrusion are crimes that often occur in neighborhoods when there is opportunity or negligence by owners and security personnel. Many studies have been carried out to improve environmental security by applying cameras as a surveillance medium. However, the camera is not optimal in detecting objects when the lighting conditions are lacking. Therefore, in this study, a monitoring and object detection system was built by applying the Illumination Invariant model. This model is used to improve the appearance of the image from light and shadow reflections. The process of detecting and identifying objects is done by using human facial features (face detection) captured by the camera. The camera used is a Logitec C270 Webcam 720p which is connected via a USB port on the Raspberry Pi 4. The Raspberry Pi 4 processes human face image data and sends the processing results to a MySQL database using the HTTP protocol. Data transmission is done using the Python Flask web framework. The system was successfully run 100% by using black box testing of all functional requirements. Tests on the object detection feature were carried out based on different lighting conditions 15 times by comparing the original image and the results of the Illumination Invariant implementation. Based on the test results obtained object detection accuracy of 86.7%
Neural Network-Based Image Processing for Tomato Harvesting Robot
Get PDFAgriculture is one of the areas that can benefit from robotics technology, as it faces issues such as a shortage of human labor and access to less arid terrain. Harvesting is an important step in agriculture since workers are required to work around the clock. The red ripe tomatoes should go to the nearest market, while the greenest should go to the farthest market. Harvesting robots can benefit from Neural Network-based image processing to ensure robust detection. The vision system should assist the mobility system in moving precisely and at the appropriate speed. The design and implementation of a harvesting robot are described in this study. The efficiency of the proposed strategy is tested by picking red-ripened tomatoes while leaving the yellowish ones out of the experimental test bed. The experiment results demonstrate that the effectiveness of the proposed method in harvesting the right tomatoes is 80%
Classification of Coffee Leaf Diseases using CNN
Get PDFIndonesia’s coffee industry plays a crucial role as a major export, making a significant contribution to the country’s economy by generating foreign exchange. The quality and quantity of coffee production depend on various factors such as humidity, rain, and fungus that can cause rust diseases on coffee leaves. These diseases can spread quickly and affect other coffee plants quality, leading to decreased production. To address this issue, CNN with VGG-19 architecture model was utilized to identify coffee plant diseases using image data and the python programming language, which in previous studies used MATLAB as their platform. In addition, VGG-19 with image enhancement and contouring data for pre-processing step has a more profound learning feature than the method used in the previous studies, AlexNet which makes the structure of VGG- 19 more detailed. The dataset used in this paper is Robusta Coffee Leaf Images Dataset which have three classes, namely health, red spider mite, and rust. The VGG-19 model attained F1-Score of 90% when evaluated using the testing data with ratio 80:20, where 80% is training data, and 20% is validation data. This paper employed 0.0001 learning rate, batch size 15, momentum 0.9, 12 training iteration, and RMSprop optimizer
Integration of Fuzzy C-Means and SAW Methods on Education Fee Assistance Recipients
Get PDFEvery year, UMTAS gets a quota for KIP tuition fee assistance provided by KEMDIKBUD. This program is intended for high school / vocational/equivalent graduates from poor and vulnerable families. The evaluation results of its implementation have problems, including the number of applicants exceeding the quota given by KEMDIKBUD and some applicants coming from well-off families. This research uses the fuzzy c-means method for data clustering and the SAW method for ranking. The results of data grouping using the fuzzy c-means method obtained the first cluster (C1) of 72 data and the second cluster (C2) of 119 data. Group C1 is closer to the provisions of aid recipients (eligible) compared to data group C2 (ineligible) because Data C1 consists of 100% DTKS recipients, 50% KIP and KKS card owners, 100% parental income <750,000, 40.28% parental dependents >=2 people and 29.17% applicants with achievements. 72 registrant data included in Data C1 are then ranked using the SAW technique to get weights, and 30 data with the highest weight will be used as a decision on recipients of KIP-Kuliah Education fee assistance according to the quota provided. The optimization of Fuzzy C-Means with SAW methods in selecting recipients of education fee assistance is objective and right on target
The Implementation of Pretrained VGG16 Model for Rice Leaf Disease Classification using Image Segmentation
Get PDFRice is an agricultural sector that produces rice which is one of the staple foods for the majority of the population in Indonesia. In the cultivation of rice plants there are also factors that affect rice production and are not realized by farmers causing that they are late in handling and diagnosing symptoms and making rice production decline. Therefore, it is necessary to have an early diagnosis of rice plants to identify them correctly, quickly and accurately. Machine learning is one of the classification techniques to detect various plant diseases such as rice plants. There are several studies on machine learning using the Convolutional Neural Network with the VGG16 model to classify rice leaf diseases and using Image Segmentation techniques on rice leaf datasets for make the image becomes a form that is not too complicated to analyze. The data used in this research is Rice Leaf Disease which consists of 3 classes including Bacterial leaf blight, Brown spot, and Leaf smut. Then segmentation is carried out using two techniques, namely threshold and k means. Then data augmentation for make dataset used has a large and varied number and training using VGG16 model with hyperparameter tuning and obtained 91.66% accuracy results for scenarios with the k-means dataset
Design and Performance of Solar-Powered Surveillance Robot for Agriculture Application
