• Suhas Jadhav
• M. U. Inamdar

In this paper, computer vision-based algorithm for tomato grading is projected which workings in four steps. First is to apply preprocessing for uploaded image or images, such preprocessing step involves different algorithm methods like gray scale conversion, histogram calculation and binarization process. Discussed such all process deeply in below section. Second step is k-means segmentation to segment input image into 3 region i.e, Background, damaged and fresh area. Then, geometric, statistical and textural features from refined defected regions are extracted. Finally, for tomato grading, valued performance of SVM (Support Vector Machine) classifiers is done. Classification is computed in two approaches which in the first one, an input tomato is classified into two categories which of healthy grade and defected grade. In the second approch, the input tomato is classified into three sectors of first grade, second grade and damaged grade. In both grading steps, SVM classifier works as the finest one with detection rate of 96.8% and 98.1% for 2 categories (healthy grade and defected grade) & 3 quality categories (first grade, second grade and damaged grade), among 90 different golden delicious apple images, respectively.

SVM Classifier, K-means Clustering, Machine learning.

"Smart Fruit Grading System Using K-Means Clustering and SVM Classifier", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 6, page no.437-442, June-2019, Available :http://www.jetir.org/papers/JETIR1907059.pdf

"Smart Fruit Grading System Using K-Means Clustering and SVM Classifier", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 6, page no. pp437-442, June-2019, Available at : http://www.jetir.org/papers/JETIR1907059.pdf