629 Hits in 3.6 sec

Structure Guided Salient Region Detector

S. Fan, F.P. Ferrie
2008 Procedings of the British Machine Vision Conference 2008  
This paper presents a novel method for detection of image interest regions called Structure Guided Salient Regions (SGSR).  ...  s Salient Region detector by exploiting image structure information. The detected SGSRs are highly distinct and very selective.  ...  , Harris-Affine detector (haraff), MSER, Intensity extrema-based Region detector (ibraff), and Edge-based Region detector (ebraff).  ... 
doi:10.5244/c.22.43 dblp:conf/bmvc/FanF08 fatcat:cffodsck7za5db2xlkuh54sqly

Recent Advances in Features Extraction and Description Algorithms: A Comprehensive Survey [article]

Ehab Salahat, Murad Qasaimeh
2017 arXiv   pre-print
This paper is dedicated to provide a comprehensive overview on the state-of-the-art and recent advances in feature detection and description algorithms.  ...  Computer vision is one of the most active research fields in information technology today.  ...  SIFT Derievatives SIFT algorithm has a local feature detector and local histogram-based descriptor.  ... 
arXiv:1703.06376v1 fatcat:rqcg77mfwnfchas36crbn7zikq

An Enhanced MSER Pruning Algorithm for Detection and Localization of Bangla Texts from Scene Images

Rashedul Islam, Rafiqul Islam, Kamrul Talukder
2019 ˜The œinternational Arab journal of information technology  
Firstly, MSER regions are generated from the input color image by converting it into gray scale image.  ...  The efficiency of text recognition depends on the efficiency of text localization. So, the main goal of the proposed method is to detect and localize text regions with high accuracy.  ...  They also developed a powerful low-level detector called CE-MSER which extends the widely used MSER by enhancing intensity contrast between text pattern and background.  ... 
doi:10.34028/iajit/17/3/11 fatcat:sngl2hu67vfcvak3gwn3b6ut4y

Beyond MSER: Maximally Stable Regions using Tree of Shapes

Petra Bosilj, Ewa Kijak, Sébastien Lefèvre
2015 Procedings of the British Machine Vision Conference 2015  
This article explores the application of a tree-based feature extraction algorithm for the widely-used MSER features, and proposes a Tree of Shapes based detector of Maximally Stable Regions.  ...  The proposed descriptor is compared to the standard MSER implementation as well as a tree-based MSER implementation, achieving competitive results in the matching setup and outperforming the baseline MSER  ...  We focus here on the Maximally Stable Extremal Regions (MSER), a fast detector based on image intensity, responding to blobs of high contrast and producing affine invariant, highly featured regions of  ... 
doi:10.5244/c.29.169 dblp:conf/bmvc/BosiljKL15 fatcat:ok2wvclkorbb5hjh6t7thhy7fe


O. Akcay, E. O. Avsar
2017 The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
Later on, the pre-processed images were evaluated using current image detection and feature extraction methods Maximally Stable Extremal Regions (MSER) and Speeded Up Robust Features (SURF) algorithms.  ...  A successful image matching is essential to provide an automatic photogrammetric process accurately.  ...  First, histogram equalization including channel-based local contrast-limited adaptive enhancement were used for intensity adjustment (Zuiderveld, 1994) .  ... 
doi:10.5194/isprs-archives-xlii-1-w1-575-2017 fatcat:a33xo5dt6bddvbzolqebndtmpu

A Color-Based Interest Operator [chapter]

Marta Penas, Linda G. Shapiro
2009 Lecture Notes in Computer Science  
Our operator is closely related to the grayscale MSER but it works on the HSV color space, as opposed to the most popular operators in the literature, which are intensity based.  ...  It combines a fine and a coarse overlapped quantization of the HSV color space to find maximally stable extremal regions on each of its components and combine them into a final set of regions that are  ...  Also, the stability of the regions is fragile and very dependent on the input parameters. In this paper we present a color interest operator, called COLOR, based on the same principles as the MSER.  ... 
doi:10.1007/978-3-642-04146-4_103 fatcat:nizs7j5ulvdqhj66kgoszw7jbu

Scene text detection via extremal region based double threshold convolutional network classification

Wei Zhu, Jing Lou, Longtao Chen, Qingyuan Xia, Mingwu Ren, Yuanquan Wang
2017 PLoS ONE  
In this paper, we present a robust text detection approach in natural images which is based on region proposal mechanism.  ...  Given a natural image, character candidates are extracted from three channels in a perception-based illumination invariant color space by saliency-enhanced MSER algorithm.  ...  We further apply color histogram-based contrast method inspired by [49] to enhance contrast of the small-size regions.  ... 
doi:10.1371/journal.pone.0182227 pmid:28820891 pmcid:PMC5562312 fatcat:albr46hkvbh2xlwjio7oe7jxpq

Classification of Wild Animals based on SVM and Local Descriptors

Slavomir Matuska, Robert Hudec, Patrik Kamencay, Miroslav Benco, Martina Zachariasova
2014 AASRI Procedia  
This presented paper proposed about local key point detectors and descriptors, bag of key points method in combination with SVM.  ...  In experiments a total 6 key points detectors, namely, SURF, SIFT, STAR, FAST, MSER and GFTT were used.  ...  SURF: detector is based on the determinant of the Hessian matrix. The discriminant value is used to classify the maximum and minimum of the function by second order derivative test [4, 7] .  ... 
doi:10.1016/j.aasri.2014.09.006 fatcat:mx6otz6wyjfabp2vrt2fw2tiyi

