Average Accuracy Level Research Articles

R-SABMNet: A YOLOv8-Based Model for Oriented SAR Ship Detection with Spatial Adaptive Aggregation

Synthetic Aperture Radar (SAR) is extensively utilized in ship detection due to its robust performance under various weather conditions and its capability to operate effectively both during the day and at night. However, ships in SAR images exhibit various characteristics including complex land scattering interference, variable scales, and dense spatial arrangements. Existing algorithms are insufficient in effectively addressing these challenges. To enhance detection accuracy, this paper proposes the Rotated model with Spatial Aggregation and a Balanced-Shifted Mechanism (R-SABMNet) built upon YOLOv8. First, we introduce the Spatial-Guided Adaptive Feature Aggregation (SG-AFA) module, which enhances sensitivity to ship features while suppressing land scattering interference. Subsequently, we propose the Balanced Shifted Multi-Scale Fusion (BSMF) module, which effectively enhances local detail information and improves adaptability to multi-scale targets. Finally, we introduce the Gaussian Wasserstein Distance Loss (GWD), which effectively addresses localization errors arising from angle and scale inconsistencies in dense scenes. Our R-SABMNet outperforms other deep learning-based methods on the SSDD+ and HRSID datasets. Specifically, our method achieves a detection accuracy of 96.32%, a recall of 93.13%, and an average level of accuracy of 95.28% on the SSDD+ dataset.

Implementation and analysis of hybrid communication for monitoring and control for android-based smart farming

Farmers are forced to water each crop individually or wait to switch off the water pump. to avoid this, water management needs to be optimized—that is, it needs to be done on schedule, in the appropriate quantity, and throughout a wide area in an effective manner. That is hybrid communication between online and offline communication. To run this system will use 3 nodes, for details 1 node in the irrigation section, 1 node in the monitoring section of land and soil quality, and 1 node in the section to control data or can be said to be a semi-server to give commands to 2 nodes offline. To see if the system still communicates with each other and runs according to the block diagram that has been designed, an Android application will be made to monitor the system. Where if there is a failed data update or data change in the results of the automation system, a notification will come out to check the system as a whole. To support this research, several sensor calibrations were carried out which resulted in an average accuracy level above 90 percent. For the overall system test, 10 experiments were carried out at different times to see the reliability of the system

Long-term forecasting for climate classification using hybrid backpropagation relevance vector machine algorithm

Abstract Climate forecasting and classification are essential in planning cropping patterns. Therefore, the purpose of this study is to evaluate the condition of hydro-climatological data for the last 10 years and determine the trend of the data and determine the climate classification. The forecasting process utilizes a long-term forecasting method based on a hybrid backpropagation-relevance vector machine (BP-RVM) algorithm. The architecture parameters consist of three hidden layers (36-73-38-19-1), learning rate 0.1, momentum 0.9, and RBF gamma of 0.01. The forecasting results show that the BP-RVM algorithm has an average accuracy level with a category of “high accurate forecasting”. The climate classification in the Central Lombok region, Indonesia is D4 climate (slightly dry category). The increase in rainfall volume and temperature from year to year has a strong correlation with a decrease in air humidity, wind speed, and length of sunshine. The results of this study can be utilized as a basis for planning cropping patterns and handling the water needs of crops grown in the dry season. Therefore, farmers can determine the amount of water needed from dams or watersheds due to evaporation that occurs during the growing season.

A Novel ML-Driven Approach to Enhance CRN Performance under Varying Network Parameters

This paper explores RL and DRL techniques for spectrum allocation in the context of CRNs, with consideration of difficulties like spectrum utilization and network performance in changing conditions. The proposed improved spectrum management model integrates RL with model-based prediction as a way of improving decision making. The results of the experiment prove that the identified approach allows for achieving an average level of accuracy of 96%, and a loss rate of 0.20, as well as of precision of 92% to 0.95. In addition, recall was extended from 0.85 to 0.90, and the F1 score was at 0.90, which indicated that the model demonstrated satisfactory performance at both precise and recall. The proposed algorithm outperformed existing machine learning models with a 96% accuracy, a low loss of 0.20, and an F1 Score of 0.90, showcasing superior reliability and adaptability. Based on these outcomes, it can be concluded that the proposed hybrid RL model is more effective in predicting the next available spectrum and more adaptable to the changes in the CRN environment than the single RL method thus, the proposed solution is suitable for real-time spectrum allocation in CRNs.

