74 research outputs found
Batik Nitik 960
No full textBatik is one of the Indonesian cultural heritages which have noble cultural and philosophical meaning in every motif. This article introduces a dataset from Yogyakarta, Indonesia, "Batik Nitik 960". Dataset was captured from a piece of fabric consisting of 60 Nitik motifs. The dataset provided by Paguyuban Pecinta Batik Indonesia (PPBI) Sekar Jagad Yogyakarta, collection of Winotosasto Batik, and the data taken in APIPS Gallery. The dataset is divided into 60 categories with a total of 960 images, and each class has 16 photos. The images were captured by Sony Alpha a6400, lighting using Godox SK II 400, and data was filtered using jpg format. Each category has four motifs and is augmented using rotation using 90, 180, and 270 degrees. Each class has a philosophical meaning which describes the history of the motif. This dataset allows the training and validation of machine learning models to classify, retrieve, or generate a new batik motif using a generative adversarial network. To our knowledge, this is the first publicly available Batik Nitik dataset with philosophical meaning. Data provides by a collaboration of PPBI Sekar Jagad Yogyakarta, Universitas Muhammadiyah Malang, and Universitas Gadjah Mada. Hope this dataset "Batik Nitik 960" can support batik research and we are open to research collaboration
Temu kembali citra menggunakan micro structure co-occurrence descriptor (Penelitian pengembangan IPTEKS) / Minarno
No full text31 hal.:bib., ill.,lamp.; 28 c
Sistem Scada untuk pembangkit listrik tenaga mikro hidro (Penelitian Pengembangan Kerekayasaan dan Teknologi) / Minarno
No full text44 hal.:bib., ill.,lamp.; 28 c
PELATIHAN PEMBUATAN DAN PERAWATAN WEBSITE BERBAHASA INGGRIS UNTUK MENINGKATKAN PENJUALAN PAKET JASA TOUR DAN TRAVEL DI KECAMATAN KARANGPLOSO MALANG
No full textPELATIHAN PEMBUATAN DAN PERAWATAN WEBSITE BERBAHASA INGGRIS UNTUK MENINGKATKAN PENJUALAN PAKET JASA TOUR DAN TRAVEL DI KECAMATAN KARANGPLOSO MALANGYuda Munarko1 & Agus Eko Minarno21,2Teknik Informatika, Fakultas Teknik, Universitas Muhammadiyah MalangE-mail: 1) [email protected] tour dan travel saat ini adalah jenis usaha yang sedang marak di area Malang dan sekitarnya. Hal ini disebabkan oleh besarnya pangsa pasar wisata di area Malang dan meningkatnya perekonomian masyarakat pada umumnya. Permasalahan yang ada adalah persaingan yang sangat ketat diantara pengusaha jasa tour dan travel, yang menimbulkan persaingan tidak sehat, sehingga berimbas pada penetapan harga yang rendah dan tidak kompetitif, terutama untuk wisatawan domestik. Pada kenyataannya, jumlah wisatawan mancanegara yang berkunjung ke Malang, dan kebutuhan mereka belum cukup terwadahi dengan maksimal oleh jasa tour dan travel yang ada. Oleh karena itu, untuk meningkatkan potensi pendapatan jasa tour dan travel, maka harus dijalin komunikasi yang baik dan efisien dengan calon wisatawan mancanegara. Namun masalah yang dihadapi pengusaha tour dan travel pada umumnya adalah kendala bahasa dan tidak adanya media komunikasi yang efisien. Masalah ini juga yang dihadapi oleh mitra, yakni Manavin Tour and Travel dan MH Tour and Travel. Pendampingan pembuatan dan perawatan website berbahasa Inggris diharapkan mampu untuk membantu mitra dalam mengatasi permasalahan Mitra. Website tersebut menjadi media untuk promosi layanan dan juga media komunikasi yang efektif dengan calon wisatawan mancanegara. Hasil yang didapatkan, dengan adanya website ini mitra sudah mendapatkan klien dari mancanegara yang berhubungan dengan bidang usaha jasa tour dan travel.Kata Kunci: Website, Tour, Trave
Klasifikasi Resting-State Dan Task-State Pada Functional Magnetic Resonance Imaging Menggunakan Cross Correlation dan Support Vector Machine
