Search

Mangrove plants are one of the plants that really help aquatic ecosystems between the sea, coast, and land. Mangrove plants provide many ecological, social, and economic benefits. In Indonesia, mangrove plants have 202 species with the same anatomy as other plants in general, consisting of roots, fruits, stems and leaves. Nowadays, the location of mangrove plants in Indonesia has experienced the fastest damage in the world due to conversion to ponds, settlements, industry and plantations. One of the efforts to restore aesthetic value and restore the ecological function of mangrove forest areas is rehabilitation using mangrove fruit. In the rehabilitation process, farmers generally use the manual method with the naked eye to determine fruit ripeness on mangrove plants, so the resulting level of accuracy is not optimal. To overcome this problem, an application is needed that can facilitate farmers in determining fruit maturity in mangrove plants so that it can help determine the maturity level of mangrove fruit. The development of this application utilizes the Deep Learning method as well as the utilization of digital image processing techniques with Grayscaling, Adaptive Threshold, Sharpening and Smoothing techniques. The results of this study are an application that can detect the level of fruit maturity in mangrove plants with an accuracy of 99.11%. With this application, determining the maturity level of fruit on mangrove plants can be easily done.

Smart farming in various worlds is not just about applying technology in terms of storing data on agricultural land. However, having a concept of measurable data based on available computational techniques trained and then generating knowledge. As an application, the agri drone sprayer can be used for the process of applying pesticides and liquid fertilizers on each side. In addition, drone surveillance is also useful in implementing smart farming such as mapping land so that farmers will know the condition of their agricultural land. However, the soil and weather sensor will also help the farmers to monitor the farmland as well. Devices with sensors can only obtain data in the form of air and soil humidity, temperature, soil pH, water content and forecasting the harvest period. So that the smart farming model can help farmers to get recommendations, in preventing the predicted damage to their land and crops. However, according to its geographical location, the application of smart farming can be a smart solution to agricultural problems in Indonesia and make the future of Indonesian Agriculture a technology-based smart agriculture.

The marsh frogs that lost their habitat will become pests that are very detrimental to the fish cultivators. A technological approach is mandatory to assure the fish ponds to always be free from the frogs' presence by detecting the frog sound as a sign of frog existence in the breeding pond. This research implemented Mel Frequency Cepstral Coefficients- Vector Quantization (MFCC-VQ) algorithm to identify the frog's sound. The captured sound signal features will be extracted using Mel Frequency Cepstral Coefficients (MFCC) algorithm. The extracted sound signal will be identified using Vector Quantization (VQ) algorithm. The result of this research is the further the distance of the microphone from the sound source and the noisier environmental conditions in the detection process, the smaller the accuracy percentage. This study's highest accuracy is 99.6% at a distance of 1 meter and 83.2% at a distance of 11 meters in normal circumstances. Through the test, it can also be concluded that the further distance of the microphone from the source, the smaller the percentages of specificity and sensitivity, and the slower the processing speed, although the difference in speed per meter is not too significant.

Online news are being spread through social media by news agencies to encourage people to read news from their site. After users log in to their site, users will continue to read the news if it is relevant to their personalized news recommendation. However, nowadays, a personalized recommendation could be provided to users if the site has records of users' browsing history, and users must have logged in to the news site. This could be problematic if the readers are anonymous users, and the site does not have enough browsing history records of these users. It will be difficult for news agencies to increase the number of daily readers on their site, especially if they have to compete with other agencies for providing a good personalized recommendation. Therefore, a personalized recommendation system is necessary for anonymous users who only visit the news site occasionally so that the website can recommend the news they prefer to read by implementing CSR Sparse Matrix Vector Multiplication and proximity processing as sparse matrix applied method. The Bayesian framework for user interest will also be implemented with the modification in its term. This research proved that personalized news recommendations could still be provided to readers with limited knowledge of the users' searching history.

Extraction is a way to get knowledge from information space, such as social networks. One efficient and concise tool for expressing knowledge related to meaning is to involve the concept of similarity. There are several similarity formulations to approach the same thing from the objects but have different tasks according to their functions. However, the measurement results reveal different meanings, although they remain in a mutually supportive position. Therefore, this paper aims to express the different meanings besides new meanings of the similarity function, which are proven based on the difference between the divisors of the similarity formulation. Different measurements of similarities produce different and new meanings with supporting simulation and clustering to manage big data.

