Dates ripen state identification deep learning technique

Jawaid Shabbir; M. Usama; A. Ali; T. Hussain; M. Mazhar Khan

Author(s)

Jawaid Shabbir Computer Engineering Department, Sir Syed University Of Engineering and Technology Karachi,,Pakistan

M. Usama Computer Engineering Department, Sir Syed University Of Engineering and Technology Karachi,,Pakistan

A. Ali Computer Engineering Department, Sir Syed University Of Engineering and Technology Karachi,,Pakistan

T. Hussain Computer Engineering Department, Sir Syed University Of Engineering and Technology Karachi,,Pakistan

M. Mazhar Khan Computer Engineering Department, Sir Syed University Of Engineering and Technology Karachi,,Pakistan

Abstract

One of the primary factors is that conventional approaches are expensive and time-consuming. An automatic system for classifying the maturity of palm dates was created in this suggested system to identify the various phases of date fruit ripeness quickly and accurately. The three stages of our system's operation are the one-stage deep learning model (You Only Look Once Yolov4) algorithm used to identify palm dates in a video frame, the centroid tracking algorithm used to keep track of each vehicle within a defined area of interest, and palm date ripen state detection algorithm. While the centroid tracking technique can effectively follow any moving item, the deep learning model (You Only Look Once Yolov4) approach is particularly accurate at detecting objects. A test using a few traffic movies demonstrates that our suggested approach can detect and identify any ripen state detection of dates in various lighting and weather scenarios. The technology is straightforward to use and put into place.

Publication Details

Page(s) 24-29

DOI DOI not available

Published Journal: International Journal of Emerging Engineering and Technology, Volume: 1, Issue: 2, Year: 2022

Request the full text from PASTIC

Keywords

deep learning, YOLO, dates.