Scribd
Upload
24 views1 page

Sample Poster Content

Uploaded by

Sriram Sri
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
24 views1 page

Sample Poster Content

Uploaded by

Sriram Sri
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 1

Expelling Undesired Targets From an Image Exploiting Surface

Synthesis and Dispersion Based Technique


Pratik Bhosale [15MCM0013] | Prof.Kumaresan P| SCORE
Motivation/ Introduction Results
Images plays very important role to keep the memory of the incident to which After following the algorithmic steps we are able to remove the unwanted areas
that image belongs. Many times the images gets corrupted or some unwanted from image and also can reproduce the image with good quality. The below figure
objects are added. The technique which removes the undesired portion from an shows the output result of the system when we successfully implemented it.
image is called Object removing and Region filling technique. There are two types
(i) Surface synthesis technique which is used to detect the characteristics of the The figure below shown is the login page. If we enter correct username and
image surface and to generate large image region.(ii) Dispersion based method to password then only we can access the system.
fill the little locales or crevices. The main aim of this technique is to produce the
image without disturbing its original quality and restore it.

SCOPE of the Project


There is a lot of scope for inpainting technique in future to reproduce the output
image as a HD image. Also the time required for inpainting and calculating the
pixel values can be reduced by using more algorithms. The other applications of
this technique is it can be effectively used in traffic surveillance cameras for
detecting the number plate and for security and photographic purpose this
technique can be used. First image is pre-processed get the information about
image such as size and type. Second, a low-resolution image is first built from the
original picture. Third, an in-painting algorithm is applied to fill-in the holes of the
low-resolution picture. Fourth, the quality of the inpainted regions is improved by
super-resolution algorithm.

Methodology
The proposed system have 8 main blocks.1.Region of Interest of an Input
Image:Undesired area or object from an input image is basically a Region of The figure below shown is the Masked Region of Interest image. The masked
Interest means the area where we wants to perform the inpainting operation. region is shown by green colour. Only that portion of the image will be removed.
2.Mapping Operation: The Mapping operation is done by using border detection
principle. The points are created at the border of the object and they are
connected sequentially. But the first point should connect to the last point then
only mapping operation will be done.3.Masking operation:Only the masked
portion will be removed from the image.

Fig 1: Block Diagram

The figure shown below is the inpainted output image.

4.Surface Synthesis Technique:In order to identify the image characteristics this


method is used. The mapped pixel values of the region will be saved and only the
masked region pixels will be removed from the image.5.Dispersion Based
Technique:The region pixel values which are removed will be filled in this
technique. In this technique K-nearest neighbor algorithm is used which will uses
the nearest pixel values to fill the removed patch or region.6.Down Sampling: It is
done by integrating the input image by x factor to set the output image pixels to
the average of the corresponding input pixel.7.Reproduced Image:After
performing all the operations new image is produced and it will be displayed on
the GUI.8.Image Restoration:The image which is produced will be stored at
desired location. The location is selected by the user where he wants to save or
restore the output image.

The below Algorithmic diagram represents sequential steps to remove the


unwanted area from image also called as Image Inpainting. Conclusion/ Summary
This project aimed at Super resolution based in-painting using Image Processing
Fig 2: Algorithmic diagram has satisfied the goal. The development and implementation of this system has
given us a great satisfaction. The in-painted images are sharp and without color
artifacts. The examples shown suggest a wide range of applications like
restoration of old photographs and damaged film, removal of superimposed text,
and removal of objects. The user needs only to provide the region to be in-
painted, the rest is automatically performed by the algorithm in a few
minutes.The results can either be adopted as a final restoration or be used to
provide an initial point for manual restoration, thereby reducing the total
restoration time by orders of magnitude.

Contact Details
[email protected]

Acknowledgments/ References
I am so much thankful of Prof.P.Kumaresan for guiding me and
encouraging me to do more research

You might also like