Root Image Processing Lab
Plant and Soil Science Department
Michigan State University
Director Dr. Alvin Smucker


Computer Algorithms for Quantifying Video Recordings of Plant Root Images

Vera Bakic (CPS), Dr. Alvin Smucker (CSS), Dr. George Stockman (CPS)

Pattern Recognition and Image Processing Lab
Computer Science Department
Michigan State University

and

Root Image Processing Lab
Plant and Soil Science Department
Michigan State University

Minirhizotron image analysis

MR-RIPL 2.0---Ridge detection algorithm

(a) Original image (b) Enhanced image (c) Peaks and gradients
(d) Centerlines (e) Centerlines and edges (f) Final segmentation

MR-RIPL 3.0---``Mole'' filter algorithm

(a) Original image (b) Enhanced image (c) Histogram equalized
(d) Mole filter response (e) With edge information
(f) Thinned (g) Centerlines (h) Final segmentation

Classification methods and data sets used

Results of MR-RIPL

Manual MR-RIPL
classification Incorrect Error in %
v2.0 v3.0 v2.0 v3.0 v2.0 v3.0
All root 65698 68935 20397 10693 31.05 15.51
Tapes bg 238496 70920 4347 6544 1.82 9.23
Tape 1 root 9234 11585 2572 2998 27.85 25.88
corn bg 70614 20787 1494 1323 2.12 6.36
Tape 2 root 8235 9096 1833 862 22.26 9.48
alfalfa bg 34244 13153 846 1758 2.47 13.37
Tape 3 root 10390 9820 3399 1113 32.71 11.33
populus bg 31665 12890 276 2269 0.87 17.60
Tape 4 root 27325 26834 10629 4498 38.90 16.76
beans bg 35313 8234 901 790 2.55 9.59
Tape 5 root 10514 11600 1964 1222 18.68 10.53
wheat bg 66660 15856 830 404 1.25 2.55
All Tapes Tape 1 Tape 2 Tape 3 Tape 4 Tape 5
Length v2.0 16050 1078 987 3123 9728 1134
(pixels) v3.0 4149 1675 +896 +1156 3708 818
Error v2.0 24.43 11.67 11.98 30.06 35.60 10.78
in % v3.0 6.02 14.45 +9.85 +11.77 13.82 7.05

Discussion of MR-RIPL results

Sample images

(a) Original image (b) MR-RIPL 2.0 (c) MR-RIPL 3.0