Visual landmark generation and redetection with a single feature per frame

S Frintrop, A B Cremers
2010 2010 IEEE International Conference on Robotics and Automation  
The approach is based on the assumption that highly discriminative regions are easily redetectable in subsequent frames as well as in frames visited from different viewpoints.  ...  We show that we can represent a large environment with a few salient landmarks and that a large percentage of these landmarks is robustly redetectable from different viewpoints.  ...  Visual localization and navigation is often based on landmarks, that means on objects or regions in the environment that serve as reference points for the robot.  ... 
doi:10.1109/robot.2010.5509798 dblp:conf/icra/FrintropC10 fatcat:o6ya76yi7jhdfdp3i4knpbocwy

Detection and Recognition of Road Traffic Signs - A Survey

Sumi K., Arun Kumar
2017 International Journal of Computer Applications  
A comparitive study is performed on these techniques and their performance is explained.  ...  This paper presents a survey on the detection and recognition of traffic signs which has a number of important application areas that include advanced driver assistance systems, road surveying and autonomous  ...  A technique based on Maximally Stable Extremal Regions (MSER) region and canny edge detector was also supervised for character recognition in traffic sign detection process.  ... 
doi:10.5120/ijca2017913038 fatcat:6gap7z6otrgubjrvdxi23za6fm

Joint Spectral Correspondence for Disparate Image Matching

Mayank Bansal, Kostas Daniilidis
2013 2013 IEEE Conference on Computer Vision and Pattern Recognition  
We show experimental results of our approach on a public dataset of challenging image pairs and demonstrate significant performance improvements over state-of-the-art. * The authors are grateful for support  ...  The lack of local intensity or gradient patterns in these images makes the application of pixellevel descriptors like SIFT infeasible.  ...  Our MSER detector does not output a detection score and so we only present repeatability numbers based on ordering by scale.  ... 
doi:10.1109/cvpr.2013.361 dblp:conf/cvpr/BansalD13 fatcat:e3zjws7wnfew5f6eriii2q6jeq

Localization accuracy of region detectors

Andreas Haja, Bernd Jahne, Steffen Abraham
2008 2008 IEEE Conference on Computer Vision and Pattern Recognition  
In contrast to existing methods that are based on the descriptor distance of region correspondences, the new measure is pre-computed on each image individually.  ...  Also, it is shown that localization accuracy linearly depends on region scale for some detectors, which may thus be used as a pre-selection criterion for the removal of error-prone regions.  ...  Introduction In this work, five popular region detectors are compared with regard to localization accuracy in position and region shape: the edge-based (EBR) and intensity-based (IBR) region detectors  ... 
doi:10.1109/cvpr.2008.4587829 dblp:conf/cvpr/HajaJA08 fatcat:usbhwwl4yrf3pni7ohucagckay

Automatic text extraction and character segmentation using maximally stable extremal regions [article]

Nitigya Sambyal, Pawanesh Abrol
2016 arXiv   pre-print
The paper proposes a novel text extraction and character segmentation algorithm using Maximally Stable Extremal Regions as basic letter candidates.  ...  Text detection and segmentation is an important prerequisite for many content based image analysis tasks.  ...  ON MSER AND CONNECTED COMPONENT METHOD TABLE II : II EXTRACTION OF RUSSIAN CHARACTER SET BASED ON MSER AND CONNECTED COMPONENT METHODconnects each character underneath it giving a single component.  ... 
arXiv:1608.03374v1 fatcat:vpd2q4bsz5alfaxaql2cnv55mu

Human Activity Recognition Based on Object Detection

Aarsh Joshi, Hetal Parmar, Khyati Jain, Chandni Shah, ,Prof. Vaishali R Patel
2017 IOSR Journal of Computer Engineering  
This technique determines the distance of the object from a human and provides the results of activities based on the distance.  ...  This paper presents a new approach by combining both the techniques for extra usability and accurate detection of activities.  ...  At a high level, MSER can be explained, by thinking of the intensity profile of an image representing a series of buckets.  ... 
doi:10.9790/0661-1902012632 fatcat:hajxtuwmcrd23pwptk7sw37oqi

Machine Vision based Traffic Sign Detection Methods: Review, Analyses and Perspectives

Chunsheng Liu, Shuang Li, Faliang Chang, Yinhai Wang
2019 IEEE Access  
In recent years, there have been a large number of TSD algorithms based on machine vision and pattern recognition. In this paper, a comprehensive review of the literature on TSD is presented.  ...  We divide the reviewed detection methods into five main categories: color-based methods, shape-based methods, color-and shape-based methods, machine-learning-based methods, and LIDAR-based methods.  ...  Hence, the color enhancement methods and suitable parameters of MSERs are very important for these methods. 2) HIGH CONTRASTED MARGIN REGIONS BASED DETECTION Unlike MSERs that can detect high-contrast  ... 
doi:10.1109/access.2019.2924947 fatcat:urejf36h4rauvgur6qvj63yzxe
« Previous Showing results 1 — 15 out of 629 results