An Android & Web based Application to Estimate Nitrogen Concentration in Rice Crop of Western Odisha
Hey Readers..
Just sharing my abstract part of Post Graduation Major Project. As the title suggests its a cross platform Web cum Android Application in the Agriculture Genre.
Publish in IEE.org
Just sharing my abstract part of Post Graduation Major Project. As the title suggests its a cross platform Web cum Android Application in the Agriculture Genre.
Publish in IEE.org
https://ieeexplore.ieee.org/document/9555875
Abstract
Summary
The developed web cum android app can automate nitrogen status estimation in rice leaf with high accuracy. However, we have achieved an error of 13.75%. This methodology can be put up to develop a farmer friendly app with a well behaved monitoring system. This application can also add an extra hand to collect real time dataset about rice production and fertilizer application. This app can be made more reliable by considering more features while N status estimation. This app can be made available efficiently to both industrial sectors as well as to a normal farmer if a well-design & discipline is followed for its development.
Abstract
In
Rice crop (Oryza sativa) the Nitrogen & Chlorophyll contents can be
estimated by Leaf Color Chart (LCC) for proper prescription about the
Fertilizer. Comparing the visual attributes with the LCC is not same always.
The accuracy is affected by human color perception. This paper proposes
development of a Web/Android based application called “N4riceleaf”. This
application captures a 2D color image of rice leaf & determines its average
color in HEX color code. Then it makes an estimation of N concentration in
accordance to the generated HEX color code of the image & the five shade
LCC achieving high accuracy. This
application runs on all devices irrespective of device & platform i.e.
Smartphone, tablet, Laptop etc. which meets a minimum requirement.
Summary
The developed web cum android app can automate nitrogen status estimation in rice leaf with high accuracy. However, we have achieved an error of 13.75%. This methodology can be put up to develop a farmer friendly app with a well behaved monitoring system. This application can also add an extra hand to collect real time dataset about rice production and fertilizer application. This app can be made more reliable by considering more features while N status estimation. This app can be made available efficiently to both industrial sectors as well as to a normal farmer if a well-design & discipline is followed for its development.
Thanks
Comments
Post a Comment