Hasil Pencarian  ::  Simpan CSV :: Kembali

"In this paper, we present a 3D shape modeling system based on Tsai Shah shape from shading (SFS) algorithm. This SFS provides partial 3D shapes, as depth maps of the object to be reconstructed. Our previously developed Projected Polygon Representation Neural Network (PPRNN) performed the reconstruction process. This neural network is able to successively refine the polygon vertices parameter of an initial 3D shape based on 2D images taken from multiple views. The reconstuction is finalized by mapping the texture of object image to the 3 D initial shape. It is known from static stereo analysis that even though multiple view images are used, obtaining 3D structure without considering of base-distance information, i.e. focal separation between different camera positions, is impossible. Unless there is something else is known about the scene. Here we propose the use of shading features to extrat the 3D depth maps by using a fasat SFS algortihm, instead of rendering the object based on bare 2D images. A beginning result of reconstructing human (mannequin) head and face is presented. From our experiment, it was shown that using only 2D images would result a poor reconstruction. While using the depth-maps provides a smoother and more realistic 3D object. "

"Aplikasi metode Multiatribut pada data poststack seismik dan hasil inversinya telah dilakukan untuk mengestimasi kecepatan interval melalui pendekatan Neural Network. Estimasi kecepatan interval yang dihasilkan tersebut digunakan untuk memprediksi tekanan formasi di Lapangan Texaco 3D, Louisiana. Tujuan dari studi ini adalah untuk mengaplikasikan pendekatan geostatistik dan analisis Multiatribut dengan keterbatasan data yang dimiliki untuk memprediksi tekanan formasi. Hasil estimasi kecepatan interval menggunakan Multiatribut (10 atribut) menunjukkan korelasi yang sangat baik yaitu rata-rata korelasi prediksi log hasil atribut dan log validasi mencapai 79%, dengan tingkat kesalahan yang kecil berkisar rata-rata 175 - 292 m/s dari kecepatan validasi. Pendekatan Neural Network menghasilkan atribut polaritas semu (apparent polarity) sebagai atribut terbaik dalam estimasi kecepatan dengan error berkisar 108 m/s (berdasarkan hasil PNN) hingga 166 m/s (berdasarkan hasil MLFN). Anomali kecepatan rendah terdeteksi pada kedalaman 2800 - 2900 m dan sekitar kedalaman 3000 m, dengan gradient tekanan rata-rata mencapai 18 ? 22 ppg.

---

Application of Multiattribute to poststack seismic data and the the seismic inversion result has been carried out to estimate the interval velocity, by using Neural Network approach. The result of estimated interval velocity is used to predict formation pressure in Texaco 3D Field, Louisiana. The purpose of this study is to apply the geostatistical approach and Multiattribute analysis to predict the formation pressure.

The results of estimated interval velocity using Multiattribute (10 attributes) show excellent correlation of the average correlation between predicted log and the real log reached 79%, with an error training and validation of a fairly small range from an average of 175-292 m/s validation of the velocity. The Neural Network approachment generating apparent polarity attribute as the best attribute of velocity estimation with errors ranging from 108 m/s (based on PNN) up to 166 m/s (based on the results of MLFN). Low velocity anomaly was detected at a depth of 2800 - 2900 m and approximately 3000 m depth, with the pressure gradient averaged 18-22 ppg."

