Zeus Numerix Logo Home | Company | Careers | Contact Us
Icon case_studies
Case Studies
Home » Case Studies » Aerospace & Defense

Using Image Processing Techniques for Estimating Rocket Tip-Off

You may download a PDF copy of this article from here.

Our customer is a defense organization working towards delivering state of the art Multi Barrel Rocket Launch system to the Indian Army. Such systems suffer from tip-off, i.e. non alignment of the projectile with the launch tube at the time when the projectile has just left the tube. Since tip-off affects the fall of shot and CEP (Circular Error of Probability), its measurement from every launch trial is extremely desirable. Even though high speed cameras (1000 fps) are being used to document the launch, manual analysis on underlying images is error prone due to distortions from lens aberration, CCD/CMOS noise, lossy image compression algorithms etc.

Zeus Numerix developed a generic software framework involving computer vision and image processing concepts. The software comprised mainly of:

  • Camera Calibration (transformation from optical characteristics & orientation)
  • Lens Distortion (modeling of radially symmetric distortions)
  • Shooting Setup (selection of video frame composition)
  • Image Processing (tracking algorithm for translation & rotation of projectile)
  • Kalman Filter (analysis with dynamics equation)
The measured tip-off values from this software were in excellent agreement with numerically estimated values and its effect on fall of shot.

Left: Effects of various types of distortion on a image, Right: Composition guide for setting up the camera.

The customer was supplied with this software which automated the process of rocket launch measurement. Extracted data is presented in easy to analyze tables and graphical plots, thus saving the designers of the rocket system a lot of precious time.

Share Social
Quick Links