Classification & Grading of Tomatoes using Image Processing Techniques
Santi Kumari Behera1, Abhijeet Mahapatra2, Amiya Kumar Rath3, Prabira Kumar Sethy4
1Santi Kumari Behera, Assistant Professor, Department of Computer Science & Engineering, V. S. S. University of Technology, Burla, Sambalpur, India.
2Abhijeet Mahapatra, M.Tech. Student, Department of Computer Science & Engineering, V. S. S. University of Technology, Burla, Sambalpur, India.
3Prof. Amiya Kumar Rath, Professor & Head, Department of Computer Science & Engineering, V. S. S. University of Technology, Burla, Sambalpur, India.
4Prabira Kumar Sethy, Assistant Professor, Department of Electronics, Sambalpur University, Burla, Sambalpur, India.
Manuscript received on 08 April 2019 | Revised Manuscript received on 15 April 2019 | Manuscript Published on 26 April 2019 | PP: 545-550 | Volume-8 Issue-6S April 2019 | Retrieval Number: F61110486S19/19©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© 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: This examination manages the programmed arrangement and reviewing of tomatoes utilizing picture-preparing systems. Mostly tomatoes are of three assortments for example cherry, classic & cylindrical variety and of all assortment have little, medium & big size. Tomato order and evaluating is exceptionally troublesome precisely and in quick way because of their huge contrast in highlight, for example, size, shape and shading because of variable states of nature condition and manual components. Programmed characterization and evaluating of tomato dependent on picture preparing methods is the best arrangement as manual forecast is absence of objectivity, exactness and has lower proficiency. Here, we utilize the picture handling parameters, for example, major axis, minor axis, bounding box, perimeter & diameter for grouping and evaluating reason for tomato and furthermore confirmed with the ground truth measure by Vernier caliper. This examination accomplished coefficient of correlation (R2 ) 0.98 for length and 0.97 for width. Again, it effectively characterizes the variety of 96.67% and grade into three classes as indicated by size is 100%.
Keywords: Classification, Grading, Tomato, and Image Processing.
Scope of the Article: Classification