"Kondisi pasca bencana adalah sebuah kondisi darurat yang membutuhkan pertolongan pertama dari tim penyelamat. Oleh karena itu, dikembangkan pemanfaatan radar yang digunakan untuk mendeteksi manusia dalam kondisi pasca-bencana. Sayangnya, banyaknya parameter yang mempengaruhi pengklasifikasian membatasi pemakaian radar 24 GHz seperti reruntuhan yang menutupi manusia. Oleh karena itu, radar dengan frekuensi yang lebih tinggi dimanfaatkan dengan frekuensi 77 GHz yaitu sinyal milimeter. Metode seperti deep learning dan backpropagation neural network sudah diterapkan pada penelitian-penelitian sebelumnya menggunakan radar sinyal milimeter. Namun, tingkat akurasi dari klasifikasi kelas dari makhluk hidup hanya mencapai 49% dengan jumlah klasifikasi 2 kelas dan 32% dengan jumlah klasifikasi 3 kelas. Oleh karena itu, dikembangkan kembali dengan metode Convolutional Neural Network. Akurasi yang didapatkan meningkat hingga mencapai 99% untuk klasifikasi 2 kelas dan 3 kelas. Namun akurasinya menurun untuk klasifikasi kelas yang lebih banyak hingga 68%. Skripsi ini mengajukan metode 3D-Convolutional Neural Network guna meningkatkan resolusi dari data yang diberikan dalam pelatihan dari model untuk meningkatkan akurasi pada klasifikasi kelas dengan model yang diajukan.

---

The post-disaster condition is an emergency that requires immediate first aid from rescue teams. Therefore, the use of radar has been developed to detect humans in post-disaster conditions. Unfortunately, the numerous parameters affecting classification, such as rubble covering humans, limit the use of 24 GHz radar. Consequently, higher frequency radar, specifically 77 GHz millimeter-wave signals, is utilized. Methods like deep learning and backpropagation neural networks have been applied in previous studies using millimeter-wave radar signals. However, the classification accuracy for living beings reached only 49% for two-class classification and 32% for three-class classification. Therefore, the method was further developed using Convolutional Neural Networks (CNN). The accuracy achieved improved to 99% for both two-class and three-class classifications, but it decreased to 68% for classifications with more classes. This thesis proposes the use of a 3D-Convolutional Neural Network method to enhance the resolution of the data used in model training, aiming to improve the accuracy of class classification with the proposed model."

"Citra hiperspektral memiliki informasi dalam rentang spektrum yang luas melebihi rentang spektrum yang ada pada citra RGB sebagai citra yang umum digunakan sehari-hari saat ini. Informasi tersebut dapat dimanfaatkan dalam berbagai macam bidang; salah satunya adalah pengukuran kadar tertentu dalam suatu objek. Namun, kamera hiperspektral sebagai alat akuisisi citra memiliki kekurangan yaitu harganya yang mahal, tidak mudah dioperasikan, ukuran hasil citra yang besar, serta memerlukan teknik dan perangkat khusus saat mengakuisisi citra. Hal tersebut berbeda dengan kamera RGB yang memiliki harga yang jauh lebih murah, hasil citra berukuran kecil, serta mudah dioperasikan. Penelitian ini melakukan implementasi sistem rekonstruksi citra hiperspektral dari citra RGB berbasis convolutional neural network ResNet pada sistem prediksi kadar fenolik daun bisbul. Terdapat proses rekonstruksi citra hiperspektral dengan target jumlah bands sebanyak 224 pada rentang panjang gelombang 400 sampai 1000 nm. Penelitian ini menggunakan algoritma model ResNet untuk model rekonstruksi citra, serta algoritma model XGBoost untuk model prediksi kadar. Performa model yang dihasilkan dalam penelitian ini adalah RMSE sebesar 0,1129 dan MRAE sebesar 0,3187 untuk model rekonstruksi citra, serta RMSE sebesar 0,5798 dan MRAE sebesar 0,1431 untuk model prediksi kadar. Citra hiperspektral hasil rekonstruksi mampu menghasilkan pola spectral signature yang serupa dengan citra hiperspektral asli.

---

Hyperspectral images have much information within their large spectrum area; larger than RGB images which are used daily nowadays. The information can be used in many applications; one of them is content measurement of an object. However, hyperspectral cameras as an image acquisition instrument have disadvantages, such as high cost, not easy to operate, large image results, and require additional equipment in its image acquisition. This is different from RGB cameras which have cheaper price, smaller in image size, and easier to operate. This study implemented a hyperspectral image reconstruction system from RGB images based on the ResNet convolutional neural network on the velvet apple leaf’s phenolic content prediction system. This study reconstructs hyperspectral images with a total target of 224 bands in the wavelength range of 400 to 1000 nm. This study uses the ResNet model algorithm for the image reconstruction model, and the XGBoost model algorithm for the content prediction. The performance of the model produced in this study is RMSE of 0.1129 and MRAE of 0.3187 for the image reconstruction model, as well as RMSE of 0.5798 and MRAE of 0.1431 for the content prediction model. The reconstructed hyperspectral image can produce the same spectral signature pattern as the original hyperspectral image."

