| dc.description.abstract | The study was conducted to assess the status and requirements of nitrogen at different growth
phases of wheat and maize through digital image analysis. The major objectives were to
study the effect of nitrogen on morpho-physiological traits; use digital image and software
for assessment morpho-physiological traits at different growth stages and establish a
relationship between morpho-physiological traits with image parameters. To fulfill the
objectives four experiments were conducted in two successive years (2016-17 and 2017-18)
at BARI, Gazipur, Bangladesh with wheat and maize crops. Wheat experiment comprised of
four wheat varieties (BARI Gom 26, BARI Gom 28, BARI Gom 29 and BARI Gom 30) and
four nitrogen levels (0.0 kg N ha-1, 50 kg N ha-1, 100 kg N ha-1, and 150 kg N ha-1) during 2016
17 and three wheat varieties (BARI Gom 26, BARI Gom 28 and BARI Gom 30) and five
nitrogen levels (0.0 kg N ha-1, 60 kg N ha-1, 100 kg N ha-1, 120 kg N ha-1 and 140 kg N ha-1)
during 2017-18. While, the maize experiment comprised of three maize varieties (hybrid 981,
BARI Hybrid Maize 9 and hybrid P3396) and three nitrogen levels (100 kg N ha-1, 200 kg N ha
1and 300 kg N ha-1) in both the years. All the experiments were laid out in split plot design
with three replications. Data on plant height (cm), Leaf Area Index (LAI), Canopy cover
(CC), Soil-Plant Analyses Development (SPAD) reading of leaf, Leaf nitrate content (LNC),
Leaf chlorophyll content (LCC), Normalized Difference Vegetation Index (NDVI), Total
dry matter (TDM), different yield contributing and yield traits of wheat and maize. CC is an
important phenotypic trait which indicates overall plant growth and helpful to predict
advanced traits like biomass and yield. Digital image analysis was used as an alternative
method to quantify the greenness of foliage as indirect measurements of crop N status and
phenotyping. The image was taken from 50 cm above the plant canopy by using modified
selfie stick. Two software’s (WCC for wheat and MCC for maize) were developed for digital
phenotyping in JAVA and Interface design in JFrame. A five-step algorithm was developed
to measure the CC of individual plots by RGB (red, green and blue). The output gave 27
digital traits value after analyzing an image including CC and NDVI. The values of canopy
cover were closely correlated with the NDVI and the ratio vegetation index. Among the
interaction treatments of wheat, maximum grain yield (3.97 t ha-1) was recorded by the
interaction of 150 kg N ha-1 and BARI Gom 30 which was statistically similar to 100 kg N
ha-1 and BARI Gom 29, 100 kg N ha-1 and BARI Gom 28 during 2016-17 where the
maximum grain yield (3.94 t ha-1) was found with treatment combination 120 kg N ha-1and
BARI Gom 26 which was statistically similar to 120 kg N ha-1 and BARI Gom 30, 120 kg
N ha-1 and BARI Gom 28 during 2017-18. Among the interaction effect of maize variety
and nitrogen level, highest grain yield (14.45 t ha-1) was recorded by the interaction of hybrid
P3396 and 300 kg N ha-1 which was statistically identical with BHM 9 and 300 kg N ha-1).
The results also indicated that, application of nitrogen fertilizer has positive effects on
physiological parameters of wheat and hybrid maize. A strong relationship among the traits
NDVI, SPAD, CC and LNC was observed across the wheat and maize varieties. The results
also suggest that in case of NDVI, SPAD and CC saturation, NDVI and CC could accurately
predict nitrogen requirements for wheat and maize. The results of this study partially
supported the hypothesis that digital image traits measured by image processing software’s
(WCC and MCC) can identify and differentiate variations in nitrogen use in wheat and maize
crops. The outcomes of study have created a new era of using digital image analysis in the
determination of N-requirement in wheat and maize in near future. | en_US |
