TY - GEN
T1 - Intelligent System to Classify Peanuts Varieties Using K-Nearest Neighbors (K-NN) and Support Vector Machine (SVM)
AU - Narendra, V. G.
AU - Govardhan Hegde, K.
PY - 2019/1/1
Y1 - 2019/1/1
N2 - In the world, India is the second biggest producer of peanuts or groundnuts, and it is also our country’s major oilseed crop. In India, the existing peanuts crop varieties are GAUG-1, Kuber, Amber, PG-1, BG-1, T-64, GAUG-10, BG-2, Chandra, Kadri-2, Chitra, Kadri-3, Prakash, T-28, Kaushal, etc. Presently, peanuts are having only 75–80% of India’s average market value. Because, the peanuts kernel quality assessment, as well as identifying varieties, are done manually by skilled labors, which leads costly. In this research, an affordable method is proposed to assess the peanuts kernel quality and identifying the different varieties quickly with undamaged, repeatability with low cost, and accurately with high distinguishing rate. Also, to meet the quality of peanuts kernel as per the international market standards and to increase the income of the former. The proposed system relies on computer vision and machine learning. The obtained overall accuracies were K-nearest neighbors (93.33%) and Support vector machine (93.82%). These percentages are discriminating peanuts variety as the best predictive model.
AB - In the world, India is the second biggest producer of peanuts or groundnuts, and it is also our country’s major oilseed crop. In India, the existing peanuts crop varieties are GAUG-1, Kuber, Amber, PG-1, BG-1, T-64, GAUG-10, BG-2, Chandra, Kadri-2, Chitra, Kadri-3, Prakash, T-28, Kaushal, etc. Presently, peanuts are having only 75–80% of India’s average market value. Because, the peanuts kernel quality assessment, as well as identifying varieties, are done manually by skilled labors, which leads costly. In this research, an affordable method is proposed to assess the peanuts kernel quality and identifying the different varieties quickly with undamaged, repeatability with low cost, and accurately with high distinguishing rate. Also, to meet the quality of peanuts kernel as per the international market standards and to increase the income of the former. The proposed system relies on computer vision and machine learning. The obtained overall accuracies were K-nearest neighbors (93.33%) and Support vector machine (93.82%). These percentages are discriminating peanuts variety as the best predictive model.
UR - http://www.scopus.com/inward/record.url?scp=85075235335&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85075235335&partnerID=8YFLogxK
U2 - 10.1007/978-981-15-0108-1_33
DO - 10.1007/978-981-15-0108-1_33
M3 - Conference contribution
AN - SCOPUS:85075235335
SN - 9789811501074
T3 - Communications in Computer and Information Science
SP - 359
EP - 368
BT - Advanced Informatics for Computing Research - 3rd International Conference, ICAICR 2019, Revised Selected Papers
A2 - Luhach, Ashish Kumar
A2 - Jat, Dharm Singh
A2 - Hawari, Kamarul Bin Ghazali
A2 - Gao, Xiao-Zhi
A2 - Lingras, Pawan
PB - Springer Paris
T2 - 3rd International Conference on Advanced Informatics for Computing Research, ICAICR 2019
Y2 - 15 June 2019 through 16 June 2019
ER -