No full textKlasifikasi Resting-State Dan Task-State Pada Functional Magnetic Resonance Imaging Menggunakan Cross Correlation dan Support Vector MachineClassification of Resting-State and Task-State In Functional Magnetic Resonance Imaging Using Cross Correlation and Support Vector MachineAgus Eko MinarnoJurusan Teknik Informatika, Fakultas Teknik, Universitas Muhammadiyah MalangJl. Tlogomas 246 Malang (0341) 464318 Email : [email protected] ABSTRACTIn the previous study identified several overlapping voxel during the resting-state and state task. Methods for determining the connectivity map involves overlapping areas, less than optimal in describing patterns of task-state as a feature. In this study, proposed a new method for the selection of features to improve accuracy and reduce computational time in determining significant voxel-state task when using a non-overlapping area. Selection of the features to determine significant voxel using cross Correlation, and take voxel correlation value which is above the average correlation. The next stage is the determination of the threshold value to determine the number of voxels are chosen as a feature. The selected features will be labeled in accordance with the given stimulus, namely picture and sentence, and then selected voxel to obtain non-overlapping between the stimulus picture with the stimulus sentence. The average yield of 6 subjects, methods that involve overlapping area using SVM classifier obtained precision, recall, and accuracy respectively 94.2%, 95.1%, 94.6% and computation time 0.021 seconds. While the method has a non-overlapping area of precision, recall, and accuracy respectively 95.0%, 95.3%, 95.1% and computation time 0.019 seconds. Feature selection methods using non-overlapping area has the accuracy and computation time better than methods that involve overlapping area, in determining the connectivity map.Keywords: feature selection, task-state, cross-correlation, voxel-based selection, non-overlappingABSTRAKPada penelitian sebelumnya teridentifikasi beberapa voxel yang overlapping pada saat resting-state dan task state. Metode untuk menentukan connectivity map melibatkan daerah yang overlapping, kurang optimal dalam menggambarkan pola task-state sebagai ciri. Pada penelitian ini, diusulkan sebuah metode baru untuk pemilihan fitur untuk meningkatkan akurasi dan mengurangi waktu komputasi dalam menentukan voxel yang signifikan pada saat task-state menggunakan metode non-overlapping area. Pemilihan fitur untuk menentukan voxel yang signifikan menggunakan cross corelation, dan mengambil voxel dengan nilai korelasi yang berada diatas korelasi rata-rata. Tahapan berikutnya adalah penentuan nilai ambang batas (threshold) untuk menentukan jumlah voxel yang dipilih sebagai fitur. Fitur yang terpilih akan diberi label sesuai dengan stimulus yang diberikan, yaitu picture dan sentence, kemudian diseleksi untuk mendapatkan voxel yang non-overlapping antara stimulus picture dengan stimulus sentence. Hasil rata-rata dari 6 subyek, metode yang melibatkan overlapping area menggunakan classifier SVM diperoleh precision, recall, dan accuracy masing–masing 94.2%, 95.1% , 94.6% dan waktu komputasi 0.021 detik. Sedangkan metode non-overlapping area memiliki precision, recall, dan accuracy masing–masing 95.0%, 95.3% , 95.1% dan waktu komputasi 0.019 detik. Pemilihan fitur menggunakan metode non-overlapping area memiliki akurasi dan waktu komputasi yang lebih baik dari metode yang melibatkan overlapping area, dalam menentukan connectivity map.Kata kunci : feature selection, task-state, cross-correlation, voxel-based selection, non-overlappin
HII: Histogram Inverted Index For Fast Images Retrieval
Get PDFThis work aims to improve the speed of search by creating an indexing structure in CBIR system. We utilised an inverted index structure that usually used in text retrieval with a modification. The modified inverted index is built based on histogram data that generated using Multi Texton Histogram (MTH) and Multi Texton Co-Occurrence Descriptor (MTCD) from 10,000 images of Corel dataset. When building the inverted index, we normalised value of each feature into a real number and considered pairs of feature and value that owned by a particular number of images. Based on our investigation, on MTCD histogram of 5,000 data test, we found that by considering histogram variable values which owned by maximum 12% of images, the number of comparison for each query can be reduced by 67.47% in a rate, the precision is 82.2%, and the rate of access to disk is 32.83%. Furthermore, we named our approach as Histogram Inverted Index (HII).