Knowledge and practical skills for cancer pain management among nurses on remote islands in Japan and related factors nationwide.

To clarify the knowledge and practical skills needed for cancer pain management among nurses on remote islands in Japan and related factors nationwide. Due to geographical factors, nurses working on remote islands in Japan have few opportunities to attend training programs, which makes it difficult to acquire the knowledge and practical skills needed to provide pain management for patients with cancer. We conducted a self-administered questionnaire survey regarding knowledge and practical skills in pain management for patients with cancer. Cross-sectional study. Nurses working in cancer pain care in medical facilities and home care on remote islands throughout Japan. We analysed 128 responses. Regarding knowledge, the average accuracy level was 49.1%. Items with a low accuracy rate included selecting medicine according to the type of pain and the patient's condition. Regarding practice, the items with low scores included analgesics appropriate for the type of pain and relating physical pain to mental, social and spiritual aspects. The most common significant factor in both knowledge and practice was related to postgraduate training. These findings suggest that to improve the knowledge and practical skills for cancer pain management among nurses on remote islands in Japan, it is necessary to incorporate clinical reasoning into basic education and establish remote education systems and consultation systems with other facilities.

Transfer Learning Model Application for Rastrelliger brachysoma and R. kanagurta Image Classification Using Smartphone-Captured Images

Prior aquatic animal image classification research focused on distinguishing external features in controlled settings, utilizing either digital cameras or webcams. Identifying visually similar species, like Short mackerel (Rastrelliger brachysoma) and Indian mackerel (Rastrelliger kanagurta), is challenging without specialized knowledge. However, advancements in computer technology have paved the way for leveraging machine learning and deep learning systems to address such challenges. In this study, transfer learning techniques were employed, utilizing established pre-trained models such as ResNet50, Xception, InceptionV3, VGG19, VGG16, and MobileNetV3Small. These models were applied to differentiate between the two species using raw images captured by a smartphone under uncontrolled conditions. The core architecture of the pre-trained models remained unchanged, except for the removal of the final fully connected layer. Instead, a global average pooling layer and two dense layers were appended at the end, comprising 1024 units and by a single unit, respectively. To mitigate overfitting concerns, early stopping was implemented. The results revealed that, among the models assessed, the Xception model exhibited the most promising predictive performance. It achieved the highest average accuracy levels of 0.849 and 0.754 during training and validation, surpassing the other models. Furthermore, fine-tuning the Xception model by extending the number of epochs yielded more impressive outcomes. After 30 epochs of fine-tuning, the Xception model demonstrated optimal performance, reaching an accuracy of 0.843 and displaying a 11.508% improvement in predictions compared to the model without fine-tuning. These findings highlight the efficacy of transfer learning, particularly with the Xception model, in accurately distinguishing visually similar aquatic species using smartphone-captured images, even in uncontrolled conditions.

Penerapan Teknologi Augmented Reality Untuk Aplikasi Pengukuran Panjang Suatu Objek

Length measurement is an activity to determine the length value of an object, surface, or other object. The measurement process generally requires conventional length measuring tools such as a ruler or ruler, tape measure, or calipers. Length measurements will experience problems when conditions do not allow the use of conventional measuring instruments due to position factors, the size of objects or surfaces that are too large and difficult for one person to reach, or differences in results obtained when measuring using non-standard units which cause uncertainty the measuring value. Augmented Reality is a technology that allows users to see the real world combined with virtual objects without changing the real world as the main environment. This technology also allows the length measurement process to be carried out using virtual objects as measurement markers and provides real-time measurement results without using conventional length measuring instruments. The results obtained are a length measurement application that can run by relying on virtual objects in the form of pointers which are placed as markers and connected by lines to obtain information in the form of the value of the measurement results. With an average accuracy level of 98.03%, this length measurement application can be used as an alternative medium for carrying out the length measurement process.

