Automated detection of nitrogen status on plants: Performance of image processing techniques

Bibliographic Details
Format: Restricted Document
_version_ 1860799526150864896
building INTELEK Repository
collection Online Access
collectionurl https://intelek.unisza.edu.my/intelek/pages/search.php?search=!collection407072
date 2019-06-12 01:42:42
eventvenue Seoul, South Korea
format Restricted Document
id 6343
institution UniSZA
originalfilename 1196-01-FH03-FRIT-19-25570.pdf
person Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML
like Gecko) Chrome/74.0.3729.169 Safari/537.36
recordtype oai_dc
resourceurl https://intelek.unisza.edu.my/intelek/pages/view.php?ref=6343
spelling 6343 https://intelek.unisza.edu.my/intelek/pages/view.php?ref=6343 https://intelek.unisza.edu.my/intelek/pages/search.php?search=!collection407072 Restricted Document Conference Conference Paper application/pdf 4 1.6 Adobe Acrobat Pro DC 20 Paper Capture Plug-in Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML like Gecko) Chrome/74.0.3729.169 Safari/537.36 2019-06-12 01:42:42 1196-01-FH03-FRIT-19-25570.pdf UniSZA Private Access Automated detection of nitrogen status on plants: Performance of image processing techniques The significant role of nitrogen element in plants growth resulting in increased usage of nitrogen fertilizer in the agriculture field. With the aim to avoid improper use of nitrogen fertilization on plants and to assist local farmers in improving plants monitoring, this paper presents an economical and non-destructive method in determining nitrogen status of Napier grass using digital image processing approach. Three authentic techniques of image segmentation Otsu, K-means clustering, and watershed transformation were applied and compared to recognize the most accurate method for segmenting leaf pixel from its background. Otsu was discovered as the most efficient technique with less time-processing. Out of 36 features extracted from the segmented image, kurtosis, skewness and standard deviation of the blue color image were the most related features in classifying nitrogen status of the images. Classifiers like KNN, decision tree, and linear discriminant were used to classify the leaves image and nitrogen status accordingly. The accuracy of 100% was recorded in classifying the leaves image using decision tree and KNN classifier. 16th IEEE Student Conference on Research and Development Seoul, South Korea
spellingShingle Automated detection of nitrogen status on plants: Performance of image processing techniques
summary The significant role of nitrogen element in plants growth resulting in increased usage of nitrogen fertilizer in the agriculture field. With the aim to avoid improper use of nitrogen fertilization on plants and to assist local farmers in improving plants monitoring, this paper presents an economical and non-destructive method in determining nitrogen status of Napier grass using digital image processing approach. Three authentic techniques of image segmentation Otsu, K-means clustering, and watershed transformation were applied and compared to recognize the most accurate method for segmenting leaf pixel from its background. Otsu was discovered as the most efficient technique with less time-processing. Out of 36 features extracted from the segmented image, kurtosis, skewness and standard deviation of the blue color image were the most related features in classifying nitrogen status of the images. Classifiers like KNN, decision tree, and linear discriminant were used to classify the leaves image and nitrogen status accordingly. The accuracy of 100% was recorded in classifying the leaves image using decision tree and KNN classifier.
title Automated detection of nitrogen status on plants: Performance of image processing techniques
title_full Automated detection of nitrogen status on plants: Performance of image processing techniques
title_fullStr Automated detection of nitrogen status on plants: Performance of image processing techniques
title_full_unstemmed Automated detection of nitrogen status on plants: Performance of image processing techniques
title_short Automated detection of nitrogen status on plants: Performance of image processing techniques
title_sort automated detection of nitrogen status on plants: performance of image processing techniques