Nyayu Latifah Husni
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Nyayu Latifah Husni has a total of 82 co-authors. Here are the top co-authors by number of shared publications:
- Ade Silvia Handayani
Politeknik Negeri Sriwijaya
25 shared publications - Irsyadi Yani
Department of Mechanical Engineering, Faculty of Engineering, Universitas Sriwijaya, Indralaya 30662, South Sumatra, Indonesia
14 shared publications - Siti Nurmaini
Universitas Sriwijaya
14 shared publications Ekawati Prihatini
4 shared publicationsEvelina
Electrical Department, Politeknik Negeri Sriwijaya
4 shared publicationsSabilal Rasyad
Program Studi Teknik Elektronika, Politeknik Negeri Sriwijaya, Indonesia
4 shared publications
Latest Publications
- Open Access
- 15 July 2026
- Published by Universitas PGRI Palembang in JURNAL AMPERE
- Vol. 11 (1) , 12-24
Artificial Intelligence for Detecting Illegal Waste Dumping in River
This study developed a river waste detection system using the YOLOv8 algorithm to recognize human and waste objects in real time. The dataset was obtained from CCTV recordings of the Sekanak River, then underwent annotation and augmentation processes to improve the model's resilience to variations in lighting and environmental conditions. The model was trained with several configurations, and the best results were obtained at 100 epochs with a mAP@0.5 of 0.88 and stable validation. The system was testedon 10 daytime and nighttime image samples to assess its detection performance in the field. Based on the testing, the system detected human objects with a precision of 90.9% and trash with a precision of 81.8%, with an overall accuracy of 86.3%. Although the results were quite accurate, detection at night showed increased noise due to minimal light and water reflections, which could lead to false detection. Overall, the application of this technology can be a solution to support automatic and sustainable river monitoring ina smart environmental monitoring system. However, improvements in dataset quality and methods for handling visual disturbances are still needed to achieve more optimal.
- 15 July 2026
- Published by Universitas PGRI Palembang in JURNAL AMPERE
- Vol. 11 (1) , 12-24
Open AccessArtificial Intelligence for Detecting Illegal Waste Dumping in River
This study developed a river waste detection system using the YOLOv8 algorithm to recognize human and waste objects in real time. The dataset was obtained from CCTV recordings of the Sekanak River, then underwent annotation and augmentation processes to improve the model's resilience to variations in lighting and environmental conditions. The model was trained with several configurations, and the best results were obtained at 100 epochs with a mAP@0.5 of 0.88 and stable validation. The system was testedon 10 daytime and nighttime image samples to assess its detection performance in the field. Based on the testing, the system detected human objects with a precision of 90.9% and trash with a precision of 81.8%, with an overall accuracy of 86.3%. Although the results were quite accurate, detection at night showed increased noise due to minimal light and water reflections, which could lead to false detection. Overall, the application of this technology can be a solution to support automatic and sustainable river monitoring ina smart environmental monitoring system. However, improvements in dataset quality and methods for handling visual disturbances are still needed to achieve more optimal.