Metoda ANFIS za klasifikaciju EEG signala pri zamišljanju pokreta različitih zglobova istog ekstremiteta

Introduction: The experimental area of brain-computer interfaces (BCIs) is expanding to include movement actions, which play a crucial part in deciphering cognitive processes. Without the need for any kind of exterior stimulation, motor imagining (MI) can be used as a powerful model for braincomputer interfaces (BCIs). A natural method of operating exterior devices is to imagine moving various joints in the same arm. These envisioned motions have similar spatial images in the motor brain, making it difficult to differentiate MI of various joints of the same leg based on EEG data. Method: A pre-existing data collection of 25 participants was utilized in this study. The participants visualized using their right limbs to carry out three different activities: visualize yourself manipulating your right hand, visualize bending your right arm, and close your eyes while you relax. To assign categories to these impulses, we turned to the adaptive neuro-fuzzy reasoning system. Results: The average level of accuracy was 90%. Conclusion: The findings demonstrate that this technique is crucial for correctly categorizing EEG data. The data collection used in this investigation consists of EEG measurements of the same limb used in muscular imaging. The new categorization method will be applied to these signals to draw conclusions.

Evaluacija performansi Veri-Q hemoglobinskog merača za procenu hemoglobina i hematokrita na mestu pružanja pomoći

Abstract: Introduction: Haemoglobin (Hb) and packed cell volume (PCV) are tests used in the assessment of anaemia. The Veri-Q haemoglobin meter is a new device in the point-of-care market used for the quantitative measurement of haemoglobin and packed cell volume. This study aimed to evaluate the performance of the Veri-Q haemoglobin meter for the assessment of haemoglobin and packed cell volume. Materials and Methods: Haemoglobin and packed cell volume estimations of one hundred and eleven participants were determined using the Veri-Q Red haemoglobin meter and statistically compared with those obtained from the conventional method (gold standard). Results: One hundred and eleven undergraduate students participated in this study, of whom 72 (64.9%) were females and 39 (35.1%) were males. The mean haemoglobin values calculated using the Veri-Q haemoglobin meter (11.97 ± 1.95 g/dl) were significantly lower than the values obtained with the conventional method (12.35 ± 1.66 g/dl) (t = 4.7549; p = 0.0001). Similar results were obtained with packed cell volume estimation. The performance indices of the Veri-Q haemoglobin meter were as follows: sensitivity (75.41% for Hb and 77.05% for PCV), specificity (28.0% for Hb and PCV), positive predictive value (PPV) (56.1% for Hb and 56.63% for PCV), negative predictive value (48.28% for Hb and 50% for PCV). The prevalence of anaemia in this study was 54.05%. Positive Likelihood Ratio (1.05 for Hb and 1.07 for PCV), Negative Likelihood Ratio (0.88 for Hb and 0.82 for PCV), Accuracy (54.05% for Hb and 54.95% for PCV). The Receiver Operative Characteristics (ROC) Area Under Curve (AUC) was 0.996 for Hb and 0.984 for PCV. A significant but weak positive correlation was found to exist between haemoglobin estimations using conventional and Veri-Q methods (r = 0.3013, P = 0.01); PCV (R = 0.2512, P = 0.07). The sensitivity of the device can be described as moderate while the specificity is low. The accuracy of the device is just average. Conclusion: The Veri-Q haemoglobin meter demonstrates an average level of accuracy and a high AUC, making it potentially useful for field epidemiological studies.

