Hasil Pencarian  ::  Simpan CSV :: Kembali

"In A neural network approach to fluid quantity measurement in dynamic environments, effects of temperature variations and contamination on the capacitive sensor are discussed, and the authors propose that these effects can also be eliminated with the proposed neural network based classification system. To examine the performance of the classification system, many field trials were carried out on a running vehicle at various tank volume levels that range from 5 L to 50 L. The effectiveness of signal enhancement on the neural network based signal classification system is also investigated. Results obtained from the investigation are compared with traditionally used statistical averaging methods, and proves that the neural network based measurement system can produce highly accurate fluid quantity measurements in a dynamic environment. Although in this case a capacitive sensor was used to demonstrate measurement system this methodology is valid for all types of electronic sensors."

"

Pengawasan bawah air sangat penting untuk memantau ekosistem laut, melindungi infrastruktur kritis, dan memastikan keamanan maritim dengan pendeteksian anomali, pelacakan aktivitas bawah air, dan perlindungan area sensitif. Namun, Kendaraan Bawah Air yang Dioperasikan dari Jarak Jauh (ROV) memiliki beberapa tantangan, salah satunya adalah arus bawah air sehingga diperlukan pengendali yang kuat untuk menjaga stabilitas. Skripsi ini memodelkan hubungan antara input dari RPM motor dengan pitch rate dan yaw rate sebagai output. Model Sistem Dinamis didapat dengan menggunakan data-data yang diperoleh selama uji lapangan di salah satu kolam uji coba di kota Bandung. Sebanyak 57,788 titik data dikumpulkan selama lima menit dan diolah menggunakan aplikasi MATLAB dengan memanfaatkan jaringan neural LSTM. Hasilnya menunjukkan bahwa dari Model Sistem Dinamis pitch rate didapatkan hasil simulasi terbaik dengan hyperparameter di dua layer LSTM, 900 Hidden Units, 1700 Epochs, 100 mini-batch size, 0.001 Initial Learning Rate, 0.8 Gradient Threshold, dan rasio training : testing sebesar 55:45, Selain itu, didapatkan nilai Root Mean Square Error (RMSE) training dan testing sebesar 0.041248 dan 0.2517. Pada Model Sistem Dinamis yaw rate didapatkan hasil simulasi terbaik dengan hyperparameter di dua layer LSTM, 950 Hidden Units, 2000 Epochs, 120 mini-batch size, 0.0005 Initial Learning Rate, 0.8 Gradient Threshold, dan rasio training : testing sebesar 55:45 dengan perolehan nilai RMSE training dan testing sebesar 0.030847 dan 0.70734. Dari simulasi yang telah dilakukan, penulis berhipotesis bahwa hasil simulasi telah cukup optimal untuk  digunakan dalam pemodelan Sistem Dinamis pada Kendaraan Bawah Air yang Dioperasikan Jarak Jauh.

---

Underwater surveillance is crucial for monitoring marine ecosystems, protecting critical infrastructure, and ensuring maritime security through anomaly detection, underwater activity tracking, and safeguarding sensitive areas. However, Remotely Operated Underwater Vehicles (ROVs) face several challenges, including underwater currents, necessitating robust controllers to maintain stability. This thesis models the relationship between input from motor RPMs and pitch rate and yaw rate as output. The Dynamic System Model is obtained using data collected during field tests in one of the trial pools in Bandung. A total of 57,788 data points were gathered over five minutes and processed using the MATLAB application, leveraging a neural LSTM network. The results indicate that for the Dynamic System Model, the best simulation results for pitch rate were achieved with hyperparameters in a two-layer LSTM: 900 Hidden Units, 1700 Epochs, 100 mini-batch size, 0.001 Initial Learning Rate, 0.8 Gradient Threshold, and a training-to-testing ratio of 55:45. Additionally, the Root Mean Square Error (RMSE) values for training and testing were 0.041248 and 0.2517, respectively. For yaw rate, the best simulation results were obtained with hyperparameters in a two-layer LSTM: 950 Hidden Units, 2000 Epochs, 120 mini-batch size, 0.0005 Initial Learning Rate, 0.8 Gradient Threshold, and the same training-to-testing ratio. The corresponding RMSE values for yaw rate were 0.030847 (training) and 0.70734 (testing). Based on the conducted simulations, the author hypothesizes that the simulation results are sufficiently optimal for use in modelling the Dynamic System of Remotely Operated Underwater Vehicles.

