HDR Image Compression by Multi-Scale down Sampling of Intensity Levels

HDR images are inherently very large in size compared to normal images. Hence, storage and communication overheads of HDR images are expensive to be used in mobile devices. Hence, invariably image compression is adopted for HDR images. In this paper, HDR image compression is achieved by down sampling the intensity levels while maintaining the dynamic range same as that of the original. This aspect retains the edge information of the images almost intact. Spatial down-sampling process is used to reduce the number of intensity samples. Consequently, this operation lowers the bit depth required to store the corresponding index file which in turn results in image compression.

CNN Feature-Based Image Copy Detection with Contextual Hash Embedding

As one of the important techniques for protecting the copyrights of digital images, content-based image copy detection has attracted a lot of attention in the past few decades. The traditional content-based copy detection methods usually extract local hand-crafted features and then quantize these features to visual words by the bag-of-visual-words (BOW) model to build an inverted index file for rapid image matching. Recently, deep learning features, such as the features derived from convolutional neural networks (CNN), have been proven to outperform the hand-crafted features in many applications of computer vision. However, it is not feasible to directly apply the existing global CNN features for copy detection, since they are usually sensitive to partial content-discarded attacks, such as copping and occlusion. Thus, we propose a local CNN feature-based image copy detection method with contextual hash embedding. We first extract the local CNN features from images and then quantize them to visual words to construct an index file. Then, as the BOW quantization process decreases the discriminability of these features to some extent, a contextual hash sequence is captured from a relatively large region surrounding each CNN feature and then is embedded into the index file to improve the feature’s discriminability. Extensive experimental results demonstrate that the proposed method achieves a superior performance compared to the related works in the copy detection task.

Współczesne nazwiska Polaków o możliwej genezie litewskiej nieobecne w LPŽ, poświadczone w kartotece antroponimicznej LKI

The aim of the article is to present the attestations of contemporary Polish surnames of Lithuanian origin which are absent from the dictionary of Lithuanian surnames (“Lietuvių pavardžių žodynas”, LPŽ), excerpted from the anthroponymic index card files that have been stored in the Lithuanian Language Institute in Vilnius and continually enlarged for several decades now. The files contain data excerpted from historical sources of the 16th to 19th centuries and consist of about 200,000 index cards (the actual number of excerpted anthroponyms is lower since some recur in various sources). Due to space limitations, generally only directly attested names have been included in the article, to the exclusion of those whose relationship with the researched name can be inferred rather than considered proven. Each listed attestation of an anthroponym (probably not in all cases an already established hereditary surname) is accompanied by information concerning its location and year (or time bracket), wherever available in the card index file. Given names or other details (e.g. the role of the person mentioned in documents, such as godmother in the data excerpted from baptismal registers) have only been included occasionally, if there was some reason to do so.

Jupyter-based service for JUNO analysis

The JUNO (Jiangmen Underground Neutrino Observatory) is designed to determine the neutrino mass hierarchy and precisely measure oscillation parameters. The estimated data volume of raw data is about 2 PB/year. The event rate of reactor anti-neutrinos is about 60/day, while the event rate of background is about O(10) Hz. The challenge is the event correlation during the analysis, where the background events could not be discarded. In order to use big data techniques to search for rare events, a Jupyter-based interactive service is developed for JUNO analysis. In this paper, an overview of this service is presented. The infrastructure is based on Jupyter and Kubernetes, which provides the user interface and resource management. In order to integrate the data processing framework and big data techniques, an index file is used as an intermediate file, which points to the interested events. Data processing framework SNiPER is used to select the candidate of neutrino signals and produce the index file. Apache Spark is then used to process such index file repeatedly with data cached in memory. With the index file produced from Spark and the complete event data files, SNiPER is used to process them and produce the final physics result. At the end of paper, the test-bed is presented and the testing result is shown.

Introduction

The background to the present work lies in the exciting archaeological climate of Afghanistan in the 1970s. Increasing numbers of foreign archaeological missions were engaged in fieldwork: following on from the pioneering work of the Délégation Archéologique Française en Afghanistan (DAFA) since 1922, British, German, Italian, Japanese, Soviet, and US missions were undertaking active research, as well as the Afghans themselves under the auspices of the Afghan Institute of Archaeology. The latest to establish a permanent presence in Kabul was the British Institute of Afghan Studies, in 1972. To keep abreast of these activities, in 1979 work on compiling a simple card-index file of archaeological sites in Afghanistan was begun for the library of the British Institute. It was designed as a quick, working reference guide to the major sites for the use of researchers who needed further information on a particular site or sites, modelled on those indexes existing at the time in the British School of Archaeology at Athens and the Institute of Archaeology at London University. The value of such a guide soon became apparent, and it was decided to expand this index into a full catalogue encompassing as many of the sites and monuments as possible that could be found from published sources. As such, all known sites, whether they were simply unidentified mounds observed in passing or major monumental and excavated sites, could be referred to quickly and a comprehensive list of publications dealing with each site be consulted, in tandem with expanding the Institute library. In its loose, unbound form it was designed not only to be consulted for reference but also to be constantly enlarged, updated, and improved by its users. As a result of expanding the index into a more comprehensive catalogue, it was suggested that a second version be prepared for publication as a gazetteer, and the original work was conceived. At the same time several colleagues offered to contribute their own unpublished field material for inclusion in the Gazetteer as a means of publishing sites hitherto accessible only in private archives. Chief of these were Jean-Claude Gardin and Bertille Lyonnet, who had recently completed their eastern Bactria surveys.

