The accuracy of the sensor is the closeness of a measured value to a known value i.e., the maximum difference between the actual value (measured by a primary or good secondary standard) and the indicated value at the output of the sensor.

It can be expressed either as a percentage of full scale or in absolute terms.

The accuracy of the Intel RealSense R200 Camera is determined by the following method.

1. The 480 laser scan points of the object and the environment pointed by the camera are obtained.
2. Consider the object/obstacle is placed static in the center view of the camera.
3. The set of points (20/30 points) corresponding to the object location is obtained.
4. The average value of these points is determined and considered as the observed distance of the object from the camera.
5. The average of a set of points is obtained because a single laser point value is not reliable since the laser point that corresponds to the center of the object is unknown.
6. The actual distance of the object from the camera is manually measured.
7. The actual distance and the observed distance are compared and the difference in the values is determined.
8. The less the error value is, the more accurate is the camera. The more the error value is, the less accurate is the camera.
9. The same steps are followed for a few more test cases and accuracy is observed.

LIGHTING CONDITION = 1

RGB Images

1. Plain Wall

   

2. Chair placed before a plain wall

   

3. Bag placed before a plain wall

   

4. Bucket placed before a plain wall

   

5. Netted mesh placed before a plain wall

   

LIGHTING CONDITION = 2

RGB Images

1. Plain wall

   

2. Chair placed before a plain wall

   

3. Bag placed before a plain wall

   

4. Bucket placed before a plain wall