"

"Kolesterol merupakan zat lilin mengandung lemak yang dibutuhkan untuk memproduksi hormon dan substansi lainnya dalam tubuh. Apabila jumlahnya berlebih, maka akan tercampur dengan subtansi lain dan membentuk plak pada dinding pembuluh darah. Kolesterol yang tertimbun pada pembuluh darah biasanya disebut kolesterol jahat atau Low Density Liporpotein (LDL) yang merupakan penyebab timbulnya risiko penyakit jantug koroner dan stroke. Untuk mengukur kadar LDL biasanya dilakukan dengan pengambilan sampel darah (invasif) dengan metode lipid profile test. Selain itu metode secara non-invasif berbasis iridologi saat ini juga dikembangkan. Penelitian ini dilakukan untuk membentuk suatu sistem deteksi kadar LDL secara non-invasif berbasis iridologi yaitu dengan citra mata serta menggunakan deep learning sebagai model klasifikasi. Salah satu indikator berlebihnya kadar LDL dalam tubuh ialah adanya cincin yang berwarna putih keabuan yang mengelilingi bagian iris atau biasa disebut corneal arcus. Sistem yang dirancang terdiri dari instrumen akuisisi citra, algoritma pemrosesan citra dan model klasifikasi deep learning. Pemrosesan yang dilakukan ialah menggunakan algoritma Circular Hough Transform (CHT) untuk proses lokalisasi dan Rubber-Sheet Normalization untuk menormalisasi bagian iris. Untuk mendapatkan bagian corneal arcus maka dilakukan segmentasi pada citra iris mata kanan dan kiri. Model CNN digunakan sebagai model klasifikasi kelas LDL tinggi dan normal sehingga menghasilkan akurasi sebesar 97%. Sehingga sistem dapat dikatakan bekerja dengan baik dalam prediksi status kadar LDL dalam tubuh.

---

Cholesterol is a waxy substance contains fat that required to produce hormones and other substances in the body. If the amount of cholesterol is excessive, it can be mixed with other substances and formed plaque on blood vessels. Cholesterol that builds up in blood vessels is usually called bad cholesterol or Low Density Liporpotein (LDL) which is the cause of the risk of coronary heart disease and stroke. Measuring LDL levels is usually done by taking blood samples (invasive) with the lipid profile test method. Other than that, a non-invasive method based on iridology was also developed. This research was focus to develop a non-invasive detection system for LDL levels status prediction based on eye image (iridology) using Convolutional Neural Network (CNN) as a classification model. One indicator of excess LDL levels in the body is the presence of a grayish white ring that surrounds the iris which is called corneal arcus. The system designed consists of image acquisition instruments, image processing algorithms and deep learning classification models which is CNN. The image processing is done using Circular Hough Transform (CHT) algorithm for the localization process and Rubber-Sheet Normalization for normalize the iris region. Segmentation is conducted to get the corneal arcus located at the outer of the iris region. This LDL levels status prediction system that used CNN as a classification model  with 5-fold cross validation results an accuracy of 97%. Those result show that the system worked in LDL levels prediction."

"Pengukuran kadar gula darah merupakan salah satu kebutuhan utama dalam penanganan diabetes. Namun, moda pengukuran kadar gula darah yang umum saat ini, dilakukan secara invasive atau perlu melukai bagian tubuh manusia untuk mendapat nilai kadar gula darahnya. Terdapat metode pengukuran non invasive tanpa melukai manusia, namun metode ini masih belum dapat diandalkan karena banyaknya factor yang mempengaruhi glukosa tersebut. Penelitian ini mencoba untuk menganalisis akurasi dan performa dari pengukuran gula darah secara non invasive menggunakan sensor infrared pada panjang gelombang 940 nm dengan dibantu oleh Artificial Neural Network dan juga untuk mengevaluasi hubungan komponen dasar dari sinyal analog dari sensor yang bersangkutan terhadap kadar gula darah menggunakan Multiple Regression. Akurasi prediksi gula darah dievaluasi menggunakan Clark Grid Error analysis Dalam analisis ini, 81% dari 97 sampel data berada pada zona yang dapat diterima secara klinis, sedangkan sisanya berada pada zona yang tidak. Hal ini belum mencukupi kebutuhan akurasi 95% yang dapat diterima berdasarkan dari standar ISO 15197, maka hasil daripada penelitian ini masih belum memberikan hasil yang baik. Evaluasi menggunakan multiple regression sendiri menghasilkan hubungan yang tidak signifikan antara komponen dari sinyal analog dengan kadar gula darah dengan nilai R-squared sebesar 0.0174, RMSE 66.9, dan P-value keseluruhan sebesar 0.801.

