24 research outputs found
Effectiveness of deep learning architecture for pixel-based image forgery detection
Get PDFDigital image forgery or forgery is easy to do nowadays. Verification of the authenticity of images is important to protect the integrity of the images from being misused. The use of a deep learning approach is state-of-the-art in solving cases of pattern recognition, the one is image data classification. In this study, image forgery detection was carried out using a deep learning-based method, the Convolutional Neural Network (CNN). The analysis of the different architecture of CNN has been done to show the effectiveness of each architecture. Two architectures were tested to know which one is more effective, architecture 1 has three convolution and pooling layers with 256 × 256 × 3 image input. While the other architecture has two convolution layers and pooling with 128 × 128 × 3 image input. The results show that the accuracy rate of the image forgery detection model in each architecture is around 80%. However, the validation accuracy is not more than 70%
Analysis of deep learning architecture for patch-based land cover classification
Get PDFIn recent years, the usage of computer vision methods for mapping land cover and land use has increased. Patch-based change detection may produce much superior results than pixel-based change detection and generate accurate change maps. Convolutional Neural Network (CNN) is an excellent option for remote sensing applications using hyperspectral data. The training of models using deep learning techniques takes a lot of time. This research aims to determine the most efficient deep learning model for patch-based land-cover classification by examining and comparing three CNN architectural models for land cover classification: LeNet-5, VGG-16, and ResNet-50. EuroSAT data derived from Sentinel-2A remote sensing imagery are utilized in this work. Comparing the three CNN architectures indicates that ResNet-50 has the highest validation accuracy, with a testing accuracy of 0.877, and a training time that is neither too quick nor too slow. The LeNet-5 model has the quickest training time but the lowest accuracy. VGG-16 has the longest training period yet has the highest test score of 0.878
Analysis of deep learning architecture for patch-based land cover classification
Get PDFIn recent years, the usage of computer vision methods for mapping land cover and land use has increased. Patch-based change detection may produce much superior results than pixel-based change detection and generate accurate change maps. Convolutional Neural Network (CNN) is an excellent option for remote sensing applications using hyperspectral data. The training of models using deep learning techniques takes a lot of time. This research aims to determine the most efficient deep learning model for patch-based land-cover classification by examining and comparing three CNN architectural models for land cover classification: LeNet-5, VGG-16, and ResNet-50. EuroSAT data derived from Sentinel-2A remote sensing imagery are utilized in this work. Comparing the three CNN architectures indicates that ResNet-50 has the highest validation accuracy, with a testing accuracy of 0.877, and a training time that is neither too quick nor too slow. The LeNet-5 model has the quickest training time but the lowest accuracy. VGG-16 has the longest training period yet has the highest test score of 0.878
The Effect of Error Level Analysis on The Image Forgery Detection Using Deep Learning
Get PDFDigital image modification or image forgery is easy to do today. The authenticity verification of an image become important to protect the image integrity so that the image is not being misused. Error Level Analysis (ELA) can be used to detect the modification in image by lowering the quality of image and comparing the error level. The use of deep learning approach is a state-of-the-art in solving cases of image data classification. This study wants to know the effect of adding ELA extraction process in the image forgery detection using deep learning approach. The Convolutional Neural Network (CNN), which is a deep learning method, is used as a method to do the image forgery detection. The impacts of applying different ELA compression levels, such as 10, 50, and 90 percent, were also compared in this study. According to the results, adopting the ELA feature increases validation accuracy by about 2.7% and give the better test accuracy. However, the use of ELA will slow down the processing time by about 5.6%
Opinion mining analysis on online product reviews using Naïve Bayes and feature selection
No full textFor profit companies that are centered on customers, every opinion of them is very important because knowing this opinion will be able to help the company to consider in making decisions about whether a product will be added to the production or not. It is necessary to know whether this opinion is positive or negative. This study discusses the classification of positive and negative opinions on online products using Naïve Bayes method. And to overcome a large number of datasets (having a lot of features/words) for increasing the accuracy of Naïve Bayes, it is combined with feature selection. In this case, the feature selection technique used is Chi-Square. This research is devoted to comparing system performance to see the performance of feature selection on small and large datasets. From the results of these tests, it is evident that feature selection combined with Naïve Bayes results in better system performance than not using feature selection for datasets that have many word features (large datasets) which can increase the accuracy around 11%. On the other hand, the use of feature selection has less effect on small datasets
ANALISA PERANCANGAN SISTEM INFORMASI KOMUNITAS OLAHRAGA FREELETICS SURABAYA MENGGUNAKAN METODE DSDM DAN RUP
