slugAnalysis of Two Image Segmentation Schemes Based on Its Error Rate
Journal Title: International Journal for Research in Applied Science and Engineering Technology (IJRASET) - Year 2014, Vol 2, Issue 4
Abstract
There are various image segmentation techniques available. Each on e is distinct according to the parameters and segmentation constraints. Also there are region based segmentation and boundary based segmentation. Here two such image segmentation methods: Graph cut and Maximal Similarity based Region Merging are considered and the two are analysed on the basis of error rate. Graph cut involves finding the distance between pixels and minimizing an energy function. MSRM approach finds the similarity between two regions using a similarity measure. The experiments are conducted and the results obtained are formulated. The criteria and pre segmentation procedure are different for the two methods. One is designed for the segmentation of texture images and the other for colour images. Graph cut combines both region based and boundary based methods whereas MSRM is a region based segmentation technique. The advantages and disadvantages of these methods are analysed on different aspects.
Authors and Affiliations
Parvathy Raghunandanan, S. John Livingston
Keywords
Related Articles
- EP ID EP17828
- DOI -
- Views 372
- Downloads 11
Construction of control chart based on six sigma initiatives for Cumulative – Sum
A control chart is a statistical device used for the study and control of repetitive process. W.A. Shewhart [6] of Bell Telephone Laboratories suggested control charts based on the 3 sigma limits. Now the companies in d...
An Effective Object Detection and Tracking by Background and Foreground Detection
the advance of technology makes video acquisition devices better and less costly, thereby increasing the number of applications that can effectively utilize digital video. Compared to still images, video sequences provi...
slugSurvey on Localizing Text in Scene Images
Localizing text method is used for extracting text regions from scene images with complex backgrounds and multiple text appearances. Natural scene images with text information are divided into two categories according t...
Locate Me: A Privacy Preserving Pseudonym Generation of Location Proof Updates Through Co-Located Devices
Today’s location-sensitive service relies on user’s mobile device to determine its location and send the location to the application. This approach allows the user to cheat by having his/her device transmit a fake locat...
Intelligent D2D Communication Using Internet of Things
Over the traditional home device in our daily life, the technology will get our life to next generation home automation. The proposed system work on sensor and mobile device to give an intelligent solution for needs, th...