Temu Kembali Citra Menggunakan Multi Texton Co-Occurrence Descriptor
Get PDFSistem temu kembali citra masih menjadi topik penelitian yang belum terselesaikan. Beberapa metode ekstraksi fitur untuk temu kembali citra telah dikerjakan sebelumnya, diantaranya Gray Level Co-Occurrence Matrix (GLCM), Texton Co- Occurrence Histogram (TCM), Multi Texton Histogram (MTH), Micro Stucture Descriptor (MSD), Enhanced Micro Strcuture Descriptor (EMSD) dan Color difference Histogram (CDH). Namun, penelitian tersebut masih memiliki precision rata-rata 40%- 60%, sehingga masih perlu dikembangkan lebih lanjut. Dibandingkan dengan TCM, MSD, EMSD dan CDH, pendekatan menggunakan MTH memiliki kompleksitas komputasi yang lebih sederhana, sehingga untuk melakukan temu kembali citra menjadi lebih cepat. Namun demikian MTH memiliki kekurangan dalam merepresentasikan fitur. Pertama, MTH hanya menggunakan fitur lokal dalam merepresentasikan citra. Kedua, dalam pendeteksian pasangan piksel menggunakan Texton, ada informasi pasangan piksel yang terlewatkan sehingga dapat mengurangi representasi citra. Penelitian ini mengusulkan pendekatan baru untuk melakukan ekstraksi fitur pada sistem temu kembali citra. Kontribusi penelitian ini yaitu menambahkan jenis Texton baru untuk mendeteksi pasangan piksel dan menambahkan fitur GLCM. Metode yang diusulkan pada penelitian ini dinamakan Multi Texton Co-Occurrence Descriptor (MTCD). MTCD melakukan ekstraksi fitur warna, tekstur dan bentuk secara simultan menggunakan Texton, kemudian menghitung representasi citra secara global dengan GLCM. Texton mendeteksi konkurensi pasangan pixel pada setiap komponen RGB dan orientasi tepi citra, sedangkan GLCM merepresentasikan citra dengan sudut pandang global yang dihasilkan dari energy, entropy, contrast dan correlation. Fitur akhir MTCD berupa histogram hasil dari deteksi Texton dan GLCM. Data yang digunakan untuk temu kembali citra menggunakan 300 data Batik dan 10.000 data Corel. Pengukuran kemiripan citra menggunakan Canberra dan pengukuran performa MTCD menggunakan precision dan recall. Data uji dipilih secara acak terdiri dari 50 d ata Batik, 2.500 untuk data Corel 5.000 dan 5.000 untuk data Corel 10.000. Berdasarkan hasil uji coba yang telah dilakukan, penambahan 2 texton baru dan fitur GLCM dapat meningkatkan precision 2,86% pada data Batik, 3,40% pada data Corel 5.000 dan 3,06% pada data Corel 10.000. MTCD lebih unggul daripada MTH untuk temu kembali citra.
=============================================================================================================================
Image retrieval system is one of a challenging topic and is not yet finalized. A number of features extraction methods has been proposed, for example Gray Level Co- Occurrence Matrix (GLCM), Texton Co-Occurrence Histogram (TCM), Multi Texton Histogram (MTH), Micro Stucture Descriptor (MSD), Enhanced Micro Structure Descriptor (EMSD) and Color difference Histogram (CDH). However, the precision rate of those methods are relatively low, between 40% and 60%. Therefore, there is a need of a new approach to improve the results. Looking to those methods, in term of computational complexity, MTH is the simplest. The problem is that there is weakness in representing image features. First, MTH using local features to representate the image. Second, The weakness occurs in the proces of detecting pairs of pixel using texton for color quatization and edge orientation quantization. This study proposes a new approach to perform features extraction in image retrieval systems. Contribution of this study is to add new types of Texton to detect pairs of pixels and adding GLCM features. The method in this study is called Multi Texton Co-Occurrence Descriptor (MTCD). MTCD works by extracting color features, texture features and shape features simultaneously using Texton, then calculates the global image representations with GLCM. Texton detects concurrency of pairs of pixels on each RGB component and the edge orientation of image, while GLCM represents the image as global viewpoint by the value of energy, entropy, contrast and correlation. Features that are detected by MTCD are presented as histogram. The data used in this study is a 300 batik data and a 10,000 Corel data. In order to measure image similarity, Canberra Distance is used. For performance measurement, precision and recall are used. Test data randomly selected consists of 50 Batik data, 2,500 for Corel 5.000 and 5.000 for Corel 10.000. Based on the results of the testing that has been done, the addition of 2 new texton and GLCM features can improve the precision 2.86%, 3 ,40% and 3,06% on Batik, Corel 5.000 and Corel 10.000 respectively. MTCD is superior than MTH for image retrieval
CBIR of Batik Images Using Micro Structure Descriptor on Android
Get PDFBatik is part of a culture that has long developed and known by the people of Indonesia and the world. However, the knowledge is only on the name of batik, not at a more detailed level, such as image characteristic and batik motifs. Batik motif is very diverse, different areas have their own motifs and patterns related to local customs and values. Therefore, it is important to introduce knowledge about batik motifs and patterns effectively and efficiently. So, we build CBIR batik using Micro-Structure Descriptor (MSD) method on Android plat- form. The data used consisted of 300 images with 50 classes with each class consisting of 6 images. Performance test is held in three scenarios, where the data is divided as test data and data train, with the ratio of scenario 1 is 50%: 50%, scenario 2 is 70%: 30%, and scenario 3 is 80%: 20%. The best results are generated by scenario 3 with precision valur 65.67% and recall value 65.80%, which indicates that the use of MSD on the android platform for CBIR batik performs well
Vehicle Classification using Haar Cascade Classifier Method in Traffic Surveillance System
Get PDFObject detection based on digital image processing on vehicles is very important for establishing monitoring system or as alternative method to collect statistic data to make efficient traffic engineering decision. A vehicle counter program based on traffic video feed for specific type of vehicle using Haar Cascade Classifier was made as the output of this research. Firstly, Haar-like feature was used to present visual shape of vehicle, and AdaBoost machine learning algorithm was also employed to make a strong classifier by combining specific classifier into a cascade filter to quickly remove background regions of an image. At the testing section, the output was tested over 8 realistic video data and achieved high accuracy. The result was set 1 as the biggest value for recall and precision, 0.986 as the average value for recall and 0.978 as the average value for precision
- …