Pengelolaan perkebunan kelapa sawit sering mengalami kendala, antara lain masalah organisme pengganggu tumbuhan (OPT) terutama masalah penyakit. Oleh karena itu, dibuatlah pendekatan untuk mengenali penyakit pada daun kelapa sawit agar dapat membantu kinerja dari para petani kelapa sawit dalam menentukan jenis penyakit pada daun sehingga mendapatkan hasil yang lebih maksimal. Deteksi tepi adalah perubahan nilai intensitas derajat keabuan yang mendadak (besar) dalam jarak yang singkat. Sobel operator digunakan untuk pengidentifikasikan pola wajah, khususnya terdapat di dalam algoritma deteksi tepi. Support Vector Machine (SVM) digunakan sebagai metode klasifikasi. Oleh karena itu, dalam penelitian ini penulis akan menerapkan metode deteksi tepi dengan menggabungkan teknik algoritma Sobel Operator untuk menghilangkan derau dan metode Support Vector Machine sebagai pengklasifikasian data penyakit pada daun kelapa sawit. The management of oil palm plantations often experiences obstacles, including problems with plant pest organisms (OPT), especially disease problems. Therefore, an approach was made to encourage the disease in the leaves of oil palm so that it can help the performance of oil palm farmers in determining the type of disease in the leaves so as to get maximum results. Edge detection is a change in the value of the sudden intensity of the degree of gray (large) in a short distance. Sobel operators are used to identifying face patterns, especially those found in edge detection algorithms. Support Vector Machine (SVM) is used as a classification method. Therefore in this study, the author will apply the edge detection method by combining the Sobel Operator algorithm technique to eliminate noise and the Support Vector Machine method as a classification of disease data on palm oil leaves.

The text summarizer reduces unnecessary information by selecting the important sentences. In multi-document summarization, there is a possibility that two or more important sentences share similar information. Including those sentences to the summary result will cause redundant information. This research aims to reduce similar sentences from multi-document that share similar information to obtain a more concise text summary. In order to accomplish the objective, this research uses the combination of several online news articles, divided into six groups. The combined articles are pre-processed to produce a clean text. After obtaining the clean text, this research utilizes the TextRank algorithm to extract the important sentences by using the similarity measurement. This process yields the summarized text. However, the summarized text is still containing similar sentences. The next process is calculating Maximal Marginal Relevance (MMR) to reduce similar sentences. The result of this process is the final text summary. The evaluation uses ROUGE-1 and ROUGE-2 with the average F-score is 0.5103 and 0.4257, respectively.

The Indonesian government has developed a database called TRUST+™ Positif which contains the list of blacklisted URLs. These URLs are blacklisted because they contain some prohibited materials or negative contents such as pornography, radicalism, fraud, racism, violence, gambling, and security threat. The government requires all the Internet Service Provider (ISP) in Indonesia to block Internet access to websites listed on TRUST+™ Positif database. The government expects this action will help reducing the spread of negative contents, especially pornography. One of the many ways to disseminate pornographic content is by publishing those articles in the websites. This research purpose is to provide the pornographic corpus in order to be the reference for pornography identification research. The corpus is built from 1,000 articles from 150 websites based on TRUST+™ Positif database. This research only extracts the sentences that related to pornography. The extraction is done based on the selected keywords. These keywords are generated by the most frequent words the articles and should be related to pornography. There are 447 keywords that have been selected manually. The result of this research is a pornographic corpus in Bahasa Indonesia that consists of 35,753 sentences.

This study aims to automatically identify and calculate the crown of oil palm trees by utilizing digital image processing. The data of this research were the images of the oil palm tree taken by a drone. The proposed algorithm to identify the oil palm tree was the Haar-Cascade Classifier. The processes to perform the identification process of the oil palm trees were the calculation of the value of an integral image, Haar-like features, Adaboost algorithm, cascade classifier, grayscale and the last is the identification stage of oil palm trees. The result of this study was that the system could identify the oil palm trees using the Haar-Cascade Classifier algorithm to calculate the number of palm trees in a field.

Image processing has the characteristic of input and output data in the form of images when the segmentation process is performed using Raspberry Pi with limited parameters; it will take a long time to process. An alternative can be applied using a cluster computer method to overcome this issue. In this study, a Raspberry cluster system will be built using Open MPI. Open MPI served as a compiler cluster by dividing the work process into several connected clusters device nodes so that the program can be run on cluster devices simultaneously. The test result showed that the program worked well in the cluster scope and the image conversion process performed better in the cluster scope than in single device.

