Comparison of Grain Counting Methods

Next: Counting patterns with neural Up: Grain counting and pattern Previous: Grain counting and pattern Contents Index

Comparison of Grain Counting Methods

Feature Neural Network Differencing Quick Segmentation

Speed slow medium fast

No. of params 3 2 1

Works with color images yes no no

Ability to separate good medium poor

closely-spaced patterns

Requires selecting yes no no

representative pattern

Provides graphs of yes / no yes / yes yes /yes

signal distribution and

size distribution

Type of pattern Any Grains/spheres Grains

The neural network algorithm is a modified and greatly simplified version of a hierarchical neural network[6]. With appropriate threshold values, it is possible to count grains or any other pattern such as cells, faces, etc., in the presence of other potentially-interfering objects. To illustrate, we will count occurrences of the letter ``x'' in a simplified image.

The left panel below shows the original image, a combination of x's, y's and z's. The pattern recognition process subtracts pixels that are part of the pattern (in this case the `x's), while tye `y' and `z' are unaffected. This subtraction also allows any patterns obscured by the pattern to be identified on a subsequent pass.

$\begin{picture}( 100,100 )(0,10) \put(50, 10){ \epsfig{file = letters1.ps, widt... ...in }} \put(170, 10){ \epsfig{file = letters2.ps, width=1.3 in }} \end{picture}$
Pattern recognition in tnimage.

Next: Counting patterns with neural Up: Grain counting and pattern Previous: Grain counting and pattern Contents Index

root 2006-11-13

Feature	Neural Network	Differencing	Quick Segmentation
Speed	slow	medium	fast
No. of params	3	2	1
Works with color images	yes	no	no
Ability to separate	good	medium	poor
closely-spaced patterns
Requires selecting	yes	no	no
representative pattern
Provides graphs of	yes / no	yes / yes	yes /yes
signal distribution and
size distribution
Type of pattern	Any	Grains/spheres	Grains