The goal is to combine these methods and assign a distance of the object.
Elements of the problem are:
Broadly, the methodology is as follows:
Elements of the problem are:
- Detect an object on the image
- Measure the object dimension in pixel count
- Create a function to map distance to arbitrary object depending on pixel count occupied by object
- Calibrate the camera (find the mapping)
- Creating depth images to easily map distance to arbitrary object
Broadly, the methodology is as follows:
- Object Recognition using OpenCV
- Stereo Vision using OpenCV
- Set up a simple scene using one camera
- Start with a Uniform Background
- Known object with known dimensions
- Calculate distance of object using a single camera
- Using two cameras implement:
- Calibrate cameras for Stereo Vision
- Take Stereo Images
- Implement object recognition using two cameras
- Measure object size, relate actual size and measured size to get distance using two cameras
- Object Location Algorithm