The soil fertility of plantation area is vital for determining the farmers' success in producing rice. Several factors of the rice plants environment are mandatory to be monitored to achieve the maximum quality of the harvests. Today, the use of control systems are widely encountered in various aspects of life, manual measurements to determine the environmental conditions of rice plants will be less efficient in terms of time. Therefore, a system is required that is able to monitor the rice crops in real time. The method implemented in the research was supported by several parameters such as changes in humidity, temperature, pH, soil moisture and light intensity. Data of every sensor will be stored in the database and farmers can receive notification on plantation area monitoring in the form of SMS Gateway. These sensors will be placed in several locations on the field to improve the measurement accuracy. Based on the 1-week test, all sensors generated stable values and average errors of pH at 1.99 temperature at 1.29%, air humidity at 0.54%, soil moisture at 0.31%, and light intensity at 0.49%. The result of the test was aligned with the architecture, and all sensors showed that the monitoring went well.

Weather and climate are the main factors that are very influential in the activities of human life. The profile measurement of air pressure, temperature, humidity, horizontal wind direction speed and other information related to weather observations in the vertical structure of the atmospheric is essential for the Meteorology, Climatology, and Geophysical Agency (BMKG). Therefore, it is compulsory to implement a tool to measure the vertical profile of the atmosphere and greenhouse gases (GHG) in real-time in Indonesia. The tool will assist the observation of climate parameters in each area so that weather forecasting can be more efficient to be conducted in various locations. In this study, authors had developed a method to measure the vertical profile of the atmosphere and greenhouse gases in real-time. The proposed method of this research has several parameters, namely air temperature, humidity, air pressure, carbon monoxide, and carbon dioxide. The design of the capacity platform will be tested at the Meteorology, Climatology, and Geophysics Agency. The results of the measurement conducted at BMKG are still within the tolerance limits of the average deviation of the measurement of the weather parameters set by the World Meteorological Organization (No.8 of 2006), which are 0.2 °C for air temperatures, 0, 3 hPa (mbar) for air pressure and 3 % of humidity.

Compression, in general, aims to reduce file size, with or without decreasing data quality of the original file. Digital Imaging and Communication in Medicine (DICOM) is a medical imaging file standard used to store multiple information such as patient data, imaging procedures, and the image itself. With the rising usage of medical imaging in clinical diagnosis, there is a need for a fast and secure method to share large number of medical images between healthcare practitioners, and compression has always been an option. This work analyses the Huffman coding compression method, one of the lossless compression techniques, as an alternative method to compress a DICOM file in open PACS settings. The idea of the Huffman coding compression method is to provide codeword with less number of bits for the symbol that has a higher value of byte frequency distribution. Experiments using different type of DICOM images are conducted, and the analysis on the performances in terms of compression ratio and compression/decompression time, as well as security, is provided. The experimental results showed that the Huffman coding technique has the capability to compress the DICOM file up to 1 : 3.7010 ratio and up to 72.98% space savings.

The Regional Revenue and Expenditure Budget (APBD in Indonesia) is a financial plan stipulated by Regional Regulations. The APBD of the North Sumatra Provincial Government is the reflection of development direction in the Province of North Sumatra since it would not be possible to execute the regional development process without being supported by adequate cash. Therefore, this study aims to predict the APBD allocation in 5 types of accounts, namely Regional Original Revenues, Balancing Funds, Other Legal Regional Revenues, Indirect Expenditures, and Direct Expenditures using time series data from 2002 to 2019. The proposed method in this study is Autoregressive Integrated Moving Average, which is a forecasting method using the time series data in the form of current and past data of the APBD. The results showed that the forecasting of the 2020 APBD value for the Regional Original Revenue account is IDR 6,989,217 Million, for the Balance Fund account is worth IDR 8045.259 Million, for Other Legitimate Income is at IDR 317,546 Million, for Indirect Shopping is worth IDR 9,308,607 Million and for Direct Shopping at IDR 5,495,597 Million.

Facial recognition is one of the best forms of security since it is biometric-based security that uses biological features of the face. One of the artificial neural networks that can be implemented in the face recognition case is Extreme Learning Machine with the assistance of Local Binary Pattern in obtaining the facial features. The testing process in this study used 150 face images for training data and 60 face images for test data. The test was carried out using several three parameters of the hidden neuron numbers, namely 10, 30, and 50, and also five parameters of the image condition, namely feature, expression, face direction, lighting, and webcam distance. The system achieved the highest accuracy of 90% using 50 hidden neurons.