Get PDFThe Freeletics Surabaya Community does not yet have an information system that can be used to convey information on community activities, health, sports events, and 12 week programs. The 12 week program is a program that members of the community can participate in. In this program, each member will be given a training menu for 12 weeks to work on. Each member who participates in this program is required to provide progress information every week, this information is weight and photos of body changes after doing exercise menu. With the information system, the exercise menu can be accessed via a mobile device and the progress of the exercise results can be uploaded and analyzed according to the needs, namely forming an ideal body and a healthy physique. In addition, it makes it easier for the community to share information on routine activities, 12 week programs, sports activities that can be followed, health information and social activities to the community so that people have other options in terms of health and fun social activities to do. The information system development method is the Dynamic System Development Method which will be integrated with the Rational Unified Process. By using these two methods, it is hoped that the development of information systems can be more effective and efficient. The final result of this research is a web-based information system, which can then be developed into a mobile application-based information system.Tujuan dari penelitian ini adalah mengembangkan sistem informasi Komunitas Freeletics Surabaya yang dapat digunakan untuk menyampaikan informasi kegiatan kemasyarakatan, kesehatan, olah raga, dan program 12 minggu. Program 12 minggu merupakan program yang dapat diikuti oleh anggota komunitas. Dalam program ini setiap anggota akan diberikan menu pelatihan selama 12 minggu untuk dikerjakan. Setiap anggota yang mengikuti program ini diwajibkan untuk memberikan informasi perkembangan setiap minggunya, yaitu informasi berat badan dan foto perubahan tubuh setelah melakukan menu senam. Dengan adanya sistem informasi menu senam dapat diakses melalui perangkat mobile dan progres hasil senam dapat diunggah dan dianalisis sesuai dengan kebutuhan yaitu membentuk tubuh yang ideal dan fisik yang sehat. Selain itu juga memudahkan masyarakat untuk berbagi informasi kegiatan rutin, program 12 minggu, kegiatan olah raga yang bisa diikuti, informasi kesehatan dan kegiatan sosial kepada masyarakat sehingga masyarakat mempunyai pilihan lain dalam hal kesehatan dan kegiatan sosial yang menyenangkan. melakukan. Metode pengembangan sistem informasi tersebut adalah Metode Pengembangan Sistem Dinamis yang akan diintegrasikan dengan Rational Unified Process. Dengan menggunakan kedua metode tersebut diharapkan pengembangan sistem informasi dapat lebih efektif dan efisien. Hasil akhir dari penelitian ini berupa sistem informasi berbasis web yang selanjutnya dapat dikembangkan menjadi sistem informasi berbasis aplikasi mobil
2D Mapping and boundary detection using 2D LIDAR sensor for prototyping Autonomous PETIS (Programable Vehicle with Integrated Sensor)
Get PDFPETIS (Programable Vehicle with Integrated Sensor) is a research project with goal make a robot that move independently with specific purpose. Due complexity of PETIS, research divide into several important sequence. In this research author focus on sense of sight for PETIS, LIDAR chosen due flexible and comprehensive. There is many LIDAR sensor in marketplace, LDS-01 as one of commercial LIDAR sensor available on market, produced by ROBOTIS as one of low-cost LIDAR sensor. Compare with another sensor that cost more than 500. On this research study focus with LDS-01 sensor reading, include hardware, software connection, and data handling. Based on this research LDS-01 as LIDAR sensor can read obstacle with minimum 29,9 cm and maximal 290,7 cm. Comparing with datasheet LDS-01 should work from 12 cm through 350 cm.
The Efficiency of Scrum Model for Developing Research and Publication Management Systems in Indonesia
Get PDFResearch, community service, and publication management systems are critical components of a higher education institution’s governance in Indonesia. Sistem Informasi Penelitian, Pengabdian, dan Publikasi (SIP3) or Information System of Research, Community Service, and Publication is the name of the developed system in this study, which is a web application-based system. SIP3 was built utilizing the scrum methodology and the Laravel framework. Scrum is an easy-to-implement Agile approach that facilitates the
rapid creation of systems or applications. This system includes four features: a researcher profile, research, community service, and publication. Development takes a short time with this scrum approach, roughly two months for these four features. SIP3 is evaluated for its effectiveness and practicality based on four criteria: system quality, information quality, user satisfaction, and benefits. The
”benefit” component receives the highest score, with the assertion that SIP3 enables more effective and efficient archiving of research data, services, and publications, as well as the ability to lower data error rates
Pengembangan aplikasi repositori publikasi penelitian dan pengabdian menggunakan framework PHP berbasis MVC
Get PDFProject ini bertujuan untuk membangun sebuah sistem informasi yang terbebas dari resiko yang mungkin terjadi ketika mengolah data secara manual seperti hilangnya berkas, pudarnya tulisan, kesulitan dalam mencari data publikasi penelitian dan pengabdian masyarakat. Sistem Informasi Publikasi Penelitian dan Pengabdian ini dibangun dalam bentuk katalog yang berisi informasi mengenai publikasi penelitian dan pengabdian yang dibuat dengan dua (2) jenis pengguna yaitu tamu (umum) dan admin. Sistem dikembangkan menggunakan framework PHP yang menggunakan pendekatan Model-View-Controller (MVC), yaitu Laravel. Sistem ini diharapkan mampu mengelola data penelitian dan pengabdian masyarakat baik dosen maupun mahasiswa yang menjadi penerapan karakteristik Ulul Albab di Universitas Islam Negeri Maulana Malik Ibrahim Malang yang salah satunya adalah konteks keluasan ilmu dalam melakukan penelitian dan pengabdian kepada masyarakat
