Top Banner
Realtime Robotic Radiation Oncology Brian Murphy 4ECE
13
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Realtime Robotic Radiation Oncology Brian Murphy 4ECE.

Realtime Robotic Radiation Oncology

Brian Murphy 4ECE

Page 2: Realtime Robotic Radiation Oncology Brian Murphy 4ECE.

Objectives This project is based on a partially

completed previous project to develop a real time system to optically track a ball in the field of view of a camera.

Once the ball was identified, a robotic arm with a torch attached will point towards the ball.

The ball represents a tumour while the torch represents a beam of radiation.

Page 3: Realtime Robotic Radiation Oncology Brian Murphy 4ECE.

Objectives The system will be further developed to

operate in real-time to track a moving target.

The last stage will be to determine the location of the object in 3D space.

The image processing and algorithm development will be performed in Matlab

The real-time processing will be attempted using Labview.

Page 4: Realtime Robotic Radiation Oncology Brian Murphy 4ECE.

Achieved to Date Imported video stream into Image

acquisition toolbox in Matlab. Processed images using the Image

processing toolbox. Commissioned Labview for image

capture. Performed simple edge detection

in Labview.

Page 5: Realtime Robotic Radiation Oncology Brian Murphy 4ECE.

Problems encountered with Labview

Software has to be installed in a specific order.

Camera Driver Labview Image Acquisition Toolbox Driver Software Vision Development Toolbox IMAQ USB setup

Page 6: Realtime Robotic Radiation Oncology Brian Murphy 4ECE.

Image Capture in Labview Simple Image Capture from USB

Page 7: Realtime Robotic Radiation Oncology Brian Murphy 4ECE.

Resulting Image

Page 8: Realtime Robotic Radiation Oncology Brian Murphy 4ECE.

What we want to do Want to locate the ball the

previous image. Perform various different image

processing techiniques to optically track the ball i.e. edge detection, edge linking, circle fit etc.

Compare Labview results to Matlab results.

Page 9: Realtime Robotic Radiation Oncology Brian Murphy 4ECE.

Edge Detection in Labview Simple edge detection program in

Labview Reads in a collection of circles from

a .bmp file

Analyses the picture and then returns the image having performed edge detection on it.

Page 10: Realtime Robotic Radiation Oncology Brian Murphy 4ECE.
Page 11: Realtime Robotic Radiation Oncology Brian Murphy 4ECE.

Currently Working On Lego mindstorms NXT

Devolop mindstorms systems to track a ball and follow it if it moves.

Problem with servo motor.

Page 12: Realtime Robotic Radiation Oncology Brian Murphy 4ECE.

Remaining Work Import all Matlab algorithms into Labview. Create Labview program to control two servo

motors using NIDAQ interface card . Integrate image processing algorithms with

the servo motors to allow system to estimate location of the ball and move the servo motors to point beams directly at the ball.

Fine tune system to allow two servo motors and two cameras run in real time.

Page 13: Realtime Robotic Radiation Oncology Brian Murphy 4ECE.

Questions?