Mobile App for Grading of Peanuts using Fuzzy Inference System
Asha Gowda Karegowda1, Pushpalatha.K.R2, Pramod.P.3
1Asha Gowda Karegowda, Associate Professor, Department of MCA, Siddaganga Institute of Technology, Tumkur, Karnataka, India.
2Pushpalatha.K.R, Assistant Professor,Department of MCA,Sri Siddhartha Institute of Technology, Sri Siddhartha Academy of Higher Education (SAHE), Tumkur, Karnataka, India.
3Pramod. P, Student, Department of MCA,Siddaganga Institute of Technology, Tumkur Karnataka, India.
Manuscript received on October 01, 2019. | Revised Manuscript received on October 15, 2019. | Manuscript published on October 30, 2019. | PP: 1262-1269 | Volume-9 Issue-1, October 2019. | Retrieval Number: A9629109119/2019©BEIESP | DOI: 10.35940/ijeat.A9629.109119
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The peanut a globally adopted vegetable protein with appreciable nutritive contents. The grading is one of the significant way to ensure that quality of the peanuts regulate the appropriate price in market. The proposed work is carried out in two phases. As part of first phase segmentation is done using canny edge detection followed by extraction of four Shape Features (SF): Area (Ar), Major Axis (MajAxis), Minor Axis (MinAxis), Perimeter (PeriMt). As part of second phase, Fuzzy Inference System (FIS) has been adopted for grading of four categories of peanuts. .The interval scaled shape features obtained in phase one are converted to Linguistic Terms (LingTerms). The fuzzy data base rules are attained using Association Rules with the constraints that the consequent can have only class label. The proposed work is developed as mobile app which takes peanut image as input and processes the image using FIS and provides the output in terms of pie chart. The pie chart provides the percentage of different qualities of peanuts (low/medium/good/fine) in the input image. The proposed work resulted in an accuracy of 95% when compared with the ground truth. The work can be applied in automation process for grading of food grains using mobile apps, which helps the lay customer to know about the quality of peanuts. The work can also be used for auto separation of peanuts for packing based on grade of the peanuts.
Keywords: Fuzzy Inference System, Mobile Application, Peanut grading, Shape Features.