Pakistani Researchers have achieved a milestone by developing an AI tool to access the sweetness of citrus fruits. Dr. Ayesha Zeb led the team of researchers from the National Centre of Robotics and Automation at the National University of Sciences and Technology. The AI tool successfully predicted fruit sweetness with over 80 percent accuracy without damaging the fruit.
Integration of Artificial Intelligence Algorithms to Access the Sweetness of Citrus Fruits
The researchers handpicked 92 citrus fruits from a farm in the Chakwal district. According to a local publication, they utilized a spectrometer to obtain spectra, patterns obtained from the bouncing light, from marked regions on the fruits’ skin. To examine the sample, the team employed a near-infrared spectroscopy technique to enable the analysis of non-visible light spectra. Of the 92 fruits, 64 were used for calibration and 28 for prediction via the spectrometer.
To improve the accuracy, the researchers integrated artificial intelligence algorithms to classify orange sweetness directly. The AI model was developed by obtaining reference values for Brix, TA, and fruit sweetness by peeling off samples from the marked areas for spectroscopy. Using the obtained spectrum, reference values, and sweetness labels, the team trained the AI algorithm on a total of 128 samples. Later, the researchers tested it with data from 48 new fruits, comparing the predicted values with actual measurements.
The Final Results
Researchers were astonished to see the results as the AI model accurately predicted the values of Brix, TA, and overall sweetness and outperformed traditional methods in sweetness prediction. The model achieved an overall accuracy rate of 81.03 percent for identifying sweet, mixed, and acidic tastes. With Pakistan being a major global producer of citrus fruits, this breakthrough has significant implications for the citrus industry in assessing fruit quality.