Automatic Fruits Identification and Disease Analysis using Machine Learning Techniques
Santi Kumari Behera1, Amiya Kumar Rath2, Prabira Kumar Sethyh3
1Santi Kumari Behera, Assistant Professor, Department of Computer Science and Engineering, Veer Surendra Sai University of Technology, Burla, Odisha.
2Prof. Amiya Kumar Rath, Professor, Department of Computer Science and Engineering, Veer Surendra Sai University of Technology, Burla, Odisha.
3Prabira Kumar Sethy, Assistant Professor, Department of Computer Science and Engineering, Veer Surendra Sai University of Technology, Burla, Odisha.
Manuscript received on 10 April 2019 | Revised Manuscript received on 17 April 2019 | Manuscript Published on 24 May 2019 | PP: 103-107 | Volume-8 Issue-6S3 April 2019 | Retrieval Number: F22190486S219/19©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: Automation plays an important role towards maintaining a standard with proper deal of money for Horti-Agricultural products. The automatic technique is very useful in large farm with various fruits for sorting and packaging purposes. Again, it is essential to categorize the fruits with evaluating its quality as per consumer demand. This paper proposes a methodology for fruits identification and disease analysis using machine-learning techniques. The identification and quality evaluation of fruits is based on different features i.e. color, texture, shape and appearance of defects/ Diseases. Image pre-processing is used for increasing the details of image. Image segmentation separate an image in to different parts and from the segmented image, Gray-Level Co-Occurrence Matrix (GLCM) is used to access needful features. The Multi class Support Vector Machine (MSVM) classifier is used to identify the type of fruits, detect the defects/disease appear on the fruit and classify the type of disease. The automated system successfully classifies five types of fruits i.e. Apple, Mango, Orange, Tomato and Pomegranate with accuracy of 94.02%. The second phase of methodology identified two most appeared common diseases i.e. Anthracnose and Fruit Rot with overall accuracy of 92.17%. Finally, to measure the severity of diseases Fuzzy Logic technique has been used and graded into three categories as per percentage of infection.
Keywords: Fruit Identification, Fruit Grading, GLCM, K-Mean, Multi Class SVM, Fuzzy Logic.
Scope of the Article: Fuzzy Logic