Image Processing in its general form pertains to the alteration and analysis of pictorial information. The objective of image processing is to visually enhance or statistically evaluate some aspect of an Image not readily apparent in its original form. This processing is used for convenience in order to reduce the complexity faced during the operations performed on an image. Edge detection is one such branch of image processing used to detect the edges of the objects in a picture by calculation the difference in brightness of that edge pixel with its surrounding pixels using gradient method. In this project, Sobel operator is used as a filter for detection of edges of projection of a door without further increasing the already complex process of image processing. This is done using MATLAB, Sobel filter and FPGA.

Project Proposal

1. High-level project introduction and performance expectation

Proposal 
In the presence environment every one are busy with their professional life. If they want to take care of elders they are depending on caretaker etc. If elder want to be an  independent, our proposed system gives solution for this challange. The proposed system will help the elder people to cross the door area easly without depending on others. It includes FPGA,camera.The camera continously capture the image and identify the door by using sobel edge detector. This dectector give the appropriate result for identifying the exact door or any other object by using image processing technique. . Edge detection is one such branch of image processing used to detect the edges of the objects in a picture by calculation the difference in brightness of that edge pixel with its surrounding pixels using gradient method. . In this project, Sobel operator is used as a filter for detection of edges of projection of a door without further increasing the already complex process of image processing. This is done using MATLAB, Sobel filter and FPGA.

 

2. Block Diagram

We are considering an image of size 564 * 588 pixel which will be the input image. The image is given to MATLAB to generate pixel data. This data is then passed as an input to the FPGA where Sobel operator algorithm is performed. Then the edge detected image is given as an output.
The operator consists of a pair of 3×3 convolution kernels as shown in Figure 1. One kernel is simply the other rotated by 90°.
 

These kernels are designed to respond maximally to edges running at 45° to the pixel grid, one kernel for each of the two perpendicular orientations. The kernels can be applied separately to the input image, to produce separate measurements of the gradient component in each orientation (call these Gx and Gy). These can then be combined together to find the absolute magnitude of the gradient at each point and the orientation of that gradient. The gradient magnitude is given by:
                                          
Typically, an approximate magnitude is computed using:
                                            
which is much faster to compute.

  KERNEL
In image processing, a kernel, convolution matrix, or mask is a small matrix. It is used for blurring, sharpening, embossing, edge detection, and more. This is accomplished by doing a convolution between a kernel and an image. The origin is the position of the kernel which is above (conceptually) the current output pixel. This could be outside of the actual kernel, though usually it corresponds to one of the kernel elements. For a symmetric kernel, the origin is usually the center element.
LINE BUFFER
These small memories used for storing one line of image will be called line-buffers.They can be built using BRAMs or distributed RAMs. So for a 512x512 image, one line-buffer will be 512 Bytes size and 3 line buffers require only 1536Bytes. In fact the size of line buffer depends only the width of the image. 
 

3. Expected sustainability results, projected resource savings

4. Design Introduction

Design introduction

In this project Sobel Operator is used as a filter on the image to detect the edges of the projection of the door in that image. The Sobel operator algorithm written in Verilog on Fpga executes the edge detection and identifies the projection of the door. In this kernels are written accordingly and image processing operation happens. In this paper the information on image processing, edge detection, sobel operator and the methods and tools used to achieve the desired result are presented in detail.   

5. Functional description and implementation

functional discription
Capturing the image converting from image into hex/binary file and applying sobel edge detector algorithm to find out exact edges of the door and determine the image of the door to avoid the obstacles.
 

6. Performance metrics, performance to expectation

7. Sustainability results, resource savings achieved

8. Conclusion

Conclusion
Through this project we have successfully implemented Edge Detection operation in Image Processing using Sobel operator. The input image having a projection of the door is edge detected and given as an output. Using FPGA than any other algorithm has more advantaging factors including parallelism, high storage than any other algorithm as there are also some difficulties in complex mathematical calculations, a lot of research needs to be done in this field to provide the optimal design. 
This can be extended to identify the type of edge detected door for a robot to be done autonomously. This opens further exploring the idea into new research areas.