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

Washroot image analysis

Step 1 Roots are washed from the soil and all or most of the debris is removed, then roots are stained dark.
Step 2 Images are videotaped, using backlighting, by a robotic camera system (in figure below). Barcodes describing the data are recorded at the beginning of every new sample.
Step 2 Images are automatically digitized from the video tape using audio tone markers and saved to the disk.
Step 3 Images are processed by WR-RIPL one by one.
Step 4 Results for individual images are organized at the level of experiment performed, date of recording, and the sample number.

Based on the previous algorithm developed for VICOM image processor.
Threshold is found from the intensity histogram (c).
Thresholded image (d) is thinned to a point to obtain centerlines (e).
Centerlines are measured for length and average width, and classified as root or debris using specified length/width ratio: (f) white lines are debris, colored lines are roots, different colors are used for different width classes.

(a) Original image (b) Enhanced image (c) Intensity histogram

(d) Thresholded image (e) Centerlines (f) Final segmentation