"Kanker paru merupakan kanker yang paling banyak ditemukan dan paling mematikan di dunia. Penentuan stadium kanker paru umumnya dilakukan oleh dokter radiologi dengan melihat pembesaran kelenjar getah bening (KGB) mediastinal. KGB mediastinal cukup sulit dideteksi secara visual dikarenakan memiliki kontras yang rendah  terhadap jaringan di sekitarnya, ukuran dan bentuknya yang bervariasi, serta tersebar di berbagai lokasi. Oleh karena itu, akhir – akhir ini dikembangkan sistem computer-aided detection (CADe) sebagai alat bantu bagi dokter radiologi untuk mendeteksi KGB mediastinal secara otomatis. Metode terbaik saat ini dalam sistem CADe KGB mediastinal tersebut menggunakan 2D convolutional neural network (CNN) yang diterapkan dari 3 sudut pandang (axial, coronal, sagittal). Namun, sifat 3D dari KGB mediastinal dihipotesakan akan lebih terwakili jika menggunakan 3D CNN. Oleh karena itu, dalam penelitian ini digunakan 3D CNN yang kemudian diubah menjadi 3D fully convolutional network (FCN)  untuk mendeteksi kandidat KGB mediastinal di dalam suatu tumpukkan citra CT. Kandidat KGB mediastinal tersebut kemudian dianalisa untuk mengurangi false positive (FP) menggunakan 3 metode, yaitu perhitungan mean HU, deteksi kontur menyerupai lingkaran, dan klasifikasi menggunakan 3D CNN. Performa terbaik dari sistem CADe KGB mediastinal ini diperoleh ketika menggunakan 3D CNN dalam tahap pengurangan FP dengan sensitivitas 77% dan 12 FP/pasien.

---

Lung cancer is the most common and the deadliest cancer in the world. Lung cancer staging usually was done by radiologist by detecting mediastinal lymph node (LN) enlargement. Mediastinal LN is difficult to be detected visually due to its low contrast to the surrounding tissues, various size and shape, and sparse location. Therefore, computer-aided detection (CADe) system has been developed as a tool for radiologist to detect medistinal LN automatically. The state of the art mediastinal LN CADe system used 2D convolutional neural network (CNN) from 3 planar views (axial, coronal, sagittal). However, the 3D features of mediastinal LN are hypothesized to be more reprenseted if 3D CNN is used. Therefore, in this experiment we used 3D CNN which is converted to 3D fully convolutional network (FCN) to detect mediastinal LN candidate in a stack of CT images. Then, the mediastinal LN candidates were analyzed using 3 methods to reduce the false positive (FP), which are the calculation of the mean HU, the blob detection, and the classification using 3D CNN. The best performance of this CADe system was achieved when the 3D CNN was used in the FP reduction stage which has 77% of sensitivity and 12 FP/ patient."