Recently, the usage of short messaging service (SMS) or text messages have been changed gradually to product or service promotion, and even fraud. The mobile phone users in Indonesia also experience the same condition. A simple approach to address this issue is creating black list of phone numbers or certain keywords and phrases. However, this approach is inefficient because the spammer might change the phone number or change the content of the text message. Meanwhile, another approach is utilizing text classification such as Naive Bayes, k-Nearest Neighbor (kNN), and Support Vector Machine (SVM) to recognize pattern of the text messages. This research proposes Twitter-LDA algorithm to identify spam text messages in Bahasa Indonesia. There are total 985 text messages divided to 774 text messages for training dataset and 211 text messages for testing dataset. These datasets consist of 860 spam and 125 ham text messages. All the text messages should be pre-processed before the training and testing process are applied. This research conducts five experiments which yield the average of f-score is 94.26% and accuracy is 96.49%. According to this result, the Twitter-LDA algorithm has demonstrated a good performance in identifying spam text messages in Bahasa Indonesia.

Cane (Saccharum officinarumL.) is a plant that is used as the main raw material in the sugar industry. One of the factors that can cause the reduction of sugar cane production and inhibit the development of the sugar industry is the late-diagnosed of leaf stain disease in sugarcane crops that cause crop failure. It is quite often the case because of the limited number of service facilities for communication and also the high cost to be incurred if the sugar cane entrepreneurs want to conduct the inspection process of the crop. Therefore, this study has a goal to identify leaf stain disease in sugarcane plants. This research used the image of sugar cane leaves that have stain spots as input for the image processing process using the Speeded-Up Robust Features (SURF) method. The identification process of leaf stains is divided into three types of leaf stain disease, namely ring stains, rust stains, and yellow stains. In this study, 60 images were used as preliminary data and 30 images for the testing process. Following the conducted test, it can be concluded that the early identification process of leaf stain disease in sugar cane has an accuracy rate of 90%.

The losses suffered by the farmers due to bird attack could reach of 15 – 50 percent of harvest yield. The peasants use conventional pesticides and this still conducted manually which is very inefficient. Therefore, this study has the purpose of developing an application of rice-eating bird voice detector. This application can automatically detect the sounds of birds that are eating the rice to facilitate the farmers to monitor and repel the birds' presence in the paddy to reduce losses during harvest seasons. Mel Frequency Cepstral Coefficients-Vector Quantization (MFCC-VQ) was used as the method for bird's voice recognition. The feature of voice signals captured via microphone will be extracted using the Mel Frequency Cepstral Coefficients (MFCC) algorithm. The extracted audio signal is then identified whether the sound is a bird or not using the Vector Quantization (VQ) algorithm. The identification result will generate the output of a firing sound as an action to cast out and scare the birds away from the fields. The result of this study is that the sound of birds was detected depending on the arrival of birds in the field such as during the morning, afternoon and evening. The result also showed that the further the distance of the microphone from the sound source, the smaller the intensity of the voice and the noisy the state of the environment on the detection process, the smaller the accuracy percentage.

Text is an element which provides information to the readers. However, not all text is informative, some are still in need to be processed to generate information. In text processing process, data is required to be inputted into the system. The input process will be easier if the text is already in digital form. The main issue is when the text is in the non-digital form such as in the form of image. This image should be converted into a form which recognized by the machine. Therefore, an approach is required to be able to identify the text on the images, in expectation to generate text that can be processed by the machine. The method proposed for this research to identify the text is Convolutional Neural Network. Before and after entering the identification process, the input image will go through several pre-processing and post-processing phases to select which text to be displayed as a result. The testing process used images of receipt taken at the distance of 10cm and 12 cm. The result showed the accuracy rates of the testing using images of receipt taken at the distance of 10cm and 12 cm are 95% and 85% respectively.

Human has their limitation of recognizing the detail of an object and requires longer time to identify it, especially for identifying the teeth which is usually done by forensic dentist. Therefore, a method is required for tooth identification in dental panoramic radiograph image. In this research, we use multilayer perceptron with Levenberg-Marquardt backpropagation training algorithm, to identify molar and premolar teeth based on the region property of isolated teeth generated from the integral projections. Several image processing techniques are also implemented to overcome the low contrast of the dental panoramic radiograph image. The experimental results show that our proposed method can help the forensic dentist to identify the molar and premolar teeth and give 98.2143% of accuracy while testing 10 dental panoramic radiograph images (consist of 56 teeth) with 20 images (consist of 152 teeth) for training data.