Dual Vegetation Index Analysis and Spatial Assessment in Kota Bharu, Kelantan using GIS and Remote Sensing

Vegetation serves as an indicator of environmental conditions in ecological classifications. In addition, vegetation index analysis can also benefit farmers and agricultural planners by optimising crop selection and irrigation practices. The spatial distribution of healthy vegetation can increase agricultural productivity. This study focuses on the Kota Bharu district in the state of Kelantan, Malaysia that aims to recognise the vegetation indices Normalized Difference Vegetation Index (NDVI) and Green Normalized Difference Vegetation Index (GNDVI). NDVI analysis measures reflected visible and near-infrared light to identify and evaluate living green plants. The Green Normalized Difference Vegetation Index (GNDVI) has a higher saturation threshold and is more sensitive to plant chlorophyll levels than NDVI. This approach works in agricultural environments with dense canopies or advanced crop development. The average accuracy level for NDVI 2023 is 78% while the average accuracy level for GNDVI 2023 is 76%. The value of kappa coefficient for NDVI and GNDVI for 2023 respectively are 0.73 and 0.72 which considered to be acceptable and represents the good correspondence.

CLASSIFICATION OF RICE QUALITY LEVELS BASED ON COLOR AND SHAPE FEATURES USING ARTIFICIAL NEURAL NETWORK BASED ON DIGITAL IMAGE PROCESSING

Rice is the staple food of most Indonesians. In identifying the quality of rice, it can be seen from physical characteristics such as the color and shape of rice, because these characteristics can make an object can be identified properly and clearly. In general, what is done in determining the quality of rice by looking at its color and shape. But usually the human eye in identifying objects is sometimes less accurate which is influenced by several factors, such as age. So, several studies were conducted that tried to solve the problem by using digital image processing. However, the accuracy results obtained are still not accurate, because the datasets used in the previous study were relatively small, namely around 80 images, although the average level of accuracy obtained was quite high, but the number of datasets used was very small so that the level of accuracy was still inaccurate. Therefore, in this study, it is proposed that the title of classification of rice quality levels using JST based on digital image processing which divides rice into 3 classifications, namely, good, good enough, and not good where in this study using 330 digital images to produce a more accurate level of accuracy. In this study, there are several stages, namely, image retrieval, preprocessing, segmentation, morphological, feature extraction, and classification using artificial neural networks. Based on the research conducted, training accuracy was produced with an average accuracy of 98,75% while the test accuracy was produced with an average accuracy of 98,89%.

Freshwater Monitoring System Design In Real-Time For Fish Cultivation

The increased demand for fish supply in aquaculture highlights the need for sophisticated and accurate monitoring systems to ensure optimal water quality. In this context, Internet of Things (IoT) technology and fuzzy logic have become promising solutions to improve efficiency and effectiveness in freshwater fish farming. This research aims to develop a real-time freshwater monitoring system that integrates IoT and fuzzy logic. This system will enable monitoring of critical parameters such as temperature, pH and water turbidity with a high degree of accuracy. Implementation of IoT sensors that are connected to one centralized network and use fuzzy logic to process the data obtained. The research also involved developing an intuitive user interface to manage the system. The developed system is able to provide real-time monitoring with a high level of accuracy. Users can easily access and analyze the generated data through the user interface provided. The design of monitoring freshwater conditions for intensive aquaculture can be prepared using pH sensor devices, turbidity sensors, and temperature sensors using ESP2866 modules to connect to IoT. From the results of 20 days of system trials, it was found that the system can determine the condition of fresh water with good readings with average level of accuracy for pH sensors of 97%, turbidity sensors of 92% and temperature sensors of 96%.
 
 

Online citizen science to develop local species identification skills and biodiversity awareness of prospective biology teacher students

Citizen Science (CS) currently involves massive public participation. CS activities have been transformed in various forms, one of which is through Online Citizen Science (OCS). This study aims to identify the outcomes of the OCS program on species identification skills (IS) and biodiversity awareness (KB) in prospective biology teacher students. This study used a survey method involving 30 research participants from two campuses in Majalengka and Cirebon. The research data were collected using non-test techniques in the form of analyzing the results of species identification and biodiversity awareness questionnaire. Data were analyzed qualitatively and described according to the findings. The results showed that OCS participants could identify species with an average level of accuracy in the good category. The level of biodiversity awareness in participants after participating in the OCS program was classified as very high. The OCS program can be an alternative in learning taxonomy that can increase awareness of biodiversity.

