A Consumer-Friendly Machine Learning Based Mechanism to Recognize the Quality of Commercially Available Fruits
Deepti C.1, Arjit Jindal2, Prudhvi Reddy3, Amrutha D.4
1Deepti C.*, PES University Electronic City Campus, Bengaluru, India.
2Arjit Jindal, PESIT Bangalore South Campus, Bengaluru, India.
3Prudhvi Reddy, PESIT Bangalore South Campus ,Bengaluru, India.
4Amrutha D, PESIT Bangalore South Campus, Bengaluru, India.
Manuscript received on January 24, 2020. | Revised Manuscript received on February 05, 2020. | Manuscript published on February 29, 2020. | PP: 3833-3839 | Volume-9 Issue-3, February 2020. | Retrieval Number: C6270029320/2020©BEIESP | DOI: 10.35940/ijeat.C6270.029320
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: Cultivators and sellers of many high-in-demand fruits traditionally preferred natural ripening after picking. Greed of hefty profits has motivated some of them to artificially hasten the ripening process at the cost of people’s health. Artificial ripening processes tend to degrade the entire quality of the fruit. The focus of this work is to describe a nondestructive method to detect artificial fruit ripening. To aid the detection, the proposed solution utilizes image processing and machine learning techniques to find the artificially ripened fruits. An input fruit image is selected as the test image. The next stage involves comparison of the features (histogram values) of the test image with the image of a naturally ripened one. A smartphone runs an android application to identify artificially ripened fruits. This work specifically concentrates on the commonly preferred Indian Mango and Indian Apple. The developed mechanism has an efficiency of 89-94% in correct detection.
Keywords: CNN based Approach, Fruit Ripening, Image Processing, Machine Learning.