Claim Missing Document
Check
Articles

Found 7 Documents
Search
Journal : STRING (Satuan Tulisan Riset dan Inovasi Teknologi)

Deteksi Kelayuan Pada Bunga Mawar dengan Metode Transformasi Ruang Warna HSI Dan HSV Dede Wandi; Fauziah Fauziah; Nur Hayati
STRING (Satuan Tulisan Riset dan Inovasi Teknologi) Vol 5, No 3 (2021)
Publisher : Universitas Indraprasta PGRI Jakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (848.265 KB) | DOI: 10.30998/string.v5i3.8464

Abstract

The rose is a plant from the genus Rosa which has more than 100 species with various colors. In the process of selecting roses, you will find roses that are still fresh and wilted. With that, we can detect the wilting of the rose by applying the HSI and HSV methods to image processing applications, the data collection process, namely by making data preparations on the Kaggle dataset, then classifying and training the data using the HSI and HSV methods. Based on the classification results of a total of 820 images of rose images, a total of 757 images were tested using HSI and HSV. The values obtained were Range at HSI, H = 0–0.5, S = 0–1, and I = 0.5372549–1 in the Fresh category, while the category HSI wilt, H = 0–0.5, S = 0-1, I = 0.5620915–1. The HSV range values are in the Fresh category H = 0–0.5, S = 0-1, V = 0-1, and the Wilt category H = 0-0.5, S = 0-1, V = 0-1. Furthermore, the success rate for testing roses with HSI reached 92.3% where the data read correctly 757 and read incorrectly 63 out of 820 sample data of roses, while testing on HSV the success rate reached 93.2% where the data read correctly 765 and read incorrectly 55 out of 820 rose flower sample data. Based on the above results, detection of wilting roses using the HSV color space transformation method is the best in data testing.
Implementasi Metode Background Subtraction dan Morfologi untuk Mendeteksi Objek Bergerak Pada Video Dede Saptoadi; Fauziah Fauziah; Nur Hayati
STRING (Satuan Tulisan Riset dan Inovasi Teknologi) Vol 5, No 2 (2020)
Publisher : Universitas Indraprasta PGRI Jakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (776.836 KB) | DOI: 10.30998/string.v5i2.7743

Abstract

Monitoring systems are a rapidly increasing need in almost all lines of public, such as roads, parks, buildings, terminals and many places that have a monitoring system. Detection adapts learning of moving objects in video observation to obtain results such as the human sense of sight. By detecting objects in the monitoring system, we can determine the frame that shows the position of the moving object. The methods that can be used to de-tect objects are background subtraction and morphology, these methods are considered appro-priate to be implemented in the designed program. The research method used is divided into two stages, namely data collection to collect CCTV video samples and search for related research references. Then the stages that the program goes through include frame extraction, implemen-tation of background subtraction, conversion of grayscale images and converting them into bi-nary form, performing opening morphological operations, making masking and implementing them into video. From the test results the program has a success rate of 93.3% of the test with bright lighting and 83.3% of the test with dim lighting.
Smart Trash Menggunakan Sensor Ultrasonik dan Arduino Uno Berbasis IoT Imam Pambudi Utomo; Fauziah Fauziah; Nur Hayati
STRING (Satuan Tulisan Riset dan Inovasi Teknologi) Vol 6, No 3 (2022)
Publisher : Universitas Indraprasta PGRI Jakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (437.805 KB) | DOI: 10.30998/string.v6i3.11986

Abstract

The Covid-19 pandemic has forced us to adhere to health protocols. Thus, making people more careful not to come into direct contact, one of which is by not opening and closing trash cans that are often used together with many people. Sometimes people are also lazy to open the lid of the trash can and let the garbage sit on the lid of the trash can so that it is scattered and can become a nest of all kinds of diseases. The purpose of this research is to make the smart trash can open and close automatically and find out the available capacity in the smart trash. The ultrasonic sensor has a minimum and maximum limit with a sensor range of 0-30 cm detecting people the smart trash will automatically open, if it exceeds the range limit the smart trash will not open. If it is detected by the sensor but the smart trash is full, it will not open. In the range of the smart trash height of 25cm to determine the status of the smart trash capacity on the LCD, if the smart trash capacity is 10-100% then it is available, if the LCD status is less than 10% then it is full. The Fuzzy Logic algorithm is used to determine the rules for the level of distance accuracy (cm). The resulting value is a distance of 17.5- 25cm status is available, if a distance of 10-17.5cm status is available, if a distance of 2.5-10cm status is available, if a distance is 0-2 ,5cm full state. On the accuracy of the Naïve Bayes algorithm using rapidminer software that the accuracy of the success of the servo motor is 87.33%.
Sistem Informasi Pemilihan Asisten Laboratorium dengan Metode Weighted Product dan Weighted Sum Model Nur Hayati; Sri Rahayu; Tri Ichsan Saputra
STRING (Satuan Tulisan Riset dan Inovasi Teknologi) Vol 6, No 1 (2021)
Publisher : Universitas Indraprasta PGRI Jakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (960.742 KB) | DOI: 10.30998/string.v6i1.8455

