0

I want to convert a image of a yellow chess board on a wall to a black and white image in which all yellow(or any shades of yellow) portions gets converted to black and I may get a perfect chessboard image to use findchessboardcorners() function in opencv.Even though it works fine with gray scale image but I want to make findchessboardcorners() function work faster as it works with true black and white images.Please can anyone suggest a method to do so in opencv.

user3807950
  • 75
  • 1
  • 9

2 Answers2

2
Haris
  • 13,645
  • 12
  • 90
  • 121
  • 1
    Something like `inRange(hsv_image, Scalar(28, 10, 10), Scalar(44, 255, 255), mask);` is needed – Bull Sep 04 '14 at 04:06
  • Thanks Haris....that's exactly I want..but I am able to follw up till your second point only..that is after using inrange function...I get a black and white image with yellow portions of image in white and rest all is black but my yellow chessbord is having black boundries around each chesssqaure.. so that part of information is getting truncated....can you please suggest what to do after applying Inrange() function? – user3807950 Sep 09 '14 at 21:02
  • Can you post the image you working with.... – Haris Sep 10 '14 at 10:05
0

Try using thresholding after converting into greyscale and then feed it to the chessboardcorners function. That is not very process intensive.

ram
  • 124
  • 1
  • 1
  • 6