35,132 Hits in 7.9 sec

Camera-based Sudoku recognition with deep belief network

Baptiste Wicht, Jean Hennebert
2014 2014 6th International Conference of Soft Computing and Pattern Recognition (SoCPaR)  
In this paper, we propose a method to detect and recognize a Sudoku puzzle on images taken from a mobile camera. The lines of the grid are detected with a Hough transform.  ...  The grid is then recomposed from the lines. The digits position are extracted from the grid and finally, each character is recognized using a Deep Belief Network (DBN).  ...  ACKNOWLEDGMENT We would like to thank all the people who contributed to the dataset by sending us Sudoku images taken from their phones, in particular Patrick Anagnostaras.  ... 
doi:10.1109/socpar.2014.7007986 dblp:conf/socpar/WichtH14 fatcat:jyxkv3z37fecfddyi2jviqc7w4

Extensible System for Optical Character Recognition of Maintenance Documents

John Anthony Labarga, Amardeep Singh, Vera Zaychik Moffitt
2018 Proceedings of the Annual Conference of the Prognostics and Health Management Society, PHM  
This is particularly the case in maintaining legacy systems, and when external factors (e.g. security concerns, environment, training procedures) make it infeasible to record data in a computer system  ...  To implement analytics or automated monitoring, these documents must later be converted to digital copies, which can be ingested into a database.  ...  Cell extraction To detect the row and column lines forming the cell grid, the Canny edge detector (Canny, 1986 ) works well.  ... 
doi:10.36001/phmconf.2018.v10i1.480 fatcat:bribomxkkbfzjnhe4sygmaz65e

Multi-scale Analysis of Discrete Contours for Unsupervised Noise Detection [chapter]

Bertrand Kerautret, Jacques-Olivier Lachaud
2009 Lecture Notes in Computer Science  
To overcome this issue, we propose an original strategy to detect locally both the amount of noise and the meaningful scales of each point of a digital contour.  ...  From a given maximal observation scale, the proposed approach does not require any parameter tuning and is easy to implement. We demonstrate its effectiveness on several datasets.  ...  We consider that our digital object O is the digitization of some Euclidean shape X at a grid step h, choosing for instance again the Gauss digitization.  ... 
doi:10.1007/978-3-642-10210-3_15 fatcat:p5o2qqoewbgbhg7hf5k4fp54ia

An Elementary Algorithm for Digital Arc Segmentation

David Coeurjolly, Laure Tougne, Yan Gérard, Jean-Pierre Reveillès
2001 Electronical Notes in Theoretical Computer Science  
After recalling the geometrical meaning of the separating circle problem, we present an incremental algorithm to segment a discrete curve into digital arcs. ?  ...  This paper concerns the digital circle recognition problem, especially in the form of the circular separation problem.  ...  Acknowledgements We would like to thank reviewers for their fruitful comments.  ... 
doi:10.1016/s1571-0661(04)80997-6 fatcat:vpgkmfm74vfnfmk6blm3vjfpy4

An elementary algorithm for digital arc segmentation

D. Coeurjolly, Y. Gérard, J.-P. Reveillés, L. Tougne
2004 Discrete Applied Mathematics  
After recalling the geometrical meaning of the separating circle problem, we present an incremental algorithm to segment a discrete curve into digital arcs. ?  ...  This paper concerns the digital circle recognition problem, especially in the form of the circular separation problem.  ...  Acknowledgements We would like to thank reviewers for their fruitful comments.  ... 
doi:10.1016/j.dam.2003.08.003 fatcat:tkxvkuqsdzb4tb5fdgr3nr4y6a

Autonomous operation of novel elevators for robot navigation

Ellen Klingbeil, Blake Carpenter, Olga Russakovsky, Andrew Y Ng
2010 2010 IEEE International Conference on Robotics and Automation  
To verify our approach, we collected a dataset of 150 pictures of elevator panels from more than 60 distinct elevators, and performed extensive offline testing.  ...  To address the task of detecting, localizing, and labeling the buttons, we use state-of-the-art vision algorithms along with machine learning techniques to take advantage of contextual features.  ...  First, we use the fact that most elevator buttons lie on a grid to infer missed detections and remove false positives.  ... 
doi:10.1109/robot.2010.5509466 dblp:conf/icra/KlingbeilCRN10 fatcat:ue26ql6rhrggxek3sw3b7weuia

EagleMine: Vision-Guided Mining in Large Graphs [article]

Wenjie Feng, Shenghua Liu, Christos Faloutsos, Bryan Hooi, Huawei Shen, Xueqi Cheng
2017 arXiv   pre-print
Given a graph with millions of nodes, what patterns exist in the distributions of node characteristics, and how can we detect them and separate anomalous nodes in a way similar to human vision?  ...  To identify potentially anomalous microclusters, EagleMine also a designates score to measure the suspiciousness of outlier groups (i.e. node clusters) and outlier nodes, detecting bots and anomalous users  ...  Output: Water-level tree T of islands. 1: T = non-empty grid set in H as root. 2: for water level r = 0 to log h max by step s do 3: H r = assign h i,j ∈ H to zero if log h i,j < r. 4: H r = H r • E.  ... 
arXiv:1710.08756v1 fatcat:bawp22rrbbhv5lauz32nt2gasm

