A Deep Local and Global Scene-Graph Matching for Image-Text Retrieval

Conventional approaches to image-text retrieval mainly focus on indexing visual objects appearing in pictures but ignore the interactions between these objects. Such objects occurrences and interactions are equivalently useful and important in this field as they are usually mentioned in the text. Scene graph presentation is a suitable method for the image-text matching challenge and obtained good results due to its ability to capture the inter-relationship information. Both images and text are represented in scene graph levels and formulate the retrieval challenge as a scene graph matching challenge. In this paper, we introduce the Local and Global Scene Graph Matching (LGSGM) model that enhances the state-of-the-art method by integrating an extra graph convolution network to capture the general information of a graph. Specifically, for a pair of scene graphs of an image and its caption, two separate models are used to learn the features of each graph’s nodes and edges. Then a Siamese-structure graph convolution model is employed to embed graphs into vector forms. We finally combine the graph-level and the vector-level to calculate the similarity of this image-text pair. The empirical experiments show that our enhancement with the combination of levels can improve the performance of the baseline method by increasing the recall by more than 10% on the Flickr30k dataset. Our implementation code can be found at https://github.com/m2man/LGSGM.

Improved feature exctraction process to detect seizure using CHBMIT-dataset

One of the most dangerous neurological disease, which is occupying worldwide, is epilepsy. Fraction of second nerves in the brain starts impulsion i.e. electrical discharge, which is higher than the normal pulsing. So many researches have done the investigation and proposed the numerous methodology. However, our methodology will give effective result in feature extraction. Moreover, we used numerous number of statistical moments features. Existing approaches are implemented on few statistical moments with respect to time and frequency. Our proposed system will give the way to find out the seizure-effected part of the brain very easily using TDS, FDS, Correlation and Graph presentation. The resultant value will give the huge difference between normal and seizure effected brain. It also explore the hidden features of the brain.

Comparison between Regenerator Model and the Instantaneous Direct Solar Radiation for Port Harcourt

This paper compares the result generated by the Regenerator technique and the Ideriah’s method for instantaneous solar radiation. It considered the generation of the actual power input from the regenerator. The actual power input to the photovoltaic panel was calculated from a power model derived from the regenerator’s standard power that is nominally provided, regenerator material and reflectivity of the material. The amount of visible radiation absorbable by photovoltaic plates was further derived in the model and a value of 0.386Pi was obtained. The Ideriah’s model was used to calculate global solar radiation from extraterrestrial solar radiation. It utilized the concept of transmittance to arrive at the amount of insolation received locally (i.e. at the point of interest). It also calculated the amount of insolation at optical thicknesses (Air mass values) representing top of the hour from 5.0am to 7.0pm. Comparison & result showed that the regenerator model gave 84% to 97% correctness or an approximated average correctness of 94% when compared to the Ideriah’s model. Chi-square was used to test the relationship between the results. Chi-square results showed a high level of significance in the similarity of the results obtained. Python software was used to simulate the results and graph presentation.

PENGARUH CAHAYA LAMPU TERHADAP POLA AGREGASI IKAN DI BAGAN TANCAP PERAIRAN KEPULAUAN SERIBU

Penelitian ini bertujuan untuk mempelajari pola agregasi ikan di bawah pengaruh cahaya lampu bagan tancap. Penelitian dilaksanakan pada bulan Mei 2005 di Kepulauan Seribu. Pengamatan pola agregasi ikan dilakukan dengan menggunakan alat akustik Simrad EY500. Data intensitas cahaya diambil dengan menggunakan light meter LI-250, serta verifikasi hasil tangkapan untuk memperkuat hasil analisis. Analisis meliputi interpretasi visual berbasis pada presentasi grafik yang merupakan dasar bagi penafsiran data dan penyusunan informasi. Pengaruh intensitas cahaya yang berbeda terhadap agregasi ikan mempunyai pola yang tidak sama, baik nilai intensitas cahaya optimumnya, besarnya agregasi, maupun posisi kedalaman untuk nilai agregasi terbesar. Proses pengumpulan ikan pada awal penyinaran dengan perlakuan jenis lampu petromak lebih cepat dibanding dengan hasil perlakuan lampu mercuri. Ikan akan beradaptasi terhadap variasi iluminasi cahaya sehingga selama proses pencahayaan terjadi migrasi. The research objective was to study fish aggregation pattern under light influence set in stationary bamboo lift net. This research was conducted on May 2005 in Seribu Islands waters. Simrad EY-500 echosounder was used to observe the fish aggregation. Light meter LI 250 was used to measure light intensity level and catch data for supporting the analysis result. Data analyzed by visual description based on graph presentation were used as data interpretation and information. The influence of different light intensity on fish aggregation had different pattern especially for optimum value of light intensity, maximumvalue of fish aggregation, and depth layer of the maximum value of fish aggregation. For fish aggregation under light of kerosene pressured lamps was quicker than mercury lamps in the early of illumination process. Fish will adapt to various light illumination, so that during illumination process the migration of fish is occured.