Get PDFAgriculture can benefit from robotics technology to overcome the drawback of limited human labor working in this sector. One of the robot applications in agriculture is a surveillance robot to monitor the condition. This paper describes a surveillance robot that is powered by a capacitor bank charged by a mini solar panel. The solar-powered robot is well-suited for deployment in open agricultural areas in Indonesia, where the irradiance is high. This potential is excellent for generating electricity and charging electric vehicles, such as those used in agriculture. The surveillance robot developed and tested in this study has been successfully deployed in an agriculture-like setting with all-terrain contours and the capacity to avoid obstacles. During high irradiance sunny weather, the shortest charging time was 2 hours. Hence, the proposed technology is effective for designing a surveillance robot for agricultural applications
Evaluation of Stratified K-Fold Cross Validation for Predicting Bug Severity in Game Review Classification
Get PDFSteam review data provides a lot of information for the game development team, either positive or negative reviews. It is essential as negative and positive reviews provide crucial information, and 7% of positive reviews contains bug reports. These bug reports were captured after the game was released, and many reports of common problems still exist. If players found an issue in the game, they could report it directly through the review feature provided by the online game platform. However, it took a long time for the development team to manually analyze and classify the reviews. This study proposed a new approach to automatically classify the reviews on Steam based on the bug severity level. Therefore, to solve this problem, we recommend a solution based on the research background indicated above. For this experiment, we analyzed reviews on two popular game titles namely, FIFA 23 and Apex Legends. We implemented three different classifiers, namely KNN, Decision Tree, and Naïve Bayes, which would be used to train a dataset to classify the bug severity level. Due to the imbalanced dataset, we performed cross-validation to reduce bias in the dataset. Performance in this model would be evaluated using accuracy rate, precision, recall, and F1 score. As a result, the experiment showed that game reviews of different game titles achieved different accuracy scores. The game review classification for FIFA 23 performed better than the game review classification for Apex Legends. The mean accuracy score of FIFA 23 was 72% with Decision Tree and Apex Legend was 64% with KNN
Design MPPT with Anfis Method on Zeta Converter with DC Load
Get PDFMaximum power point tracking (MPPT) for PV (Photovoltaic) systems is provided in this research using artificial intelligence-based control. The design of MPPT system with Anfis Method on the Zeta Converter with DC Load is used to optimize the work of the Photovoltaic which will be used for DC load sources. The MPPT process consists of four main stages, namely module training data, determining input and output data, determining the number and type of membership functions and ANFIS training data. Zeta converter works like a buck boost, which can increase or decrease the voltage which is an advantage in designing systems with very volatile Photovoltaic sources. Zeta Converter is used to get higher efficiency, smaller input and output current ripple values and smaller core losses in the inductor. To improve the efficiency of system performance, An MPPT algorithm for the adaptive neuro fuzzy inference system (ANFIS) that is programmed into a microcontroller controls the zeta converter. ANFIS control is used because the response is faster and more effective. The combined simulation's findings demonstrate that the ANFIS control was successful, and the system can now produce the best possible power from Photovoltaic ipanelsiiniMPPT mode by boosting efficiency by up to 19.96%
PID Controller Implementation on Animal Experimental Treadmill for Heart Medicine Purpose
Get PDFExperimental animals such as rats are often used for medical research and therapy, such as cardiologists who use a special treadmill to measure the heart health of rats by training walking or running in order to determine the appropriate dose for individuals before being applied to their patients. This research designed a system that is operated by the speed of a DC motor. To control the system, it is proposed to implement a Proportional Integral Derivative (PID) control that is able to stabilize the rotation of the DC motor based on the BPM value recorded by the encoder sensor. The value is used as feedback to the PID control, so that it can control the speed of the DC motor and work optimally and stably under load or no load. Adding a limit switch as a fatigue zone to determine the final duration. This system was tested on several objects, namely 4-month-old rats with a mass of 211 grams, 224 grams, 230 grams, and 240 grams and 2-month-old rats with a mass of 24 grams, 27 grams, 28 grams, and 30 grams. The results show that the speed reading using PID control is in accordance with the constants Kp = 17, Ki = 7, and Kd = 1. This test has a percentage overshoot (%) of 5% and an average rise time value of 0.14 seconds. System performance with a percentage accuracy of 90% starting from a setpoint of 35 m/min
Anomaly Detection Analysis with Graph-Based Cyber Threat Hunting Scheme
Get PDFAs advanced persistence threats become more prevalent and cyber-attacks become more severe, cyber defense analysts will be required to exert greater effort to protect their systems. A continuous defense mechanism is needed to ensure no incidents occur in the system, one of which is cyber threat hunting. To prove that cyber threat hunting is important, this research simulated a cyber-attack that has successfully entered the system but was not detected by the IDS device even though it already has relatively updated rules. Based on the simulation result, this research designed a data correlation model implemented in a graph visualization with enrichment on-demand features to help analysts conduct cyber threat hunting with graph visualization to detect cyber-attacks. The data correlation model developed in this research can overcome this gap and increase the percentage of detection that was originally undetected / 0% by IDS, to be detected by more than 45% and can even be assessed to be 100% detected based on the anomaly pattern that was successfully found