Research Article
DINOSARC: Color Features Based on Selective Aggregation of Chromatic Image Components for Wireless Capsule Endoscopy
Algorithm 2
Distances on SARC (DINOSARC) salient point detection.
| | Input: Images Ic (M × N), c = a, b; ISARC (M × N) | | | Output: List of salient points I[] | | (1) | //Initialize | | (2) | i ⟵ 0; | | (3) | d[] ⟵ null; | | (4) | I[] ⟵ null; | | (5) | //Sample ISARC and evaluate saliency | | (6) | For each ISARC ≠ 0, do | | (7) | templarge[] ⟵ s × s neighborhood centered at ; | | (8) | tempsmall[] ⟵ s/2 × s/2 neighborhood centered at ; | | (9) | | | (10) | | | (11) | | | (12) | | | (13) | i ⟵ i + 1; | | (14) | | | (15) | | | (16) | End For | | (17) | //Filter salient points upon their proximity | | (18) | For i = 1 to length (d[]) do | | (19) | If d[i] ≤ average (d[]) then | | (20) | remove(d[i]); | | (21) | remove(I[i]); | | (22) | End If | | (23) | End For |
|