Graph Description as an Issue in L2 Academic English Writing

Young people’s interest in taking international exams such as IELTS results from student mobility and their willingness to appraise language abilities. In this paper, Academic Writing Task One of IELTS is examined. This task implies candidates comprehending graphic information and processing it in written discourse. The gap between a host of graph description tests and an insignificant number of efficient teaching methods has provided a rationale for the current study. It focuses on graph description as a cognitive, psychological and educational process and employs the analysis method in the theoretical section. Based on the action research method, drawing on 25 students’ written samples, the study has quested for peculiar language problems detrimental to processing the graph description task. The data have revealed the key pillars of successful written graph presentation: the combination of all four main skills, i.e. reading, listening, writing, and speaking; skills transfer; critical thinking and writing; the appropriate use of style; graphic literacy. It is concluded that the “constant nudging” method, a skills transfer, the use of appropriate vocabulary for describing trends alongside academic functional phrases and grammar features, the analysis of mass media information with numeric data are solutions to graph description issues. The findings have implications for preparing students for IELTS. This article is published under the Creative Commons Attribution 4.0 International License.

FRACTAL GEOMETRY OF EQUILIBRIUM PAYOFFS IN DISCOUNTED SUPERGAMES

This paper examines the pure-strategy subgame-perfect equilibrium payoffs in discounted supergames with perfect monitoring. It is shown that the equilibrium payoffs can be identified as sub-self-affine sets or graph-directed iterated function systems. We propose a method to estimate the Hausdorff dimension of the equilibrium payoffs and relate it to the equilibrium paths and their graph presentation.

Pengaruh Lama Pengasinan Terhadap Kadar Protein Putih Telur Itik

Background : Egg is one of the animal protein source, which has delicious taste, easy to digest and highly nutritious. Besides its affordable price, its supply availability is unquestionable as well. However, due to its short storability, it requires special treatment, such as preserving, to store it for long period. One way to preserve the egg is by pickling egg, which generally requires seven to ten days of marinating. During the process of marinating, there will be a visual change of egg white and yolk. Their structures will be more solid (the occurrence of thickening process) because salinization will lead to protein denaturalization. Consequently, it has an influence as well towards the content of egg white protein of duck egg. This study is aimed to explore the impact of various time of pickling egg towards egg white protein of duck egg. Method : The study where takes place in a laboratories, is a true experimental study for the reason that the researcher must provide intervention, hence all of potentially confounding variables are manageable. Samples that had been used in this study are duck eggs which were bought from North Brebes. This study is expected to generate data from four various time of pickling egg and control (no treatment). Since there are four samples, accordingly the number of data resulted are twenty. The resulted data will be descriptively presented in table, graph, presentation, and narration. Result : Protein level examination within duck white egg shows changes in protein levels that occurs in every variation of pickling egg time, where the average results of the assay of duck egg white protein is 14.94% without treatment (control), in five days of pickling time is 13.68%, in seven days of pickling time is 13.29%, in nine days of pickling time is 12.87% and eleven days of pickling time is 12.78%. Conclusion : There is a significant impact among the period of pickling time to the protein level degradation of duck white egg. Keywords : Duck egg, period of pickling time, level protein of duck white egg.

Relationship of Flux Harmonics and Partitioning Method of Zero Voltage Vectors

Aiming the traditional space vector PWM with equal duration of the two zero vectors, the paper introduces the principle of SVPWM. The harmonic is analyzed based on formula expression, numerical computation, and graph presentation. The optimal zero vector partitioning method is proposed. Finally, the interpolation method and discretization of the optimal method are proposed and discussed.

Number Enumeration of Planar Pin-Joined Driving Mechanisms

This paper deals with the development of a new method for the morphological number enumeration for planar pin-jointed driving mechanisms applied in robotic design. The method is based on linkage graph presentation by using geometrical symmetries of kinematic chains and combinatorial analysis. The restricting criteria being the position of: the frame, the end-effector and the actuators (motors) of the robot, different cases of symmetries for planar pin-jointed driving mechanisms of mobility 1 and 2 are addressed. New expressions for calculating the number of different possibilities to position the frame, the end-effector and the actuators in a mechanism are presented, thus reducing the number of the possible solutions by avoiding those that are isomorphic. A further consequence of the present work is the ability for it to be extended to mechanisms with three degrees of mobility and more.