I am trying to do image detection in C ++. I have two images:
Image Scene: 1024x786 Man: 36x49
And I need to identify this person from the scene. I tried to use correlation, but the image is too noisy and therefore does not give correct / accurate results.
I thought / studied the methods that would best solve this problem, and this seems the most logical:
- Gaussian filters
- Convolution
- Fft
Basically, I would like to move the noise around the images, so I can use Correlation to find a person more efficiently.
I understand that FFT will be difficult to implement and / or may be slow, especially with the size of the image used.
Can anyone suggest any pointers to a solution to this? What will be the best technique / algorithm?