Distance Estimation using stereo Images

Final project by

Ori Zakin & Ohad Eliyahoo

ohade@cs.bgu.ac.il zakino@gmail.com


Introduction

Range estimation is required for many applications such as: military, robotics, safety equipment, etc.

 

Current range estimation techniques require use of an active device such as a laser or radar. The main drawbacks of an active approach are:

• Expensive

• Use in military scenarios can compromise the measurers position.

 • Requires dedicated hardware For many applications a passive approach would be much better suited, one idea for such an approach is to use stereo photos.

 

 

Approach and Method

Our approach uses stereo photos in order to estimate the distance.

We divide the algorithm into several phases:

1.System configuration:

The system requires the user to enter the cameras specifications this parameters are critical for more accurate results.

 

2.Supply two different images of same object:

With the camera specified above take to picture one from the left of the object and the second from the right. be sure that you measured correctly the distance between the cameras.

 

3.Select object for distance estimation:

Now you have to mark by rectangle the object to estimate distance to.

 

4.Identify selected object in second image using image processing:

With cross-correlation function the system will find your object in the second image

 

5.Perform calculations and return result:

The system uses all the parameters above and with trigonometrically and geometrical formula calculate the distance.

 

 

Results

Equipment:

1. Canon A-95 digital camera.

a. Focal length – 7.8-23.4 (all experiments were with 7.8)

b. CCD width (Length of frame) – 7.1.

2. Measuring tape 3m

3. 1 Labrador male 3 year old – named Sub (measurer).

The experiments were conducted in the following fashion:

a. We placed an object at a known distance from the measuring point.

b. We took 2 photographs from the left and right of the object with different distances between the cameras locations.

c. We scaled down the images (to 25%) after an unsuccessful attempt on a full sized image set resulted in crashing the computer (several times).

d. We then ran our implementation on the images.

 e. The following table shows the results returned from our program compared to real world distance.

 
Distance between cameras Real distance Distance Estimated Delta
80 180 177.972 2.028
110 900 721.765 178.235
116 290 271.645 18.359
220 365 355.944 9.056
270 800 797.55 2.45
270 669 667.394 1.606

 

Conclusions

1. Supplying accurate data regarding the cameras parameters is crucial for the algorithms success.

2. Our project dealt only with horizontal distance between the cameras and the object, and mainly with the first case described above in which the object is located between the lenses.

3. As seen in the above table, increasing the distance between the cameras has a significant impact on the accuracy of the estimation.

4. We conducted many experiments, and the results shown above are a representative subset.

5. There is more research to be done in this field.

 

Additional Information

References

1. DISTANCE ESTIMATION ALGORITHM FOR STEREO PAIR IMAGES by EDWIN TJANDRANEGARA and YUNG-HSIANG LU

2.http://tangentsoft.net/fcalc/help/AoV.htm

3.http://www.canon.co.jp/Imaging/enjoydslr/p_2_012.html

4.http://www.halfbakery.com/idea/Lens_20width_2flength_20ratings

5.http://www.sweeting.org/mark/lenses/canon.php

6. http://hyperphysics.phy-astr.gsu.edu/hbase/geoopt/image.html

7. http://www.steves-digicams.com/2004_reviews/a95.html

8. Wikipedia

9. Google