- Open Access
- 1 May 2026
- Published by Universitas Serang Raya in Wikrama Parahita : Jurnal Pengabdian Masyarakat
- Vol. 10 (1) , 21-30
Strengthening Small-Scale Snakehead (Channa Striata) Aquaculture through the Implementation of the IoT-Based “Channa Sense” Monitoring System
Small-scale snakehead (Channa striata) farmers commonly rely on manual and periodic water quality monitoring, which often results in delayed responses to environmental fluctuations and high fry mortality rates. This community service program aimed to strengthen technological literacy and improve hatchery management practices through the implementation of an Internet of Things (IoT)-based “Channa Sense” real-time monitoring system. The intervention adopted a structured three-phase approach consisting of pre-implementation assessment, participatory workshop and system installation, and post-implementation evaluation. The program involved 37 participants representing farmers, entrepreneurs, and community members, with baseline data collected from 30 small-scale farmers across Palembang, Indralaya, and Musi Banyuasin. Pre-intervention findings showed that 56.76% of participants were unaware of the technology and none had prior experience with digital monitoring systems. Following experiential learning activities, 64.86% of participants reported a moderate to full understanding of the system, and recognition of Channa Sense as a water quality monitoring device increased from 13.51% to 60.42%. At the partner hatchery (Kandang Om Bobby), real-time monitoring reduced fry mortality from approximately 40% to 5–10%, representing a survival improvement of 30–35%. The findings indicate that participatory and context-adapted IoT interventions can effectively bridge digital literacy gaps while generating measurable operational benefits in small-scale aquaculture. However, adoption intentions remained moderate due to cost and maintenance concerns. Continued mentoring, cost optimization, and cooperative-based implementation strategies are recommended to ensure long-term sustainability and broader community uptake
- 1 May 2026
- Published by Universitas Serang Raya in Wikrama Parahita : Jurnal Pengabdian Masyarakat
- Vol. 10 (1) , 21-30
Open AccessStrengthening Small-Scale Snakehead (Channa Striata) Aquaculture through the Implementation of the IoT-Based “Channa Sense” Monitoring System
Small-scale snakehead (Channa striata) farmers commonly rely on manual and periodic water quality monitoring, which often results in delayed responses to environmental fluctuations and high fry mortality rates. This community service program aimed to strengthen technological literacy and improve hatchery management practices through the implementation of an Internet of Things (IoT)-based “Channa Sense” real-time monitoring system. The intervention adopted a structured three-phase approach consisting of pre-implementation assessment, participatory workshop and system installation, and post-implementation evaluation. The program involved 37 participants representing farmers, entrepreneurs, and community members, with baseline data collected from 30 small-scale farmers across Palembang, Indralaya, and Musi Banyuasin. Pre-intervention findings showed that 56.76% of participants were unaware of the technology and none had prior experience with digital monitoring systems. Following experiential learning activities, 64.86% of participants reported a moderate to full understanding of the system, and recognition of Channa Sense as a water quality monitoring device increased from 13.51% to 60.42%. At the partner hatchery (Kandang Om Bobby), real-time monitoring reduced fry mortality from approximately 40% to 5–10%, representing a survival improvement of 30–35%. The findings indicate that participatory and context-adapted IoT interventions can effectively bridge digital literacy gaps while generating measurable operational benefits in small-scale aquaculture. However, adoption intentions remained moderate due to cost and maintenance concerns. Continued mentoring, cost optimization, and cooperative-based implementation strategies are recommended to ensure long-term sustainability and broader community uptake
- Open Access
- 1 May 2025
- Published by Atlantis Press SARL
- p. 374-384
Machine Learning Application of Flood Detecting and Mapping in Urban Area with Geographic Information Data
- 1 May 2025
- Published by Atlantis Press SARL
- p. 374-384
Open AccessMachine Learning Application of Flood Detecting and Mapping in Urban Area with Geographic Information Data
- Open Access
- 27 February 2024
- Published by Atlantis Press SARL
- p. 303-310
Design And Implementation of Smart Infusion Pump For Telemedicine
- 27 February 2024
- Published by Atlantis Press SARL
- p. 303-310
Open AccessDesign And Implementation of Smart Infusion Pump For Telemedicine
- Open Access
- 1 October 2023
- Published by Department of Computer Engineering, Faculty of Computer Science, Universitas Sriwijaya in Computer Engineering and Applications Journal
- Vol. 12 (3) , 181-188
Littering Activities Monitoring using Image Processing
Littering is a human behavior that become a habit since childhood. Even though there are rules that prohibit this behavior, the community still continues to do so. In order to limit this bad behavior, a device that can monitor and provide notifications is needed. In this research, a device is offered that can identify human activities in real time using webcam-based image processing. Then, it is processed by machine learning using the Recurrent Neural Network (RNN). The monitoring device produced in this research works by comparing the captured image data with a dataset. The captured image data will then be extracted features and form several coordinate points on the human body, then the system will classify these human activities into the category of normal activities or littering activities. This device will provide an output in the form of a warning every time the activity of littering is detected.
