Update:
Final Progress Report is downloaded in the File (optional) below

View uploaded report

Digital Imaging Technique to Detect and Rate Iron Deficiency chlorosis (IDC) in Soybeans
Oveis Hassanijalilian, Hans Kandel, Sreekala G. Bajwa, John Nowatzki, Saravanan Sivarajan, Ted Helms
Based on USDA statistics, 5.75 million acres of soybeans were planted in the state of North Dakota in 2015. Soybean yield in the state of North Dakota was almost 185.9 million bushels. One of the factors that reduces soybean yield is lack of useable iron which causes Iron Deficiency Chlorosis (IDC). In 2013 at The North Central U.S., total land area where soybean was grown was approximately 4.7 million acres, where IDC caused 375,000 ton loss in soybean grain production at a value of $120 million per year.
The most effective step to manage this problem is to select an IDC tolerant variety suitable to a specific area as IDC occurrence varies widely under different environment conditions. Tolerance of different varieties to IDC can be determined through IDC visual rating by experts in variety research trials. Some scientists suggest that visual ratings at an early stage could help farmers identify its severity and apply iron chelates to prevent yield loss. But visual rating is too subjective, and requires experts. Leaf color can be used to identify chlorosis in soybean leaves, so digital imaging technique can detect its presence when the crop is growing. Digital imaging technique is used in this study for rating IDC in soybeans with a camera which does not require expertise.
The digital image captured has different bands of Red, Green and Blue (RGB). Research studies at the University of Arkansas showed that the amount of red and blue color scheme indicates how green an image or a plant looks. Therefore, they suggested a vegetation index called Dark Green Color Index (DGCI) based on the color scheme or values of Hue, Saturation and Intensity (HSI). It has been used on corn to detect the nitrogen deficiency.
The greener soybean plots are, the higher the value of DGCI will get. The average value of DGCI for the whole plot was used as a base value to rate IDC. Then by segmenting the image and detecting the low DGCI values for chlorosis, the rating was increased. IDC scores are from 1 to 5 with 0.5 point increments which 1 is no chlorosis, and 5 is severe chlorosis and dead tissues.
Chlorophyll meter values for individual leaflets were also measured. Digital images also were captured from the same leaflets with two different smartphones to determine the ability of the technique to estimate the chlorophyll amount. The DGCI values determined by digital image processing technique was significantly correlated (r2 = 0.89) to Chlorophyll meter reading of the same leaflets. Moreover, the correlation between DGCI values of two different smartphones was significantly high (r2 = 0.98) which means DGCI can be accountable for differences of different cameras.
In summary, these days a lot of people carrying smartphones. They can be used to rate IDC in soybean using DGCI. We are planning to use DGCI accompanied by Unmanned Aerial Vehicle (UAV) to delineate IDC and to detect the growing pattern of IDC in the field. As a result, susceptible regions of the fields can be detected for future management.