Klasifikasi Penyakit Padi berdasarkan Citra Daun Menggunakan Model Terlatih Resnet101

Rice is a staple food source for most countries in the world, including Indonesia. The problem of rice disease is a problem that is quite crucial and is experienced by many farmers. Approximately 200,000 - 300,000 tons per year the amount of rice attacked by pests in Indonesia. Considerable losses are caused by late-diagnosed rice plant diseases that reach a severe stage and cause crop failure. The limited number of Agricultural Extension Officers (PPL) and the Lack of information about disease and proper treatment are some of the causes of delays in handling rice diseases. Therefore, with the development of information technology and computers, it is possible to identify diseases by utilizing Artificial Intelligence, one of which is by using recognition methods based on image processing and pattern recognition technology. The purpose of this research is to create a Machine Learning model by applying the model architecture from Resnet101 combined with the model architecture from the author. The model proposed in this study produces an accuracy of 98.68%.

Garbage Classification Using Ensemble DenseNet169

Garbage is a big problem for the sustainability of the environment, economy, and society, where the demand for waste increases along with the growth of society and its needs. Where in 2019 Indonesia was able to produce 66-67 million tons of waste, which is an increase from the previous year of 2 to 3 million tons of waste. Waste management efforts have been carried out by the government, including by making waste sorting regulations. This sorting is known as 3R (reduce, reuse, recycle), but most people do not sort their waste properly. In this study, a model was developed that can sort out 6 types of waste including: cardboard, glass, metal, paper, plastic, trash. The model was built using the transfer learning method with a pretrained model DenseNet169. Where the optimal results are shown for the classes that have been oversampling previously with an accuracy of 91%, an increase of 1% compared to the model that has an unbalanced data distribution. The next model optimization is done by applying the ensemble method to the four models that have been oversampled on the training dataset with the same architecture. This method shows an increase of 3% to 5% while the final accuracy on the test of dataset is 96%.

Implementasi BGP dan Resource Public Key Infrastructure menggunakan BIRD untuk Keamanan Routing

The validity of the routing advertisements sent by one router to another is essential for Internet connectivity. To perform routing exchanges between Autonomous Systems (AS) on the Internet, a protocol known as the Border Gateway Protocol (BGP) is used. One of the most common attacks on routers running BGP is prefix hijacking. This attack aims to disrupt connections between AS and divert routing to destinations that are not appropriate for crimes, such as fraud and data breach. One of the methods developed to prevent prefix hijacking is the Resource Public Key Infrastructure (RPKI). RPKI is a public key infrastructure (PKI) developed for BGP routing security on the Internet and can be used by routers to validate routing advertisements sent by their BGP peers. RPKI utilizes a digital certificate issued by the Certification Authority (CA) to validate the subnet in a routing advertisement. This study aims to implement BGP and RPKI using the Bird Internet Routing Daemon (BIRD). Simulation and implementation are carried out using the GNS3 simulator and a server that acts as the RPKI validator. Experiments were conducted using 4 AS, 7 routers, 1 server for BIRD, and 1 server for validators, and there were 26 invalid or unknown subnets advertised by 2 routers in the simulated topology. The experiment results show that the router can successfully validated the routing advertisement received from its BGP peer using RPKI. All invalid and unknown subnets are not forwarded to other routers in the AS where they are located such that route hijacking is prevented.

Pengamanan Data melalui Model Super Enkripsi Autokey Cipher dan Transposisi Kolom

One of the essential instruments in the cyber era is data. Therefore, maintaining data security is an important thing to do. One way that can be done to maintain data security is through cryptography. In cryptography, two basic techniques are commonly used, namely substitution techniques and transposition techniques. One of the weaknesses of the basic cryptographic techniques is the lower level of data security. This study proposed a super encryption model in securing data by combining cryptographic algorithms with substitution techniques, i.e., autokey cipher and transposition, i.e., columnar transposition cipher. This study used the Avalanche Effect method as a measurement tool for the proposed super encryption model. The test results have shown that the proposed super encryption model can provide a better level of security. The avalanche effect test on the five data test shows that the average AE value of the proposed super encryption model is 30.76%. This value is higher than the single autokey cipher algorithm of 1.66% and column transposition with a value of 18.03%. Other results from the five data test have shown that the proposed model has a high level of accuracy of 100% in terms of the decryption process results, which is the same as the initial data before going through the encryption process.

Penerapan Convolutional Neural Networks untuk Mesin Penerjemah Bahasa Daerah Minangkabau Berbasis Gambar

One of efforts by the Indonesian people to defend the country is to preserve and to maintain the regional languages. The current era of modernity makes the regional language image become old-fashioned, so that most them are no longer spoken. If it is ignored, then there will be a cultural identity crisis that causes regional languages to be vulnerable to extinction. Technological developments can be used as a way to preserve regional languages. Digital image-based artificial intelligence technology using machine learning methods such as machine translation can be used to answer the problems. This research will use Deep Learning method, namely Convolutional Neural Networks (CNN). Data of this research were 1300 alphabetic images, 5000 text images and 200 vocabularies of Minangkabau regional language. Alphabetic image data is used for the formation of the CNN classification model. This model is used for text image recognition, the results of which will be translated into regional languages. The accuracy of the CNN model is 98.97%, while the accuracy for text image recognition (OCR) is 50.72%. This low accuracy is due to the failure of segmentation on the letters i and j. However, the translation accuracy increases after the implementation of the Leveinstan Distance algorithm which can correct text classification errors, with an accuracy value of 75.78%. Therefore, this research has succeeded in implementing the Convolutional Neural Networks (CNN) method in identifying text in text images and the Leveinstan Distance method in translating Indonesian text into regional language texts.