Babylon Tower is a three-dimensional puzzle which consists of six discs that are arranged piled up. Each disc consists of six columns of small balls along the side in six kinds of different colors. Babylon Tower can be played by rotating and sliding it. The goal of the game of Babylon Tower is to sort each ball of the same color on the same column and sort the balls in each column based on the brightness of the color of the ball. There are two methods that can be used to find a solution of the game of Babylon Tower, namely disc by disc and column by column. The method proposed in this research is depth-first search column by column. In this research, it is shown that the proposed method is capable of finding step by step to reach a solution of the game of Babylon Tower.

The growing number of users and development of applications that utilize the computer network or the internet has an impact on increasing amount of data transferred through the network. With this increasing data transfer results in the increase of traffic (flow) data. A monitoring system is needed to analyze computer network by capturing on the flow of network data packets. In this study, the monitoring system analyzed and identified types of flooding and packet loss using random early detection (RED) algorithm. RED method compares the packet's counter value by its type with a user defined threshold value. The monitoring and packet capturing system was implemented on the campus wireless LAN network at the University of Sumatra Utara during active hours. As the result of the system implementation performed for several consecutive days, it was found that the active packets included TCP, UDP, and Ethernet. For the most common intrusion mode was detected as TCP / SYN flooding and packet loss (97.04%) on TCP packets with loss percentage in the network was around 8.837%

In performing an assignment, every employee is expected to complete the work to some level. However, the misuse of working hours happens regularly in their working environment, which mean they shorten their workload and working hour. This problem led to the need to build a system as an alternative in monitoring the presence of the employee in their working environment. The type of the work that requires monitoring due to its presence in the location are customer service, call center, cashier, etc. Monitoring the presence of the employee can be done using tools such as RFID. Trilateration method was used to determine object location using signal value received of an object from multiple transmitters. The signal value can be then converted to a distance value between the object and the transmitters also can be formed three circles with the radius of each circle is the distance value between the object with the transmitters. The result from our system is the employee's location on building virtual maps. System will determine whether the employee location is in their working environment or not then it will produce daily reports on the employee working hours. In the end, system will sum up the workload obtained from the results of its employee presence monitoring.

Hypertensive retinopathy refers to a disease condition that damages the retinal vascularization in hypertensive patients, so that it lead to the blindness and death. It can affect the retina function because of a swelling in the retinal vascular to be. Usually, this condition is identified manually by ophthalmologist through the fundoscopy examination using ophthalmoscope. Therefore, a method for identifying hypertensive retinopathy automatically based on the retinal fundus images is needed. In this research, we proposed Probabilistic Neural Network (PNN) for identifying hypertensive retinopathy through retinal fundus images. Before the identification process, the image processing and feature extraction used in the proposed method are box counting and invariant moments. Based on the experiment result, the proposed method has been able to identify hypertensive retinopathy with the accuracy reached 100%.

When diagnosing a disease, besides the physical examination, blood analysis is a reliable method. This because blood has components that contains a lot of key informations. Morphogical examination of peripheral blood smears is one of important lab examinations and has to be evaluated properly. But abnormal red blood cell shapes that found by a health analyst is not always the same as other analyst because of precision factor, concentration, and lack of knowledge. Besides that, morphogical examination of peripheral blood smears still done manually by health anaylists and they considered less efficient because they took a lot of time. To solve that problem, a method to classify red blood cell types that detects abnormal shapes of cells from certain disease. In this paper, radial basis function network is used as method to classify abnormal red blood cell types. Several stage before executing classification process is input image, pre-processing, feature extract with canny edge detection. Research result shows that by using this method, the accuracy to classify abnormal red blood cell types is 83.3%.

Hierarchical Graph Neuron (HGN) is an extension of network-centric algorithm called Graph Neuron (GN), which is used to perform parallel distributed pattern recognition. In this research, HGN scheme is used to classify intrusion attacks in computer networks. Patterns of intrusion attacks are preprocessed in three steps: selecting attributes using information gain attribute evaluation, discretizing the selected attributes using entropy-based discretization supervised method, and selecting the training data using K-Means clustering algorithm. After the preprocessing stage, the HGN scheme is then deployed to classify intrusion attack using the KDD Cup 99 dataset. The results of the classification are measured in terms of accuracy rate, detection rate, false positive rate and true negative rate. The test result shows that the HGN scheme is promising and stable in classifying the intrusion attack patterns with accuracy rate reaches 96.27%, detection rate reaches 99.20%, true negative rate below 15.73%, and false positive rate as low as 0.80%.