- 1 October 2023
- Published by Department of Computer Engineering, Faculty of Computer Science, Universitas Sriwijaya in Computer Engineering and Applications Journal
- Vol. 12 (3) , 181-188
Open AccessLittering Activities Monitoring using Image Processing
Littering is a human behavior that become a habit since childhood. Even though there are rules that prohibit this behavior, the community still continues to do so. In order to limit this bad behavior, a device that can monitor and provide notifications is needed. In this research, a device is offered that can identify human activities in real time using webcam-based image processing. Then, it is processed by machine learning using the Recurrent Neural Network (RNN). The monitoring device produced in this research works by comparing the captured image data with a dataset. The captured image data will then be extracted features and form several coordinate points on the human body, then the system will classify these human activities into the category of normal activities or littering activities. This device will provide an output in the form of a warning every time the activity of littering is detected.
- Open Access
- 30 September 2023
- Published by Poltekkes Kemenkes Surabaya in International Journal of Advanced Health Science and Technology
- Vol. 3 (4) , 274-282
Revolutionizing IV Infusions: Empowering Care with the DripControl+ App for Real-Time Monitoring and Precision Management
The demand for an effective and precise monitoring and control system for intravenous infusion therapy has increased due to concerns regarding medication errors and inefficiencies associated with current manual monitoring methods employed by nurses, particularly when caring for multiple patients across different rooms. This research aims to enhance intravenous infusion therapy by developing a real-time monitoring and control system. The system utilizes IoT technology and advanced sensors, including the load cell sensor for infusion volume detection, the FC-33 optocoupler sensor for precise infusion drop monitoring, and also a servo motor as an actuator to bend the infusion hose. Integrated with the NodeMCU ESP32 microcontroller, the system empowers healthcare professionals with the user-friendly DripControl+ app to remotely monitor and cont rol the infusion process. The results indicate a seamless collaboration among the system components. The FC-33 Optocoupler sensor exhibits an outstanding accuracy rate of 99.39%. The load cell sensor achieves an impressive 99.61% accuracy. The servo motor precisely follows predetermined positions. These outcomes effectively highlight the system's ability to accurately control the infusion drip rate with exceptional precision. The FC-33 optocoupler sensor and servo motor play crucial roles in achieving this accuracy. With an impressive average accuracy of 97.99%, the system has proven to be highly efficient. However, it should be noted that sudden changes in infusion speed could impact the accuracy of the readings. The future research could focus on refining the system's ability to respond to abrupt changes in infusion speed through advanced algorithms, such as machine learning.
- 30 September 2023
- Published by Poltekkes Kemenkes Surabaya in International Journal of Advanced Health Science and Technology
- Vol. 3 (4) , 274-282
Open AccessRevolutionizing IV Infusions: Empowering Care with the DripControl+ App for Real-Time Monitoring and Precision Management
The demand for an effective and precise monitoring and control system for intravenous infusion therapy has increased due to concerns regarding medication errors and inefficiencies associated with current manual monitoring methods employed by nurses, particularly when caring for multiple patients across different rooms. This research aims to enhance intravenous infusion therapy by developing a real-time monitoring and control system. The system utilizes IoT technology and advanced sensors, including the load cell sensor for infusion volume detection, the FC-33 optocoupler sensor for precise infusion drop monitoring, and also a servo motor as an actuator to bend the infusion hose. Integrated with the NodeMCU ESP32 microcontroller, the system empowers healthcare professionals with the user-friendly DripControl+ app to remotely monitor and cont rol the infusion process. The results indicate a seamless collaboration among the system components. The FC-33 Optocoupler sensor exhibits an outstanding accuracy rate of 99.39%. The load cell sensor achieves an impressive 99.61% accuracy. The servo motor precisely follows predetermined positions. These outcomes effectively highlight the system's ability to accurately control the infusion drip rate with exceptional precision. The FC-33 optocoupler sensor and servo motor play crucial roles in achieving this accuracy. With an impressive average accuracy of 97.99%, the system has proven to be highly efficient. However, it should be noted that sudden changes in infusion speed could impact the accuracy of the readings. The future research could focus on refining the system's ability to respond to abrupt changes in infusion speed through advanced algorithms, such as machine learning.