ANALISIS KEPUASAN KONSUMEN TERHADAP KUALITAS PRODUK MENGGUNAKAN ALGORITMA NAIVE BAYES

Current technological developments have brought changes to human lifestyles, such as lifestyle, selection of antic goods and communication equipment such as smartphones and other devices. Nowadays it is known that the collection of branded items is also a lifestyle trend such as bags, rings, clothes and other fashions. Batam as an industrial city encourages investors to invest in Batam, because Batam is the most strategic city and is an international trade route directly adjacent to Singapore. The problem that often occurs at PT Tas Import Sukses in serving consumers is that the goods sent are not on demand, so consumers suspect that the products traded do not have good quality. And if there is a problem, the company responds with a very slow time. The purpose of this study is to analyze consumer satisfaction with the quality of products at PT Tas Import Sukses using the Naïve Bayes method, to determine the level of consumer satisfaction with the quality of bag products. And the last is to find out the level of accuracy in determining product satisfaction and quality with the Naïve Bayes method. While the results of this study are that satisfaction is strongly influenced by the level of service, while the second order is product quality, where the average level of satisfaction accuracy using the naïve bayes algorithm is 91%, pression 89% and recall 88.33%.

Design and Develop an Automatic System for Monitoring Electrical Quality and Controlling Power Factor in Household Electrical Loads Using Capacitors

Excessive use of electronic equipment can cause problems with electricity quality and the power factor in household electrical loads. One way to overcome this problem is capacitor applied. Capacitors can be used to increase the power factor and improve the quality of electricity in household electrical loads. The test results without load showed an average power factor of 0.80, while the test results after repairs showed a power factor of 0.88. This research also evaluated the performance of voltage and current sensors integrated in the system. The current and voltage sensors can work well and are able to transmit data according to the actual values with an accuracy rate of 97.23%. The average accuracy level for current sensors reached 98.91% and 99.35%, while the accuracy level for voltage sensors obtained an average accuracy level of 99.95%.

Support Vector Machine Application for Classification of Tempe Fermentation Maturity with Information Gain Selection Feature

Tempe is one of the ingredients of traditional Indonesian cuisine. In making tempe, a soybean fermentation process is needed which is generally still carried out in an open environment so that the maturity time becomes slow and erratic. Therefore, in the tempe fermentation process, a detector is needed to find out optimal maturity in tempe. This detection effort makes it possible to use image processing by utilizing various feature extractions through the classification process. This research utilizes a variety of image features, namely texture features using the GLCM method and various color features, namely RGB, HSV, LAB, CMYK, YUV, HIS, HCL, LCH. However, with so many features, it causes a high computational load, so that in this study the Information Gain approach was used to select features. Furthermore, the classification process is carried out using the Support Vector Machine (SVM) method with variations of linear, polynomial, gaussian and sigmoid kernels. Tempe objects in the fermentation process are divided into unripe, ripe and rotten classes with a total of 410 images as a dataset. The test results (SVM+IG) on the Sigmoid kernel produce the fastest time accuracy with a computational result of 2.18 seconds on a 30:70 split ratio, the longest split ratio is 80:20 which is 2.50 seconds on a Linear kernel and produces the highest accuracy of 96 ,74%. Furthermore, in the SVM test without using Information Gain on the gaussian kernel, it produced the fastest time accuracy of 2.28 seconds, and the longest at a split ratio of 40:60, namely 3.00 seconds in the polynomial kernel. Thus the result of using SVM+IG is that the average level of accuracy when using (SVM+IG) is faster than the SVM process without IG which obtains slower computation time. Based on the description above, this study aims to apply the SVM method to classify tempe fermented images with feature selection using Information Gain.

The grammar of typically developing Oromo-speaking 3-year-olds

This study investigates the grammatical skills of typically developing Oromo-speaking preschool-age children and lays the foundation for a language assessment tool for Oromo, a Cushitic language spoken in Ethiopia. The current study used a standard picture–based elicitation task that evaluated children’s accuracy in producing grammatical utterances. Language samples were collected from 44 monolingual 3- to 4-year-old Oromo children with typical language development. The percentage of grammatical utterances (PGU) was computed and error types were analyzed. The average level of grammatical accuracy as measured by PGU was 69%, with a large amount of variability. The data for Oromo 3-year-olds showed a variety of infrequent error types. The results of the study provide crucial foundational knowledge on grammatical development in Oromo, which could form the basis of future language assessments and diagnostic materials.

