A. Vadivel, Shamik Sural and

1
COLOR-TEXTURE FEATURE EXTRACTION USING SOFT
DECISION FROM THE HSV COLOR SPACE

• A. Vadivel, Shamik Sural and
• A. K. Majumdar
• INDIAN INSTITUTE OF TECHNOLOGY KHARAGPUR, INDIA.
• shamik_at_sit.iitkgp.ernet.in

2
ANALYSIS OF THE HSV COLOR SPACE

• Three dimensional representation of the HSV color
  space
• Central vertical axis represents intensity, I
• Hue, H, is an angle in the range 0,2p relative
  to the red axis.
• Saturation, S, is the depth or purity of color

3
VISUAL PERCEPTION IN THE HSV COLOR SPACE
(a)
(b)
Variation of color perception with (a) saturation
(Decreasing from 1 to 0 right to left) for a
fixed value of intensity and (b) intensity
(Decreasing from 255 to 0 right to left) for a
fixed value of saturation.
4
SATURATION THRESHOLDING (HARD DECISION)
Weight 1 if Sgt0.25 0 if Slt0.25

• Each pixel is treated as a gray color pixel or a
  true color pixel based on its saturation

weight
Saturation
0.25
0.0
5
PROBLEMS IN HARD DECISION
Weight 1 if Sgt0.25 0 if Slt0.25
weight
Saturation
0.25
0.0
Mixture of Gray and True color Near saturation
threshold
6
TRUE COLOR AND GRAY COLOR WEIGHTS IN SOFT
DECISION
True Color Weight WH (S,I)
Gray Color Weight wI (S,I) 1 wH (S,I)
7
COMBINING COLOR AND TEXTURE - COLTEX

• Check Pixel proximity
• Consider Diagonal, Horizontal and Vertical
  neighbors
• Features for each neighborhood captured in a two
  dimensional matrix

p(0.4,0.9,200) q(1.6p, 0.0, 180)
Each cell represents a combination of current
pixel color (true or gray) and neighboring pixel
color (true or gray)
8
COMBINING COLOR AND TEXTURE - COLTEX
Matrix cells affected by neighbors p and q are
QH and QI determine quantization levels Quanta
of update for the cells are
9
PERFORMANCE MEASURE
Recall
Precision

• Relevant Set is not known for Large Uncontrolled
  Data Sets

Perceived Precision
10
NUMBER OF COLOR AND TEXTURE COMPONENTS
Recall and Precision on a database of 2015
images.
11
EXPERIMENTAL RESULTS ON 28,000 IMAGES
12
CONCLUSIONS AND FUTURE WORK

• Effective way to capture pixel color information
  using the HSV color space
• Relative importance of Hue and Intensity
  determined based on Saturation
• True color weight to capture color information
• Gray color weight to capture texture information
• COLTEX combines color and texture
• Implement other color-texture methods
• Comparison on a larger data set

13
Thank You