"Penelitian ini memiliki tujuan untuk mencari model machine learning yang dapat mengenali kegiatan yang dilakukan pengguna ATM, serta mencari algoritma terbaik untuk mengetahui kapan suatu kegiatan pengguna ATM dimulai dan selesai pada suatu video. Terdapat sembilan jenis aktivitas berbeda yang ingin dideteksi. Penelitian ini dapat dibagi dalam dua fase, yaitu fase mencari rentang waktu aktivitas pada video yang disebut fase deteksi aktivitas, dan fase mengenali aktivitas tersebut yang disebut fase pengenalan aktivitas. Pada fase pengenalan aktivitas, penulis mengajukan suatu rancangan arsitektur 3D CNN, serta melakukan eksperimen terhadap parameter pada arsitektur tersebut. Setelah melakukan beberapa eksperimen, didapatkan model terbaik dengan kernel berukuran 3 x 3 x 3, menggunakan input video dengan piksel berukuran 20 x 20 per frame, dan menggunakan dua lapis layer ekstraksi fitur. Pada fase deteksi aktivitas, penulis mengajukan suatu rancangan fungsi deteksi aktivitas, yang mengikuti framework ‘classification lalu post-processing’ yang merupakan salah satu framework untuk deteksi aktivitas (Yao et al., 2018), serta melakukan eksperimen terhadap parameter pada fungsi tersebut. Setelah melakukan beberapa eksperimen, didapatkan performa terbaik dengan parameter teta sebesar 20, dan konstanta C sebesar 365. Pada kedua eksperimen tersebut, terdapat beberapa kesalahan yang dilakukan, sehingga diperlukan eksperimen lanjutan dimana kesalahan tersebut tidak dilakukan. Kesalahan tersebut adalah model kemungkinan besar masih underfit, dan terdapat permasalahan pada pemotongan video manual pada dataset. Setelah menyelesaikan kesalahan tersebut, model untuk fase pengenalan aktivitas mendapatkan akurasi sebesar 93.94%, presisi sebesar 96.36%, recall sebesar 93.94%, dan f-score sebesar 93.69%. Pada sisi lain, dalam fase deteksi aktivitas didapatkan akurasi sebesar 94.44%, presisi sebesar 96.30%, recall sebesar 96.30%, dan f-score sebesar 94.07%.

---

This research aims to find a machine learning model that can recognize the activities of ATM users, and find the best algorithm to find when each ATM user activity starts and finishes on a video. There are nine different types of activities that this study want to detect. This research can be divided into two phases, namely the phase of detecting for a time span of activity on a video that is called the activity detection phase, and the phase of recognizing that activity that is called the activity recognition phase. In the activity recognition phase, I propose a 3D CNN architecture design, and conduct experiments on the parameters of the architecture. After carrying out several experiments, the best model is obtained with a kernel with dimensions of 3 x 3 x 3, using video input with pixels measuring 20 x 20 per frame, and using two layers of feature extraction layer. In the activity detection phase, I propose an activity detection function, which follows the ‘classification then post-processing’ framework, which is one of the frameworks for activity detection (Yao et al., 2018), and conducts experiments on the parameters of the function. After carrying out several experiments, the best performance was obtained with a theta parameter of 20, and a constant C of 365. In both experiments, there were some errors made, so that further experiments were needed to be done, where the errors were not carried out. The error is that the model is most likely still in underfit phase, and there are problems with manual video clipping on the dataset. After resolving these errors, the model for the activity recognition phase gained an accuracy of 93.94%, a precision of 96.36%, a recall of 93.94%, and an f-score of 93.69%. On the other hand, in the activity detection phase an accuracy of 94.44% is obtained, a precision of 96.30%, a recall of 94.44%, and an f-score of 94.07%."

"Estimation of porosity (and other petrophysical parameters) from well logs are important yet difficult task encountered in geophysical formation evaluation and reservoir engineering. Motivated by recent result of artificial neural network (ANN) modelling offshore eastern Canada and North sea, we have developed neural nets for converting well logs in the Field-X, West Java, to porosity. We use back propagation artificial neural network (BPANN) to model porosity of the area. The porosity ANN is a simple three layer network using sonic, density and resistivity logs for input.Optimum network's parameters, like type of activation function, number of facts, and number of neurons also have been investigated through series of trials and errors of network. The network, initially developed for basin-scale problems, perform sufficiently accurate to meet normal requirements. There is strong similarity (R=0.964) between the predicted porosity from BPANN with density-derived porosity (which has been used as a substitute for core plug porosity due to not enought core data available). A major adventage is that no a priori knowledge of the rock material and pore fluids is required. Real-time conversion based on measurements while drilling (MWD) is thus an obvious application."