Breast Skin Line Segmentation on Digital Mammogram using Fractal Approach

S. Don, Dugki Min
2016 Indian Journal of Science and Technology  
The second stage employed a fractal based approach for segmentation and the third step detects the border region from the segmented image.  ...  Methods/Analysis: A three stage processing pipeline was developed to segment the breast skin line.  ...  for 14 Do a line fit of (log(R/ε)), log(((R/ε 3 )E_ε) 15 where R is a window of suitable size 16 end for Detection of Breast Boundary using Fractal Dimension In our approach, modified variation method  ... 
doi:10.17485/ijst/2016/v9i31/85420 fatcat:eenu42tzrfft3jgap6dchcl42q

Making real games virtual: Tracking board game pieces

Steven Scher, Ryan Crabb, James Davis
2008 Pattern Recognition (ICPR), Proceedings of the International Conference on  
The same game is often played in real and virtual worlds. We integrate in-person and on-line playing of board games such as Go, bringing the real world into the virtual world.  ...  A player may record an in-person game by placing their camera on the table next to the game board, taking photos of the game.  ...  The Hough accumulator representation of the detected lines may be directly used to detect the grid, as in [1] and [6] .  ... 
doi:10.1109/icpr.2008.4761307 dblp:conf/icpr/ScherCD08 fatcat:56uqjiklyzhtrchxqtpzqxcc3q

Fake News Detection with Different Models [article]

Sairamvinay Vijayaraghavan, Ye Wang, Zhiyuan Guo, John Voong, Wenda Xu, Armand Nasseri, Jiaru Cai, Linda Li, Kevin Vuong, Eshan Wadhwa
2020 arXiv   pre-print
This is a paper for exploring various different models aiming at developing fake news detection models and we had used certain machine learning algorithms and we had used pretrained algorithms such as  ...  TFIDF and CV and W2V as features for processing textual data.  ...  In our outlier detections and removals, Isolation Forest has been applied to three different features. Generated from TFIDF, CV, WV.  ... 
arXiv:2003.04978v1 fatcat:tvvsqytmlnhblanlgxritjqjzm

Multiple Lanes Identification using Novel Region-Based Iterative Seed Method

But, due to the varying road conditions, it is very difficult to detect the lane. The computer vision and machine learning approaches are presents in most of the articles.  ...  In this proposed method, the target grids are selected form the road lane. Distance is calculated for road and lane.  ...  The information is in three types word-marking, pictogram marking and line-marking. Word-markings are usually used to deliver a text message to drivers.  ... 
doi:10.35940/ijitee.i1125.0789s419 fatcat:bm3kextd2vcnlfnqktunwze6yq

Recent improvements in tensor scale computation and its applications to medical imaging

Ziyue Xu, Milan Sonka, Punam K. Saha, Josien P. W. Pluim, Benoit M. Dawant
2009 Medical Imaging 2009: Image Processing  
Recently, we have improved the t-scale computation algorithm by: (1) optimizing digital representations for LoG and DoG kernels for edge detection and (2) ellipse fitting by using minimization of both  ...  Preliminary results demonstrate the method's potential in estimating deformation between two images. Downloaded from SPIE Digital Library on 09 Dec 2010 to  ...  In this paper, we improve the t-scale computation algorithm by -(1) optimizing digital LoG and DoG kernels for edge detection at varying kernel sizes and (2) ellipse fitting by using minimization of both  ... 
doi:10.1117/12.811360 dblp:conf/miip/XuSS09 fatcat:egx3kixxnzg6jppaomw22h4j2q

GPU Accelerated Automated Feature Extraction From Satellite Images

K. Phani Tejaswi, D. Shanmukha Rao, Thara Nair, Prasad A.V.V
2013 International Journal of Distributed and Parallel systems  
The image is passed through a hybrid median filter to remove the salt and pepper noise from the image.  ...  Finally the user is at liberty to apply the image smoothing step depending on the noise content in the extracted image.  ...  ACKNOWLEDGMENTS The authors desire to acknowledge the support and excellent guidance made available by Dr. V. K. Dadhwal, Director, NRSC in successfully implementing this work.  ... 
doi:10.5121/ijdps.2013.4201 fatcat:ru2shy3eonc4tewq57lvabc36e

Service-oriented grid computing system for digital rights management (GC-DRM)

Min-Jen Tsai, Yuan-Fu Luo
2009 Expert systems with applications  
In this paper, we proposed a service-oriented digital rights management (DRM) platform based on grid computing (called GC-DRM) which is in the compliance of Grid Portal standards by using porlet.  ...  In addition, GC-DRM has been applied to analyze the robustness of digital watermark by filter bank selection and the performance can be improved in the aspect of speedup, stability and processing time  ...  Acknowledgement This work was supported by the National Science Council in Taiwan, Republic of China, under Grant NSC96-2416-H009-015 and NSC97-2416-H009-034.  ... 
doi:10.1016/j.eswa.2009.02.066 fatcat:bau3ebx7snblbbz5jdjmk7qqii

Watermarking road maps against crop and merge attacks

Kai Jiang, Kenny Q. Zhu, Yan Huang, Xiaobin Ma
2013 Proceedings of the first ACM workshop on Information hiding and multimedia security - IH&MMSec '13  
The approach achieves almost 100% detection accuracy for moderate crop and merge attacks, and over 80% accuracy with more than 95% of the original map cropped and removed.  ...  Furthermore, the method introduces very little distortion to the original map: to effectively protect a 23.5MB Minneapolis-St.Paul map against crop and merge as well as other common attacks, only 423 bits  ...  The standard watermarking framework for digital road maps (shown in Figure 2 ) also adopts a two-step approach.  ... 
doi:10.1145/2482513.2482526 dblp:conf/ih/JiangZHM13 fatcat:u77l3omgcnevxc6uo33yhshju4
« Previous Showing results 1 — 15 out of 35,132 results