Related Experiment Video
Updated: Jun 30, 2026
11:02
Fruit Volatile Analysis Using an Electronic Nose
Published on: March 30, 2012
21.5K
TinyML-Sensor for Shelf Life Estimation of Fresh Date Fruits.
Ramasamy Srinivasagan1, Maged Mohammed2,3, Ali Alzahrani1
1Department of Computer Engineering, College of Computer Sciences and Information Technology, King Faisal University, Al Hofuf 36362, Saudi Arabia.
Sensors (Basel, Switzerland)
|August 26, 2023
Summary
This study uses machine learning and low-cost sensors to predict fresh date shelf life. Modified atmosphere packaging and vacuum sealing at 5°C significantly extended shelf life, offering a cost-effective solution for producers.
Area of Science:
- Agricultural Science
- Food Science
- Data Science
Background:
- Fresh dates have a limited shelf life, leading to spoilage and economic losses.
- Accurate shelf life estimation is crucial for producers, suppliers, and consumers.
- Modified atmosphere packaging (MAP) can extend fresh date shelf life by slowing ripening.
Purpose of the Study:
- To apply fast, cost-effective, non-destructive machine learning (ML) techniques for predicting fresh date shelf life.
- To optimize storage conditions for maximizing the shelf life of fresh dates.
- To enable timely consumption and preserve the nutritional value of dates.
Main Methods:
- Utilized Visible-Near-Infrared (VisNIR) spectral sensors to capture internal physicochemical attributes.
- Employed TinyML-compatible regression models to predict fruit development stages and estimate shelf life.
- Stored dates under vacuum and MAP (20% CO2, N balance) at 5°C and 24°C, monitoring attributes weekly.
Main Results:
- Vacuum packaging at 5°C extended shelf life to 53 days; MAP at 5°C extended it to 44 days.
- At 24°C, vacuum packaging yielded 44 days and MAP yielded 23 days of shelf life.
- The transition from Khalal to Rutab stage marked the beginning of shelf life decrease.
Conclusions:
- Machine learning and VisNIR sensors offer a non-destructive method for real-time shelf life estimation of fresh dates.
- Low-temperature storage (5°C) significantly enhances the effectiveness of both vacuum and MAP for extending date shelf life.
- Edge Impulse facilitates the deployment of TinyML models for practical, low-cost shelf life prediction applications.
Related Concept Videos
01:27Principles of Food Preservation
Food spoilage results from microbial growth, enzymatic activity, and environmental factors that gradually degrade the sensory, nutritional, and safety qualities of food. Preservation techniques aim to slow or halt these processes to extend shelf life and maintain product quality.A key concept in food microbiology is the microbial growth curve, which includes four phases: lag, exponential (log), stationary, and death. During the lag phase, bacteria adjust to their environment without significant...
01:26Methods of Controlling Food Spoilage
Food spoilage is caused by microbial growth or by chemical and physical changes, all of which affect the taste, texture, and safety of food.Temperature-Based PreservationRefrigeration at 0–4 °C slows microbial growth and enzyme activity, making it ideal for short-term storage. However, certain spoilage organisms—such as psychrotrophs like Listeria monocytogenes—can still proliferate at these temperatures. Freezing below -18 °C further slows biological processes by forming ice crystals, which...