LoG and Structural Based Arbitrary Oriented Multilingual Text Detection in Images/Video

2022 ◽  
pp. 987-1003
Author(s):  
H. T. Basavaraju ◽  
V.N. Manjunath Aradhya ◽  
D. S. Guru ◽  
H. B. S. Harish
Keyword(s):  
Performance Measures ◽  
Research Work ◽  
Text Detection ◽  
Double Line ◽  
Precise Meaning ◽  
Clear Explanation ◽  
Text Information ◽  
Multilingual Text ◽  
High Level ◽  
F Measure

Text in an image or a video affords more precise meaning and text is a prominent source with a clear explanation of the content than any other high-level or low-level features. The text detection process is a still challenging research work in the field of computer vision. However, complex background and orientation of the text leads to extremely stimulating text detection tasks. Multilingual text consists of different geometrical shapes than a single language. In this article, a simple and yet effective approach is presented to detect the text from an arbitrary oriented multilingual image and video. The proposed method employs the Laplacian of Gaussian to identify the potential text information. The double line structure analysis is applied to extract the true text candidates. The proposed method is evaluated on five datasets: Hua's, arbitrarily oriented, multi-script robust reading competition (MRRC), MSRA and video datasets with performance measures precision, recall and f-measure. The proposed method is also tested on real-time video, and the result is promising and encouraging.

LoG and Structural Based Arbitrary Oriented Multilingual Text Detection in Images/Video

2018 ◽  
Vol 7 (3) ◽  
pp. 1-16 ◽  
Author(s):  
Basavaraju H. T. ◽  
Manjunath Aradhya V.N. ◽  
Guru D. S. ◽  
Harish H. B. S.
Keyword(s):  
Research Work ◽  
Text Detection ◽  
Double Line ◽  
Precise Meaning ◽  
Clear Explanation ◽  
Geometrical Shapes ◽  
Challenging Research ◽  
Text Information ◽  
Multilingual Text ◽  
High Level

Text in an image or a video affords more precise meaning and text is a prominent source with a clear explanation of the content than any other high-level or low-level features. The text detection process is a still challenging research work in the field of computer vision. However, complex background and orientation of the text leads to extremely stimulating text detection tasks. Multilingual text consists of different geometrical shapes than a single language. In this article, a simple and yet effective approach is presented to detect the text from an arbitrary oriented multilingual image and video. The proposed method employs the Laplacian of Gaussian to identify the potential text information. The double line structure analysis is applied to extract the true text candidates. The proposed method is evaluated on five datasets: Hua's, arbitrarily oriented, multi-script robust reading competition (MRRC), MSRA and video datasets with performance measures precision, recall and f-measure. The proposed method is also tested on real-time video, and the result is promising and encouraging.

MFECN: Multi-level Feature Enhanced Cumulative Network for Scene Text Detection

2021 ◽  
Vol 17 (3) ◽  
pp. 1-22
Author(s):  
Zhandong Liu ◽  
Wengang Zhou ◽  
Houqiang Li
Keyword(s):  
Text Detection ◽  
Detection Algorithms ◽  
Features Fusion ◽  
Irregular Shapes ◽  
Scene Text Detection ◽  
Scene Text ◽  
Text Information ◽  
Multi Level ◽  
Public Datasets ◽  
High Level

Recently, many scene text detection algorithms have achieved impressive performance by using convolutional neural networks. However, most of them do not make full use of the context among the hierarchical multi-level features to improve the performance of scene text detection. In this article, we present an efficient multi-level features enhanced cumulative framework based on instance segmentation for scene text detection. At first, we adopt a Multi-Level Features Enhanced Cumulative ( MFEC ) module to capture features of cumulative enhancement of representational ability. Then, a Multi-Level Features Fusion ( MFF ) module is designed to fully integrate both high-level and low-level MFEC features, which can adaptively encode scene text information. To verify the effectiveness of the proposed method, we perform experiments on six public datasets (namely, CTW1500, Total-text, MSRA-TD500, ICDAR2013, ICDAR2015, and MLT2017), and make comparisons with other state-of-the-art methods. Experimental results demonstrate that the proposed Multi-Level Features Enhanced Cumulative Network (MFECN) detector can well handle scene text instances with irregular shapes (i.e., curved, oriented, and horizontal) and achieves better or comparable results.