Utilization of Drone with Thermal Camera in Mapping Digital Elevation Model for Ie Seu'um Geothermal Manifestation Exploration Security

Geothermal energy is a viable alternative energy source, particularly in Indonesia. This study was conducted at Ie Seu’um, Mount Seulawah Agam, which is a potential site for a geothermal power plant with an estimated electrical output of 150 megawatts. The objective of this study was to analyze and construct a digital elevation model (DEM) map of the geothermal manifestations. We analyzed water temperature, FLIR (Forward Looking Infrared) temperature, and temperature data from Landsat 8 satellite imagery. To map the heat signature of geothermal features, we utilized the DJI Phantom 4 Standard equipped with the FLIR One Gen 2 sensor. Additionally, we used the Milwaukee Mi306 to calculate the water temperature. Each test was conducted three times to obtain an optimal average level of accuracy. The DEM map was created to assess the level of safety in geothermal manifestation exploration. Elevation and slope values were analyzed to generate a 3D map display, providing a clearer image of the research site. In conclusion, drones prove to be an excellent method for ensuring the safety of exploration in geothermal manifestation areas.

Time-resolved EEG signal analysis for motor imagery activity recognition

Accurately characterizing brain activity requires detailed feature analysis in the temporal, spatial, and spectral domains. While previous research has proposed various spatial and spectral feature extraction methods to distinguish between different cognitive tasks, temporal feature analysis for each separate brain region and frequency band has been largely overlooked. This study introduces two novel approaches for recognizing cognitive activity: temporal entropic profiling and time-aligned common spatio-spectral patterns analysis. These approaches capture and use discriminative short-lived signal segments for motor imagery activity recognition. In Approach-1, we evaluated nine different measures to determine timing parameters that showed altered behavior associated with maximal inter-activity differences, which we then used in a machine-learning framework. In Approach-2, we used the best-performing signal characteristic measures from Approach-1 to determine the optimum latency of each channel at each frequency band for a CSP-based activity recognition strategy. We evaluated both approaches on two online available motor imagery EEG datasets and achieved average recognition accuracy levels of 86%. We compared our methods with four established BCI methods. The performance results show that our approaches exceeded the benchmark methods' performances, with notable improvements in the proposed time-aligned common spatio-spectral patterns approach. This study demonstrates that motor imagery recognition performance is improved when a temporal analysis is adopted alongside spatio-spectral neural feature analysis and that timing parameters associated with the maximal entropic difference of EEG segments to the cognitive tasks varied between different brain regions and subjects.

Design and Build Voltage and Current Monitoring Parameters Device of Rechargeable Batteries in Real-Time Using the INA219 GY-219 Sensor

Research on the manufacture of devices for monitoring battery voltage and current parameters has been carried out and developed using different types of sensors. This research have been released a device that can monitor battery voltage and current parameters in real time in the charge and discharge process using one sensor. The device is made using a microcontroller in the form of an Arduino Mega XPro 2560 R3, with an input in the form of an INA219 GY-219 sensor to monitor battery voltage and current. The output of device is in the form of automatic control of the charge and discharge circuit through the LCD and relay to obtain data from monitoring the battery voltage and current parameters. The batteries used as materials to be tested in this study are rechargeable Li-ion types. The device can monitor the voltage and current parameters of the battery as it should and can work automatically according to the control program created and designed by the manufacturer. Based on the test results of the INA219 GY-219 sensor in reading the voltage, the average sensor accuracy level is 99.96%, the sensor error average rate is 0.034%, and the sensor precision average is 99.97%. Meanwhile, in the reading current, the average sensor accuracy level is 98.39%, the sensor error average rate is 1.608%, and the sensor precision average is 98.98%.