Abstract

The process of selecting assistants at a faculty is generally based on requirements such as GPA, year of class, even some are accepted subjectively because of the direct assessment of a lecturer who is also the head of the laboratory. This form of election isn’t problematic as long as the student is indeed worthy of being an assistant. However, the documentation of such an admissions system is often not traceable. Therefore, an application for the assistant selection process is indispensable to make the process more documented and objective. Based on the conditions for accepting assistants that have previously been implemented, the Weighted Product decision support system method can be used in the weighting process because it is more specific to the value weights in each criterion and the Weighted Sum Model can be used for the ranking process so that the decision results can be precise, fast and objective. The results obtained were 100% test from the interface side and then the results of the study were in the form of a website-based information system for selecting laboratory assistants based on the criteria for GPA, practical course scores, specificity course scores and question scores.
Aplikasi Spoxtech Untuk Penyandang Tuna Rungu – Wicara Menggunakan Algoritma Hidden Markov Model dan Metode Finite State Automata (FSA) Destivanesha Rina; Fauziah Fauziah; Nur Hayati
STRING (Satuan Tulisan Riset dan Inovasi Teknologi) Vol 5, No 3 (2021)
Publisher : Universitas Indraprasta PGRI Jakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (941.588 KB) | DOI: 10.30998/string.v5i3.7690

Abstract

Communication is a very important thing to establish social relations. In communicating, deaf and mute persons have difficulty conveying what information they want to convey. The Android-based Spoxtech application is an application specially designed for deaf and speech impaired persons. As the name implies, Spoxtech is taken from the word speech to text - text to speech. This application can make it easier for users to communicate because it is equipped with a voice menu that functions to detect the spoken voice which is then converted into text and a typed menu which functions to detect writing which is then converted into voice form. This application was created using the Hidden Markov Model algorithm in generating sound into text and using the finite state automata method in generating text into sound by performing two-level decapitation. This results of this study were carried out by testing the whitebox and blackbox. In this research, the text normalization test on the typewriter menu was also carried out, namely each word containing currency, time, temperature, and number units. The accuracy value obtained in testing this application uses Indonesian by testing the voice menu and typing menu and testing the two-level decapitation, the value is 100% of the 300 words in Indonesian.
uSocial Realtime Berbasis Android Menggunakan Volley dan Algoritma BruteForce Azizah Azizah; Fauziah Fauziah; Nur Hayati
STRING (Satuan Tulisan Riset dan Inovasi Teknologi) Vol 5, No 2 (2020)
Publisher : Universitas Indraprasta PGRI Jakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (714.259 KB) | DOI: 10.30998/string.v5i2.7751

Abstract

Communication can be done anywhere and anytime, making it easier for people to exchange information. Communicating today is very easy by utilizing various applications or media by using internet facilities such as Instagram, Facebook, WhatsApp Messenger, and others. The author builds an application that can connect users in communicating, exchanging information, and sharing the science that is the uSocial application. The app is created by applying the Firebase, Volley, and RecyclerView methods. These three methods are useful for running applications directly or In Real-time. Algorithm testing is done by comparing the Brute Force algorithm with the Horspool, and Knuth Morris Pratt algorithms. The results of Brute Force algorithm testing on word searches performed based on the length of search sentences resulting in 100% accuracy for 500 data trials as well as shortening the time for posts searches then better using Brute Force algorithms. The test results using the Whitebox method tested using three parameters, namely cyclomatic complex city, region, and independent path, produce 17. The program's flow and logic follow ANSI standards and there is no need to change the flow or plan back. At the same time, the results on the Blackbox testing of the functions in the application managed to produce the desired output.
Analisis Sentimen Opini Masyarakat Terhadap Kuliner DKI Jakarta dengan Metode Naïve Baiyes dan Support Vector Machine Muhammad Fadel Sasongko; Fauziah Fauziah; Nur Hayati
STRING (Satuan Tulisan Riset dan Inovasi Teknologi) Vol 7, No 3 (2023)
Publisher : Universitas Indraprasta PGRI Jakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30998/string.v7i3.13931

Abstract

Culinary matter is an integral part of people’s everyday lives, especially for those who like to taste a variety of flavors and uniqueness of different food. It is easier to find information about the culinary you want to try through sites and social media. With the information, visitors can get an idea of the taste, price, and location of culinary treats. In the DKI Jakarta area, social media play an important role in providing data on people's assessments of culinary treats. Considering the use of social media platforms such as Instagram, Facebook, and Twitter, the researchers apply the methods of Naive Bayes and Support Vector Machine that later produce a result that can be used as a basis for assessing culinary treats in Jakarta. This research takes data from three sources, namely Twitter, Zomato and Quora. The data collection period for Twitter is from March 01, 2022-June 30, 2022, while for Zomato and Quora from September 2021 to June 2022, during which 2520 data are collected. The data will be then processed to draw a conclusion. The accuracy test on three sources show the accuracy score of Naïve Bayes is higher than that of the Support Vector Machine, with the accuracy of Naïve Bayes of 76.00% and that of the support vector machine of 74.00%, resulting in more positive responses than negative responses and indicating that culinary treats in Jakarta is considered good by the community.