Page 46 - tech fest 2025 אשדוד חוברת תקצירים

46
Classification of Date Ripeness Stages Using Computer Vision
EEE-B-05
Noam Navon; noamnavonphima@gmail.com
Advisor: Dr. Tom Trigano
SCE - Shamoon College of Engineering, Ashdod
Identifying the ripeness stages of dates from images enables targeted hormonal treatments that maximize yield and quality. To achieve this, image processing and statistical learning methods are applied, with this project focusing primarily on computer vision. Visual features of the fruit—such as color distribution, size, and texture—were extracted from the images using image processing techniques. These features formed the basis for integration with biological measures and previous findings, serving both to support research on the role of abscisic acid in the date growth process, and to build a reliable classifier for distinguishing between ripeness categories corresponding to pre-, duo- , and post-hormonal treatment stages.
Keywords: agriculture, computer vision, image processing, statistical learning
Feature Extraction of Dates for Measurement
EEE-B-06
Elhanan Kadosh; elhanankadosh1@gmail.com Harel Shpunt; shpunt.harel@gmail.com
Advisor: Dr. Tom Trigano
SCE - Shamoon College of Engineering, Ashdod
The project aimed to identify the ripe stage of dates using images and measurable metrics taken from the fruit. It specifically focused on the statistical learning aspect of the algorithm. The first phase examined which specific measurable characteristics would contribute to identifying the fruit's ripeness level, in order to achieve an accurate assessment based on the input data. In the next phase, these findings were integrated with visual data, and an algorithm was developed that uses the collective information to determine the level of ripeness. The result was an automated and accurate system that allows for quick and reliable identification of ripeness stages at any given time and under various conditions.
Keywords: agriculture, data analysis, machine learning