Development and Achievements of SSM Processes for High Performance Components

2016 ◽  
Vol 256 ◽  
pp. 319-327 ◽  
Author(s):  
Mario Rosso ◽  
Ildiko Peter ◽  
Ivano Gattelli
Keyword(s):  
Industrial Development ◽  
High Performance ◽  
Rear Axle ◽  
Research Work ◽  
Industrial Applications ◽  
Alternative Methods ◽  
Front Suspension ◽  
Large Numbers ◽  
Automotive Parts ◽  
High Level

During the last decades under the enthusiastic and competent guidance of Mr Chiarmetta SSM processes attained in Italy at Stampal Spa (Torino) an unquestionable high level of industrial development with the production of large numbers of high performance automotive parts, like variety of suspension support, engine suspension mounts, steering knuckle, front suspension wheel, arm and rear axle. Among the most highlighted findings SSM processes demonstrated their capability to reduce the existing gap between casting and forging, moreover during such a processes there are the opportunity to better control the defect level.Purpose of this paper is to highlight the research work and the SSM industrial production attained and developed by Mr G.L. Chiarmetta, as well as to give an overview concerning some alternative methods for the production of enhanced performance light alloys components for critical industrial applications and to present an analysis of a new rheocasting process suitable for the manufacturing of high performance industrial components.

Oscillatory Behaviors of Delayed p53 Regulatory Network with microRNA 192 in DNA Damage Response

2021 ◽  
Vol 31 (02) ◽  
pp. 2150020
Author(s):  
Chunyan Gao ◽  
Fangqi Chen
Keyword(s):  
Dna Damage ◽  
Time Delay ◽  
Dna Damage Response ◽  
Positive Feedback ◽  
Regulatory Network ◽  
Feedback Loop ◽  
Research Work ◽  
Positive Feedback Loop ◽  
Damage Response ◽  
High Level

This study develops a general model of delayed p53 regulatory network in the DNA damage response by introducing microRNA 192-mediated positive feedback loop based on the existing research work. Through theoretical analysis and numerical simulation, we find that the delay as a bifurcation parameter can drive the p53-Mdm2 module to undergo a supercritical Hopf bifurcation, thereby producing oscillation behavior. Moreover, we demonstrate how the positive feedback loop formed by p53* and microRNA 192 (miR-192) with the feature of double-negative regulation produces oscillations. Further, a comparison is given to demonstrate that microRNA 192-mediated positive feedback loop affects the robustness of system oscillations. In addition, we show that ataxia telangiectasia mutated kinase (ATM), once activated by DNA damage, makes p53* undergo two Hopf bifurcations. These results reveal that both time delay and miR-192 play tumor suppressing roles by promoting p53 oscillation or high level expression, which will provide a perspective for promoting the development of anti-cancer drugs by targeting miR-192 and time delay.

scholarly journals Prediction and Analysis of Extracting Relations using Spacy Model

2020 ◽  
Vol 8 (6) ◽  
pp. 3281-3287
Keyword(s):  
Natural Language ◽  
Information Extraction ◽  
Performance Measures ◽  
Text Summarization ◽  
Language Understanding ◽  
Language Generation ◽  
Automatic Text Summarization ◽  
Structured Information ◽  
Automatic Text ◽  
F Measure

Text is an extremely rich resources of information. Each and every second, minutes, peoples are sending or receiving hundreds of millions of data. There are various tasks involved in NLP are machine learning, information extraction, information retrieval, automatic text summarization, question-answered system, parsing, sentiment analysis, natural language understanding and natural language generation. The information extraction is an important task which is used to find the structured information from unstructured or semi-structured text. The paper presents a methodology for extracting the relations of biomedical entities using spacy. The framework consists of following phases such as data creation, load and converting the data into spacy object, preprocessing, define the pattern and extract the relations. The dataset is downloaded from NCBI database which contains only the sentences. The created model evaluated with performance measures like precision, recall and f-measure. The model achieved 87% of accuracy in retrieving of entities relation.

