TDA4VH-Q1: LDC Affine flipping

Dinesh bharathi

Part Number: TDA4VH-Q1

Hi Team,

I was trying to do the flipping on LDC but when I use the matrix [-4096 0 0; 0 4096 0; 0 0 0] It is not reflecting in the output. Is the affine limited to flipping? or am missing something here.
Output Image:

Regards,
Dinesh

over 1 year ago

0 Gang Hua over 1 year ago

TI__Guru 59010 points

Hi Dinesh,

The (h, v) coordinates start from 0 to image width and height.
Your matrix goes from (h, v) to (-h, v) which is out of the input image boundary.

I suppose you need to have "c = image height".

0 Dinesh bharathi over 1 year ago in reply to Gang Hua

Intellectual 751 points

Hi Gang,

In such case it is only applying translation in the horizontal direction not the expected flipping, the generic method for flipping is [-1 0; 0 1] Right?

0 Gang Hua over 1 year ago in reply to Dinesh bharathi

TI__Guru 59010 points

Dinesh bharathi said:
the generic method for flipping is [-1 0; 0 1] Right?

Yes, that is correct for [a,b; d, e].

But, you still need to use correct "c" and "f" values.

0 Dinesh bharathi over 1 year ago in reply to Gang Hua

Intellectual 751 points

I have used

[-4096 0 1920
0 4096 1280
0 0 0]

[4096 0 1920
0 4096 1280
0 0 0]

[4096 0 1280
0 4096 1920
0 0 0]

[-4096 0 1280
0 4096 1920
0 0 0]

All of them, Something happens but definitely not flipping.