Hasil Pencarian  ::  Simpan CSV :: Kembali

"ABSTRAKImage edge detection plays a crucial role in image analysis and computer vision, it is defined as the process of finding the boundaries between objects within the considered image. The recognized edges may further be used in object recognition or image matching. In this paper a Canny image edge detector is used which gives acceptable results that can be utilized in many disciplines, but this technique is time consuming especially when a big collection of images is analyzed. For that reason, to enhance the performance of the algorithms, a parallel platform allowing speeding up the computation is used. The scalability of a multicore supercomputer node, which is exploited to run the same routines for a collection of color images from 2100 to 42000 images is investigated."

"This book offers a comprehensive introduction to advanced methods for image and video analysis and processing. It covers deraining, dehazing, inpainting, fusion, watermarking and stitching. It describes techniques for face and lip recognition, facial expression recognition, lip reading in videos, moving object tracking, dynamic scene classification, among others.The book combines the latest machine learning methods with computer vision applications, covering topics such as event recognition based on deep learning，dynamic scene classification based on topic model, person re-identification based on metric learning and behavior analysis. It also offers a systematic introduction to image evaluation criteria showing how to use them in different experimental contexts.The book offers an example-based practical guide to researchers, professionals and graduate students dealing with advanced problems in image analysis and computer vision."

"Kanker merupakan salah satu penyakit yang memiliki angka kematian tergolong tinggi di dunia. Analisis dengan menggunakan citra histopatologi merupakan gold standar yang dilakukan untuk diagnosa kanker. Kehadiran machine learning dan deep learning memanfaatkan data untuk dilatih dan kemudian akan menghasilkan metode untuk memprediksi atau identifikasi kanker. Kebutuhan data di dalam machine learning terlebih lagi deep learning sudah seharusnya tersedia. Namun, permasalahan yang kerap kali terjadi jika melakukan penelitian dengan data medis histopatologi adalah keterdiaan data yang terbatas. Salah satu kebaruan dari disertasi ini adalah telah berhasil melakukan modifikasi dan algoritma sliding window untuk mengatasi keterbatasan data citra histopatologi yang disebut dengan conditional sliding windows. Selain itu, disertasi ini juga telah berhasil merancang arsitektur deep learning untuk menghasilkan metode identifikasi status kanker dengan citra histopatologi dengan akurasi dapat dibandingkan dengan metode terkini yang berkembang. Penggunaan conditional sliding window mampu menghasilkan beberapa skenario dataset citra histopatologi yang akan digunakan sebagai dataset untuk proses pelatihan. Arsitektur yang dikembangkan adalah convolutional neural network (CNN) yang kami sebut dengan CNN-7-5-7. Dibandingkan dengan arsitektur deep learning seperti Alexnet dan DenseNet, CNN 7-5-7 menghasilkan performa yang lebih konsisten dan juga relatif lebih cepat dalam pelatihan. Apabila dibandingkan dengan model dengan data hasil pembangkitan Generative Adversarial Network (GAN).

---

Cancer is a disease that has a relatively high mortality rate in the world. Analysis using histopathological images is the gold standard for cancer diagnosis. The presence of machine learning and deep learning utilizes data to be trained and will produce methods to predict or identify cancer. The data needs in machine learning, especially deep learning, should be available. However, the problem that often occurs when conducting research with histopathological medical data is the limited availability of data. One of the novelties of this research is the successful modification and sliding window algorithm to overcome the limitations of histopathological image data which is called conditional sliding windows. In addition, this dissertation has also succeeded in designing a deep learning architecture to produce a method of identifying cancer status with histopathological images with an accuracy comparable to the latest developed methods. The use of conditional sliding windows is able to produce several scenarios of histopathological image datasets that will be used as datasets for the training process. The architecture developed is a convolutional neural network (CNN) which we call CNN-7-5-7. Compared to deep learning architectures such as Alexnet and DenseNet, CNN 7-5-7 delivers more consistent performance and is also relatively faster in training. When compared with the model with the generated Generative Adversarial Network (GAN) data."

"Parallel MATLAB for Multicore and Multinode Computers is the first book on parallel MATLAB and the first parallel computing book focused on the design, code, debug, and test techniques required to quickly produce well-performing parallel programs.MATLAB is currently the dominant language of technical computing with one million users worldwide, many of whom can benefit from the increased power offered by inexpensive multicore and multinode parallel computers. MATLAB is an ideal environment for learning about parallel computing, allowing the user to focus on parallel algorithms instead of the details of implementation."

"Scientific computing has often been called the third approach to scientific discovery, emerging as a peer to experimentation and theory. Historically, the synergy between experimentation and theory has been well understood: experiments give insight into possible theories, theories inspire experiments, experiments reinforce or invalidate theories, and so on. As scientific computing has evolved to produce results that meet or exceed the quality of experimental and theoretical results, it has become indispensable.Parallel processing has been an enabling technology in scientific computing for more than 20 years. This book is the first in-depth discussion of parallel computing in 10 years; it reflects the mix of topics that mathematicians, computer scientists, and computational scientists focus on to make parallel processing effective for scientific problems. Presently, the impact of parallel processing on scientific computing varies greatly across disciplines, but it plays a vital role in most problem domains and is absolutely essential in many of them. Parallel Processing for Scientific Computing is divided into four parts: The first concerns performance modeling, analysis, and optimization; the second focuses on parallel algorithms and software for an array of problems common to many modeling and simulation applications; the third emphasizes tools and environments that can ease and enhance the process of application development; and the fourth provides a sampling of applications that require parallel computing for scaling to solve larger and realistic models that can advance science and engineering."