- 14 February 2022
- Published by Atlantis Press SARL
- p. 348-354
Application of Health Detector
- Open Access
- 1 April 2020
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 1500 (1)
Implementation of DSK TMS320C6416T Module in Modified Stethoscope for Lung Sound Detection
This paper presents a design of digital stethoscope that is modified using DSK TMS320C6416T. This modified stethoscope is proposed for helping the doctor to make a true decision and the same perception on the symptoms of someone’s disease. Using the conventional stethoscope usually gives drawback, such as the doctors usually makes misinterpretation on the auscultation. The reason is because of the errors that occur due to the difference on the sensitivity of the ear and the experiences of the doctor. This research offers a DSK TMS 320C6416T as an instrumentation system for identification and analysis on the lung sound. The modified design of the digital stethoscope that is proposed in this research gives benefit to the users because it provides the result in the form of the signals that can be monitored in the PC. The results of this designed stethoscope showed the interpretation of graphs in the form of the spectrogram and scalogram which will help the simplification on the analysis of lung sounds as well as to minimize or avoid the limitations of hearing problems in the process of auscultation. The method used in lung sound analysis in this research includes Fourier Transformation that is then developed into a Short-Time Fourier transform (STFT) and continued to the Wavelet Transform. Based on the test data for normal lung sounds, it was found that the working frequency range was 90 to 1265 Hz. The adventitious sounds were found with frequency range from 60 to 1980 Hz. In this research, it was also found that the lung sounds were at the frequency range from 90 to 1015 Hz for direct measurement to patients.
- 1 April 2020
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 1500 (1)
Open AccessImplementation of DSK TMS320C6416T Module in Modified Stethoscope for Lung Sound Detection
This paper presents a design of digital stethoscope that is modified using DSK TMS320C6416T. This modified stethoscope is proposed for helping the doctor to make a true decision and the same perception on the symptoms of someone’s disease. Using the conventional stethoscope usually gives drawback, such as the doctors usually makes misinterpretation on the auscultation. The reason is because of the errors that occur due to the difference on the sensitivity of the ear and the experiences of the doctor. This research offers a DSK TMS 320C6416T as an instrumentation system for identification and analysis on the lung sound. The modified design of the digital stethoscope that is proposed in this research gives benefit to the users because it provides the result in the form of the signals that can be monitored in the PC. The results of this designed stethoscope showed the interpretation of graphs in the form of the spectrogram and scalogram which will help the simplification on the analysis of lung sounds as well as to minimize or avoid the limitations of hearing problems in the process of auscultation. The method used in lung sound analysis in this research includes Fourier Transformation that is then developed into a Short-Time Fourier transform (STFT) and continued to the Wavelet Transform. Based on the test data for normal lung sounds, it was found that the working frequency range was 90 to 1265 Hz. The adventitious sounds were found with frequency range from 60 to 1980 Hz. In this research, it was also found that the lung sounds were at the frequency range from 90 to 1015 Hz for direct measurement to patients.