scholarly journals THE MAIN RESERVE OF ACCELERATED EFFECTIVE OPENING OF OIL AND GAS FIELDS IN UKRAINE

2019 ◽  
pp. 31-37
Author(s):  
V. T. Kryvosheyev ◽  
V. V. Makogon ◽  
Ye. Z. Ivanova
Keyword(s):  
Shale Gas ◽  
Oil And Gas ◽  
Research Work ◽  
Oil And Gas Industry ◽  
The United States ◽  
Resource Base ◽  
Hydrocarbon Production ◽  
Oil And Gas Fields ◽  
High Level ◽  
Gas Fields

Economic hardship in Ukraine during the years of independence led to a sharp reduction of exploration work on oil and gas, a drop in hydrocarbon production, a decrease in inventories and a sharp collapse of research work to ensure the growth of hydrocarbon reserves.The hydrocarbon potential of various sources of Ukrainian subsoil is quite powerful and can provide future energy independence of the country. Potential hydrocarbon resources in traditional traps of various types are exhausted by only 25 %. Ukraine has recently experienced so-called “shale gas boom”. The experience of extraction of shale gas in desert areas of the United States can not be repeated in densely populated Ukraine in the absence of such powerful shale strata, resource base, necessary infrastructure, own technologies and techniques and economic, environmental and social risks.Taking into account the fuel and energy problems of the state, we constantly throughout the years of independence oriented the oil and gas industry and the authorities on the active use of our own reserves and opportunities for accelerated opening of new oil and gas fields.The results of geological exploration work in the old oil and gas basins at the high level of their study indicate that deposits in non-structural traps dominate among open deposits.A complex of sequence-stratigraphical, lithology-facies and lithology-paleogeographical studies is being successfully used to forecast undeformational traps in well-studied oil and gas bearing basin of the Ukraine – the Dniprovsko-Donetsky basin. The authors predict wide development of stratigraphic, lithologic, tectonic and combined traps in terrigenous sediments of Tournaisian and Visean age, reef-carbonate massifs of the lower Tournaisian, lower and middle Visean age and others. They should become the basis for exploration of oil and gas fields for the near and medium term and open the second breath of the basin.

Genetic Diversity Analysis among 27 Indian Goat Populations using Microsatellite Markers

2020 ◽  
Author(s):  
Rajkumar Sah ◽  
Santpal Dixit
Keyword(s):  
Genetic Diversity ◽  
Association Studies ◽  
Research Work ◽  
Diversity Analysis ◽  
National Bureau ◽  
Genetic Diversity Analysis ◽  
Well Differentiated ◽  
Inbreeding Level ◽  
High Level ◽  
Indian Goat

Background: Livestock genetic diversity studies focus on their within and diversity, breed history, adaptive variations, ancestral information, site of domestication and parentage testing and assess the genetic uniformity, admixture or subdivision, inbreeding, or introgression in the population which is helpful in breed formation and their sustainable utilization.Methods: The present research work was conducted during the year 2016-17 at National Bureau of Animal Genetics Resources, Karnal-132001. STR data of 25 markers on 1237 random samples of 27 goat populations was used for analysis. The genetic diversity analysis of new population viz: Narayanpatna, Raighar, Kalahandi, Malkangiri of Odisha state and Rohilkhandi (UK) and their association studies with other Indian goat breeds was performed.Result: It was found that used markers are highly polymorphic- and the studied breeds/population showed great diversity and distributed mostly on the basis of physio-geographical condition and type of production but among new populations diversity was least which might be due to exchange of animal for breeding purposes. The studied new goat populations were well differentiated from other goat breeds which might be due to physio-geographical condition and breeding practices, so these may be considered as separate breeds/populations. In conclusion, the results showed high level of conserved genetic diversity in the Indian goat breeds. The smaller and isolated new population showed less diversity and a higher inbreeding level as compared to registered breeds.