Peningkatan Kinerja Pencarian Dokumen Tugas Akhir Menggunakan Porter Stemmer Bahasa Indonesia dan Fungsi Peringkat Okapi BM25

Proses pencarian dokumen yang menggunakan information retrieval akan menerima query dan mengembalikan dokumen yang relevan dengan query pencarian tersebut. Relevansi diperhitungkan dari relevansi kata pada query dan kumpulan dokumen yang dicari. Pada sistem pencarian yang tidak mempertimbangkan variasi morfologi kata mengakibatkan dokumen yang mempunyai kata yang merupakan variasi dari kata pada query tidak dianggap sebagai dokumen hasil pencarian. Proses stemming dilakukan untuk mengenali variasi morfologi tersebut, dengan cara melakukan perubahan pada kata-kata berimbuhan dengan cara penghapusan awalan dan akhiran suatu kata menjadi kata dasarnya. Proses stemming dilakukan pada proses indexing, sehingga akan mengurangi ukuran dari index file. Hal itu dapat mengurangi waktu pencarian dan kebutuhan memori. Dokumen hasil pencarian akan ditampilkan sesuai nilai peringkat relevansi dokumen dengan query yang diberikan. Pemberian peringkat dilakukan dengan cara memberikan bobot pada dokumen. Dokumen yang mempunyai relevansi kata yang tinggi dengan query, akan diberikan bobot yang lebih besar. Pada sistem pencarian Tugas Akhir pada Universitas X, belum dilakukan proses stemming dan indexing. Untuk meningkatkan kinerja pencarian Tugas Akhir tersebut akan ditambahkan proses stemming dan indexing, serta pengurutan peringkat dokumen hasil pencarian. Proses stemming menggunakan porter stemmer bahasa Indonesia karena dokumen TA yang dicari berbahasa Indonesia, proses indexing menggunakan inverted index. Serta pengurutan dokumen hasil menggunakan fungsi peringkat Okapi BM25. Dari hasil uji coba, proses stemming dan fungsi peringkat yang dilakukan memberikan hasil pencarian yang lebih baik sesuai relevansi query. Penggunaan stemming dan inverted index menghemat penggunaan memori serta dapat mempercepat proses pencarian secara signifikan.

QUERY SUPPORT FOR GMZ

Generic text-based compression models are simple and fast but there are two issues that needs to be addressed. They cannot leverage the structure that exists in data to achieve better compression and there is an unnecessary decompression step before the user can actually use the data. To address these issues, we came up with GMZ, a lossless compression model aimed at achieving high compression ratios. The decision to design GMZ (Khandelwal and Rajan, 2017) exclusively for GML's Simple Features Profile (SFP) seems fair because of the high use of SFP in WFS and that it facilitates high optimisation of the compression model. This is an extension of our work on GMZ. In a typical server-client model such as Web Feature Service, the server is the primary creator and provider of GML, and therefore, requires compression and query capabilities. On the other hand, the client is the primary consumer of GML, and therefore, requires decompression and visualisation capabilities. In the first part of our work, we demonstrated compression using a python script that can be plugged in a server architecture, and decompression and visualisation in a web browser using a Firefox addon. The focus of this work is to develop the already existing tools to provide query capability to server. Our model provides the ability to decompress individual features in isolation, which is an essential requirement for realising query in compressed state. We con - struct an R-Tree index for spatial data and a custom index for non-spatial data and store these in a separate index file to prevent alter - ing the compression model. This facilitates independent use of compressed GMZ file where index can be constructed when required. The focus of this work is the bounding-box or range query commonly used in webGIS with provision for other spatial and non-spatial queries. The decrement in compression ratios due to the new index file is in the range of 1–3 percent which is trivial considering the benefits of querying in compressed state. With around 75 % average compression of the original data, query support in compressed state and decompression support in the browser, GMZ can be a good alternative to GML for WFS-like services.