- Open Access
- 1 April 2020
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 1500 (1)
Analysis Model for Orifice Flow Meter Correction Factor in Measuring in-Pipe Natural Gas Flow based on Numerical Simulation
Orifice Flow meter (OFM) is widely used for measuring the flow in industry. However, in many cases, the measurement value of OFM some times will differ from the other meter, for instance Ultrasonic Meter (USM), that also measure the flow of the same pipe. The inaccuracy of the flow measurement using OFM can be caused by: i). the geometry of the OFM hole that experiences the shape changes due to abrasion or deformation as the result of the gas flow pressure on the inner surface of OFM metal holes; ii). the presence of a liquid fraction in the measurement of gas flow. To match the value measurement between the OFM and USM, a numerical simulation process of natural gas flow through OFM is needed. In this research, an analysis model for Orifice Flow Meter Correction Factor in Measuring in-Pipe Natural Gas Flow based on Numerical Simulation, is proposed. This model can be used to correct the error measurement of the OFM. The analysis process is conducted by analysing the Pearson correlation and testing the Kolmogorov-Smirnov data normalization on 9 populations of flow measurement data using OFM and USM. However, only 4 of the data that matches to the Pearson correlation criteria and the Kolmogorov-Smirnov normalization. Based on the 4 populations of data, a correction factor model was developed using a regression analysis consisting of two factors, namely OFM and USM correction factors that refer to the actual data from the measurement of gas flow rate. Based on variations in gauge pressure of 393.65 psig, 404.67 psig and 413.37 psig at a temperature of 80.710F and numerical simulation results in the form of different pressure and flow discharges, the resulting correction factors are in the form of 3rd degree polynomial equations.
- 1 April 2020
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 1500 (1)
Open AccessAnalysis Model for Orifice Flow Meter Correction Factor in Measuring in-Pipe Natural Gas Flow based on Numerical Simulation
Orifice Flow meter (OFM) is widely used for measuring the flow in industry. However, in many cases, the measurement value of OFM some times will differ from the other meter, for instance Ultrasonic Meter (USM), that also measure the flow of the same pipe. The inaccuracy of the flow measurement using OFM can be caused by: i). the geometry of the OFM hole that experiences the shape changes due to abrasion or deformation as the result of the gas flow pressure on the inner surface of OFM metal holes; ii). the presence of a liquid fraction in the measurement of gas flow. To match the value measurement between the OFM and USM, a numerical simulation process of natural gas flow through OFM is needed. In this research, an analysis model for Orifice Flow Meter Correction Factor in Measuring in-Pipe Natural Gas Flow based on Numerical Simulation, is proposed. This model can be used to correct the error measurement of the OFM. The analysis process is conducted by analysing the Pearson correlation and testing the Kolmogorov-Smirnov data normalization on 9 populations of flow measurement data using OFM and USM. However, only 4 of the data that matches to the Pearson correlation criteria and the Kolmogorov-Smirnov normalization. Based on the 4 populations of data, a correction factor model was developed using a regression analysis consisting of two factors, namely OFM and USM correction factors that refer to the actual data from the measurement of gas flow rate. Based on variations in gauge pressure of 393.65 psig, 404.67 psig and 413.37 psig at a temperature of 80.710F and numerical simulation results in the form of different pressure and flow discharges, the resulting correction factors are in the form of 3rd degree polynomial equations.
- Open Access
- 1 April 2020
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 1500 (1)
Modified Design of Water Metering System
The analog flow meter has been widely used in Regional Water Supply Company (PDAM) to measure the amount of water flow consumed by PDAM customer. This analog flow meter has an inadequacy in converting water consumed into rupiah billing system. In this research, a modified design of water metering system was proposed. A system involving water flow sensor to detect water flow velocity in the form of pulses as the inputs of microcontroller was analyzed. The output of the flow sensor will be converted as information of how much water consumed and how much money (in Rupiah Currency) should be paid to the water company. These outputs will be delivered through SMS using SIM 800L. The result of the experiments show that the proposed modified design of the water metering system could fulfill the standard of the metering system.
- 1 April 2020
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 1500 (1)
Open AccessModified Design of Water Metering System
The analog flow meter has been widely used in Regional Water Supply Company (PDAM) to measure the amount of water flow consumed by PDAM customer. This analog flow meter has an inadequacy in converting water consumed into rupiah billing system. In this research, a modified design of water metering system was proposed. A system involving water flow sensor to detect water flow velocity in the form of pulses as the inputs of microcontroller was analyzed. The output of the flow sensor will be converted as information of how much water consumed and how much money (in Rupiah Currency) should be paid to the water company. These outputs will be delivered through SMS using SIM 800L. The result of the experiments show that the proposed modified design of the water metering system could fulfill the standard of the metering system.