Lung cancer is the most common cause of cancer-related human deaths in both men and women. Doing chest radiography is one of the first steps was investigated for lung cancer identification. In this study, theChest X-Ray imageswould be used as an input in pre-processing and continuing the feature extraction to be used in neural network process as a method proposed to identify lung cancer using backpropagation algorithm. Pre-processing is a first step to do some process such as scaling and greyscaling, the segmentation was K-means clustering process, then do the feature extraction usingGrey-Level Co-occurrence Matrix (GLCM) and identification using Backpropagation done by this study to identify the lung cancer. This study showed that the proposed method was able to identify the presence of cancer with an accuracy of 75%, sensitivity 75% and specificity of 75%.

The development of information technology in the future will continue to grow and full of innovations, all shortcomings in an application or equipment will continue to be minimized, all ways to simplify and efficiently a technology are continuously looked for and developed by experts in their respective fields. The frequent use of 3D objects and animation in medical learning media is felt to be incomplete without the direct interaction of users with modern computers so that users are more interested in learning for Ear, Nose and Throat. Therefore, in this study, an interaction system was made on the Ear Nose Throat medical learning application using Microsoft HoloLens. In this study used interactions that use hand gesture and voice input. The results of this study, users will be able to interact with the Ear, Nose and Throat using Microsoft HoloLens. With this extensive application, it will facilitate users in Ear, Nose and Throat medical learning more interactively with the unique features of applications that use voice input and hand gesture.

Children education adheres to the emphasis of playing while learning or learning while playing. Playing using the body muscles can stimulate the body senses of a child and assist the child in exploring the surroundings. Children discover and learn new things or skills through playing a game. In creating an interactive learning experience to stimulate the brain of a child, a solar system is included in the formal education curriculum packed in an interactive media using Kinect where the children will play an active role in studying the solar system. The Kinect sensor detects 23 joints of a child to be used to move the 3-dimensional astronaut object directly on the body (coordinates mapper) and uses the background removal on the image to separate the background and the child pixels, so the child seems to be beside the planet. Based on the test results using the respondents of 50 children, it was shown that the system effectiveness level for the children is 86%.

In this paper, we propose an approach to detect and geotag advertisement billboard in real-time condition. Our approach is using AlexNet’s Deep Convolutional Neural Network (DCNN) as a pre-trained neural network with 1000 categories for image classification. To improve the performance of the pre-trained neural network, we retrain the network by adding more advertisement billboard images using inductive transfer learning approach. Then, we fine-tuned the output layer into advertisement billboard related categories. Furthermore, the detected advertisement billboard images will be geotagged by inserting Exif metadata into the image file. Experimental results show that the approach achieves 92.7% training accuracy for advertisement billboard detection, while for overall testing results it will give 71.86% testing accuracy.

Tuberculosis is an airborne contagious disease caused by the Bacterium Mycobacterium Tuberculosis which means major threat to global health. If Tuberculosis is not treated promptly and quickly, it will cause the disease to get worse. For that, the treatment should not be done carelessly. Therefore, a system is needed to identify the Tuberculosis diseases to help the doctors as an early stage in diagnosing the disease. In this research, Probabilistic Neural Network (PNN) method was used to identify the presence of Tuberculosis by system to help the doctor to make the decision. The stages before doing an identification such as the image acquisition, pre-processing, and feature extractions using Invariant Moment. This study was conducted using train data of 105 normal X-Ray images and 105 images of pulmonary Tuberculosis X-Rays. For the testing data used of 50 X-Ray images. The results showed that the proposed method was able to identify the Tuberculosis disease with an accuracy of 96%.

Otolaryngology is a medical science that focuses on the problems of the ears, nose and throat. Otolaryngologists in their roles are people who have been trained in medicine and surgery (surgery) in the ENT department. To get maximum results, a good practice is needed from the beginning to the end of the examination. 3D digital model is one of the services to facilitate pupils in choosing the desired education method. 3D Object is a depiction of an object that is equipped with a perception of space. 3D Object is an object that is designed and drawn the length, width, and height (x, y, z axis) of the object so that it looks more real. The result showed that each organ was detected by displaying a 3-dimensional object in accordance with the original object. Mixed reality is a technique to merge a virtual simulation and real-world environment. 3D digital model in mixed reality technology is one way to make the ENT learning more attractive. With the existence of a 3D digital models in mixed reality environment, it will be easier for pupils and the surgeons to perform a maximum organ learning.