Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Vibro-acoustic emission and heat stimulation effect on the detection of codling moth larvae in apples

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan

Citation:  2021 ASABE Annual International Virtual Meeting  2100070.(doi:10.13031/aim.202100070)
Authors:   Alfadhl Y Khaled, Chadwick A Parrish, Nader Ekramirad, Kevin D Donohue, Raul T Villanueva, Akinbode A Adedeji
Keywords:   Apples, Codling moth, Heat stimulation, Machine learning, Vibro-acoustic emission

Abstract. Codling moth (CM) larva is one of the most problematic insect pest species that plague the apple industry in the United States, especially with regards to international export. As the apple industry continues to grow, there is an ever-growing need for faster and more precise non-destructive techniques for detecting insect infestations. Vibro-acoustic emission sensing has proven to be an effective method for non-invasive detection. However, internal CM larvae infestation detection in apples can be difficult because their activity pattern is not stationary, resulting in a nontrivial chance of the larvae being dormant during the sample‘s evaluation period. One solution to this is to attempt to stimulate the larvae, encouraging it to be active during a known evaluation period. Our proposed method is heat stimulation, where the apple (and larvae, if present) are placed in an environment that slightly elevates its temperature (30°C) above the room temperature for a period before data are acquired. For heat stimulation to be meaningful, the temperature must be high enough to stimulate the larvae, but low enough to where it does not cause chemical property changes in the fruit or lessen the potential usable life of the fruit. The proposed work describes an experimental study where both control and CM larvae infested apples had vibro-acoustic data collected, with and without heat stimulation. The results show that heat stimulation does increase the accuracy of detection at the longer signal time (120 s, 60 s, and 10 s). Different classifier models were applied to recognize vibro-acoustic signals from CM infested apples, and the Ensemble model illustrated the highest performance, with overall test accuracy of 96.64%. This suggests that there is merit in heat stimulation, and industrial potential as the heat stimulation applied does not appear to damage the fruit.

(Download PDF)    (Export to EndNotes)