A New Approach to Supply Chain Performance Measurement

2016 ◽  
pp. 111-128
Author(s):  
Amit Kumar Marwah ◽  
Girish Thakar ◽  
R. C. Gupta
Keyword(s):  
Factor Analysis ◽  
Supply Chain ◽  
Performance Measures ◽  
Research Work ◽  
Primary Data ◽  
Supply Chain Performance ◽  
New Approach ◽  
Factors Affecting ◽  
Manufacturing Organizations ◽  
Overall Reliability

Existing research work has established that many of today's manufacturing organizations have failed to develop a comprehensive supply chain performance measures. In this chapter, the authors intend to empirically assess the effects of supplier buyer relations and human metrics on supply chain performance in the context of Indian manufacturing organizations. After rigorous literature review, total 18 variables have been identified which are later on reduced in number by factor analysis. As a pilot study, primary data is collected from 100 manufacturing organizations by means of a questionnaire and a scale is developed. On a sample size of 100, the proposed hypotheses are tested by applying two-tailed tests. t-test and factor analysis resulted in 5 factors, 2 related to supplier-buyer relations and 3 related to human metrics. The overall reliability of the scale comes out to be 0.697. The research work provides a new approach to the manufacturing organizations to understand the factors affecting supply chain performance. The present study is limited to Indian manufacturing organizations.

Analytical Model for High–Level Area Estimation of FPGA Design

2016 ◽  
Vol 7 (2) ◽  
pp. 35-44
Author(s):  
Rachna Singh ◽  
Arvind Rajawat
Keyword(s):  
Early Stage ◽  
Estimation Error ◽  
Research Work ◽  
Performance Estimation ◽  
Estimation Accuracy ◽  
Accurate Estimation ◽  
Fpga Design ◽  
Estimation Model ◽  
Area Estimation ◽  
High Level

FPGAs have been used as a target platform because they have increasingly interesting in system design and due to the rapid technological progress ever larger devices are commercially affordable. These trends make FPGAs an alternative in application areas where extensive data processing plays an important role. Consequently, the desire emerges for early performance estimation in order to quantify the FPGA approach. A mathematical model has been presented that estimates the maximum number of LUTs consumed by the hardware synthesized for different FPGAs using LLVM.. The motivation behind this research work is to design an area modeling approach for FPGA based implementation at an early stage of design. The equation based area estimation model permits immediate and accurate estimation of resources. Two important criteria used to judge the quality of the results were estimation accuracy and runtime. Experimental results show that estimation error is in the range of 1.33% to 7.26% for Spartan 3E, 1.6% to 5.63% for Virtex-2pro and 2.3% to 6.02% for Virtex-5.

scholarly journals A Scene Text Detector for Text with Arbitrary Shapes

2020 ◽  
Vol 2020 ◽  
pp. 1-11
Author(s):  
Weijia Wu ◽  
Jici Xing ◽  
Cheng Yang ◽  
Yuxing Wang ◽  
Hong Zhou
Keyword(s):  
State Of The Art ◽  
Recognition Task ◽  
Text Detection ◽  
Supplementary Information ◽  
Complex Environment ◽  
Irregular Shapes ◽  
Scene Text Detection ◽  
Scene Text ◽  
F Measure ◽  
Instance Segmentation

The performance of text detection is crucial for the subsequent recognition task. Currently, the accuracy of the text detector still needs further improvement, particularly those with irregular shapes in a complex environment. We propose a pixel-wise method based on instance segmentation for scene text detection. Specifically, a text instance is split into five components: a Text Skeleton and four Directional Pixel Regions, then restoring itself based on these elements and receiving supplementary information from other areas when one fails. Besides, a Confidence Scoring Mechanism is designed to filter characters similar to text instances. Experiments on several challenging benchmarks demonstrate that our method achieves state-of-the-art results in scene text detection with an F-measure of 84.6% on Total-Text and 86.3% on CTW1500.

Sign in / Sign up

Export Citation Format

Share Document