---

Measuring blood sugar levels is one of the main needs in managing diabetes. However, the current common method of measuring blood sugar levels is carried out invasively or requires injuring parts of the human body to obtain blood sugar levels. There are non-invasive measurement methods without injuring humans, but this method is still not reliable because of the many factors that influence glucose. This research attempts to analyze the accuracy and performance of non-invasive blood sugar measurements using an infrared sensor at a wavelength of 940 nm assisted by an Artificial Neural Network and also to evaluate the relationship of the basic components of the analog signal from the sensor in question to blood sugar levels using Multiple Regression. The accuracy of blood sugar predictions was evaluated using Clark Grid Error analysis. In this analysis, 81% of the 97 data samples were in the clinically acceptable zone, while the rest were in the zone that was not. This does not meet the acceptable 95% accuracy requirement based on the ISO 15197 standard, thus the results of this research still do not provide relatively good results. Evaluation using multiple regression itself produced an insignificant relationship between the components of the analog signal and blood sugar levels with an R-squared value of 0.0174, RMSE 66.9, and an overall P-value of 0.801."

"Sustainable urban environments : an ecosystem approach presents fundamental knowledge of the built environment. Approaching the topic from an ecosystems perspective, it shows the reader how to combine diverse practical elements into sustainable solutions for future buildings and cities. From urban ecology to material, water and energy use, from urban transport to livability and health. The authors introduce and explore a variety of governance tools that support the transformation process, and show how they can help overcome institutional barriers. The book concludes with an account of promising perspectives for achieving a sustainable built environment in industrialized countries. Offering a unique overview and understanding of the most pressing challenges in the built environment, Sustainable urban environments helps the reader grasp opportunities for integration of knowledge and technologies in the design, construction and management of the built environment. "

"The urban environment - buildings, cities and infrastructure - represents one of the most important contributors to climate change, while at the same time holding the key to a more sustainable way of living. The transformation from traditional to sustainable systems requires interdisciplinary knowledge of the re-design, construction, operation and maintenance of the built environment. Sustainable Urban Environments: An Ecosystem Approach presents fundamental knowledge of the built environment. Approaching the topic from an ecosystems perspective, it shows the reader how to combine diverse practical elements into sustainable solutions for future buildings and cities. You'll learn to connect problems and solutions at different spatial scales, from urban ecology to material, water and energy use, from urban transport to livability and health. The authors introduce and explore a variety of governance tools that support the transformation process, and show how they can help overcome institutional barriers. The book concludes with an account of promising perspectives for achieving a sustainable built environment in industrialized countries. Offering a unique overview and understanding of the most pressing challenges in the built environment, Sustainable Urban Environments helps the reader grasp opportunities for integration of knowledge and technologies in the design, construction and management of the built environment. Students and practitioners who are eager to look beyond their own fields of interest will appreciate this book because of its depth and breadth of coverage."

"

Data LiDAR banyak menggantikan data dua dimensi untuk merepresentasikan data geografis karena kekayaan informasi yang dimilikinya. Salah satu jenis pemrosesan data LiDAR adalah segmentasi semantik tutupan lahan yang mana telah banyak dikembangkan menggunakan pendekatan model deep learning. Algoritma-algoritma tersebut menggunakan representasi jarak Euclidean untuk menyatakan jarak antar poin atau node. Namun, sifat acak dari data LiDAR kurang sesuai jika representasi jarak Euclidean tersebut diterapkan. Untuk mengatasi ketidaksesuaian tersebut, penelitian ini menerapkan representasi jarak non-Euclidean yang secara adaptif diupdate menggunakan nilai kovarian dari set data point cloud. Ide penelitian ini diaplikasikan pada algoritma Dynamic Graph Convolutional Neural Network (DGCNN). Dataset yang digunakan dalam penelitian ini adalah data LiDAR Kupang. Metode pada penelitian ini menghasilkan performa nilai akurasi 75,55%, di mana nilai akurasi ini lebih baik dari algoritma dasar PointNet dengan 65,08% dan DGCNN asli 72,56%. Peningkatan performa yang disebabkan oleh faktor perkalian dengan invers kovarian dari data point cloud dapat meningkatkan kemiripan suatu poin terhadap kelasnya.

---

LiDAR data widely replaces two-dimensional geographic data representation due to its information resources. One of LiDAR data processing tasks is land cover semantic segmentation which has been developed by deep learning model approaches. These algorithms utilize Euclidean distance representation to express the distance between the points. However, LiDAR data with random properties are not suitable to use this distance representation. To overcome this discprepancy, this study implements a non-Euclidean distance representation which is adaptively updated by applying their covariance values. This research methodology was then implemented in Dynamic Graph Convolutional Neural Network (DGCNN) algorithm. The dataset in this research is Kupang LiDAR. The results obtained performance accuracy value of 75.55%, which is better than the baseline PointNet of 65.08% and Dynamic Graph CNN of 72.56%. This performance improvement is caused by a multiplication of the inverse covariance value of point cloud data, which raised the points similarity to the class.

"