Analisis Algoritma Shi-Tomasi Dalam Pengujian Citra Senyum Pada Wajah Manusia

Problems in image processing to obtain the best smile are strongly influenced by the quality, background, position, and lighting, so it is very necessary to have an analysis by utilizing existing image processing algorithms to get a system that can make the best smile selection, then the Shi-Tomasi Algorithm is used. the algorithm that is commonly used to detect the corners of the smile region in facial images. The Shi-Tomasi angle calculation processes the image effectively from a target image in the edge detection ballistic test, then a corner point check is carried out on the estimation of translational parameters with a recreation test on the translational component to identify the cause of damage to the image, it is necessary to find the edge points to identify objects with remove noise in the image. The results of the test with the shi-Tomasi algorithm were used to detect a good smile from 20 samples of human facial images with each sample having 5 different smile images, with test data totaling 100 smile images, the success of the Shi-Tomasi Algorithm in detecting a good smile reached an accuracy value of 95% using the Confusion Matrix, Precision, Recall and Accuracy Methods.

Implementasi Metode Unsupervised Learning Pada Sistem Keamanan Dengan Optimalisasi Penyimpanan Kamera IP

Security monitoring systems using face recognition can be applied to CCTV or IP cameras. This is intended to improve the security system and make it easier for users to track criminals is theft. The experiment was carried out by detecting human faces for 24 hours using different cameras, namely an HD camera that was active during the day and a Night Vision camera that was active at night. The application of Unsupervised Learning method with the concept of an image cluster, aims to distinguish the faces of known or unknown people according to the dataset built in the Raspberry Pi 4. The user interface media of this system is a web-based application built with Python Flask and Python MySQL. This application can be accessed using the domain provided by the IP Forwarding device which can be accessed anywhere. According to the test results on optimization of storage, the system is able to save files only when a face is detected with an average file size of ± 2.28 MB for 1x24 hours of streaming. So that this storage process becomes more efficient and economical compared to the storage process for CCTV or IP cameras in general.

Sistem Keamanan Gedung Menggunakan Kinect Xbox 360 Dengan Metode Skeletal Tracking

The incidence of fire and theft is very threatening and causes disruption to people's lifestyles, both due to natural and human factors resulting in loss of life, damage to the environment, loss of property and property, and psychological impacts. The purpose of this study is to create a building security system using Kinect Xbox 360 which can be used to detect fires and loss of valuable objects. The data transmission method uses the Internet of Things (IoT) and skeletal tracking. Skeletal detection uses Arduino Uno which is connected to a fire sensor and Kinect to detect suspicious movements connected to a PC. Kinect uses biometric authentication to automatically enter user data by recognizing objects and detecting skeletons including height, facial features and shoulder length. The ADC (Analog to Digital Converter) value of the fire sensor reading has a range between 200-300. The fire sensor detects the presence of fire through optical data analysis containing ultraviolet, infrared or visual images of fire. The data generated by Kinect by detecting the recognition of the skeleton of the main point of the human body known as the skeleton, where the reading point is authenticated by Kinect from a range of 1.5-3 meters which is declared the optimal measurement, and if a fire occurs, the pump motor will spray water randomly. to extinguish the fire that is connected to the internet via the wifi module. The data displayed is in the form of a graph on the Thingspeak cloud server service. Notification of fire and theft information using the delivery system from input to database

Numerical Approach of Symmetric Traveling Salesman Problem Using Simulated Annealing

The aim of this paper is to elaborate the performance of Simulated Annealing (SA) algorithm for solving traveling salesmen problems. In this paper, SA algorithm is modified by using the interaction between outer and inner loop of algorithm. This algorithm produces low standard deviation and fast computational time compared with benchmark algorithms from several research papers. Here SA uses a certain probability as indicator for finding the best and worse solution. Moreover, the strategy of SA as cooling to temperature ratio is still given. Thirteen benchmark cases and thirteen square grid symmetric TSP are used to see the performance of the SA algorithm. It is shown that the SA algorithm has promising results in finding the best solution of the benchmark cases and the squared grid TSP with relative error 0 - 7.06% and 0 – 3.31%, respectively. Further, the SA algorithm also has good performance compared with the well-known metaheuristic algorithms in references.

Raised Cosine Filter Implementation on Digital Video Broadcasting Satellite 2 (DVB-S2)

The development of digital video broadcasting is still continue recently and was done by many parties. One of the project regarding this research was DVB project. There was three areas in digital video broadcasting. One of them was Digital Video Broadcasting Satellite Second Generation (DVB-S2). The development of this project is not focus only in video broadcasting but also focus in applications and mutlimedia services. The objective of this research was to implement raised cosine filter in DVB-S2 using matlab simulink in order to optimize SNR and BER value. Parameters used in this project was QPSK mode and LDPC with 50 iteration. Those parameters was chosen to maintain originality of data that sent in noisy channel. The result showed that by implementing raised cosine filter could optimized BER value of the system. The higher SNR value would give the lower BER value. In static video, the best SNR value when using a filter is 0.9 dB with a BER value of 0.000004810 while for dynamic video the SNR is 0.9 with a BER value of 0.00001030.