Prediction of black soldier fly larval sex and morphological traits using computer vision and deep learning

Sarah Nawoya; Quentin Geissmann; Henrik Karstoft; Kim Bjerge; Roseline Akol; Andrew Katumba; Cosmas Mwikirize; Grum Gebreyesus

doi:10.1016/j.atech.2025.100953

Prediction of black soldier fly larval sex and morphological traits using computer vision and deep learning

Sarah Nawoya^*, Quentin Geissmann, Henrik Karstoft, Kim Bjerge, Roseline Akol, Andrew Katumba, Cosmas Mwikirize, Grum Gebreyesus

^*Corresponding author for this work

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaper › Journal article › Research › peer-review

1 Citation (Scopus)

Abstract

The growing interest in insect farming as a sustainable protein alternative has given rise to the commercial production of key species like the Black Soldier Fly (BSF), primarily for use in livestock, fish, and pet nutrition. Despite the heightened interest in BSF production, there is a need for increased efficiency, particularly in the context of large-scale measurement of various traits for selective breeding as well as management optimization. The unique insect production systems, coupled with the challenges posed by their small size, fragility, and metamorphic life cycle underscores the necessity for innovative approaches to streamline production. This study explores the potential of computer vision (CV) in predicting the larval sex and morphological traits of BSF, offering a non-invasive, rapid, and automated method for trait measurement. The study explores algorithms utilizing You-Only-Look-Once (YOLOv8) in detection and segmentation, ResNet for feature extraction and classification, and regression analysis mechanisms. We assess the ability of our models to predict larval weight from images through morphometric weight prediction and CNN-regression approaches. A notable contribution of this study is the pioneering effort to classify BSF larval sex using CV and deep learning (DL). In the analysis of larval weight prediction, a coefficient determination (R²) of up to 0.80 between measured and predicted weight was achieved using the morphometric weight prediction approach, along with an R² of 0.71 through the CNN-regression approach. Additionally, the sex prediction module demonstrated an F1 score of 0.75 and a prediction accuracy of 74 %. These results underscore the feasibility of leveraging CV techniques for predicting the sex and body traits of BSF larvae, representing a significant advancement toward the automation of selective breeding in the context of insect farming.

Original language	English
Article number	100953
Journal	Smart Agricultural Technology
Volume	11
ISSN	2772-3755
DOIs	https://doi.org/10.1016/j.atech.2025.100953
Publication status	Published - Aug 2025

Keywords

Automated Phenotyping
Black soldier fly
Computer vision
Deep learning
Insect farming
Selective breeding

Access to Document

10.1016/j.atech.2025.100953

Cite this

@article{96b0d3aa2600473fa2a012a733ee8787,

title = "Prediction of black soldier fly larval sex and morphological traits using computer vision and deep learning",

abstract = "The growing interest in insect farming as a sustainable protein alternative has given rise to the commercial production of key species like the Black Soldier Fly (BSF), primarily for use in livestock, fish, and pet nutrition. Despite the heightened interest in BSF production, there is a need for increased efficiency, particularly in the context of large-scale measurement of various traits for selective breeding as well as management optimization. The unique insect production systems, coupled with the challenges posed by their small size, fragility, and metamorphic life cycle underscores the necessity for innovative approaches to streamline production. This study explores the potential of computer vision (CV) in predicting the larval sex and morphological traits of BSF, offering a non-invasive, rapid, and automated method for trait measurement. The study explores algorithms utilizing You-Only-Look-Once (YOLOv8) in detection and segmentation, ResNet for feature extraction and classification, and regression analysis mechanisms. We assess the ability of our models to predict larval weight from images through morphometric weight prediction and CNN-regression approaches. A notable contribution of this study is the pioneering effort to classify BSF larval sex using CV and deep learning (DL). In the analysis of larval weight prediction, a coefficient determination (R2) of up to 0.80 between measured and predicted weight was achieved using the morphometric weight prediction approach, along with an R2 of 0.71 through the CNN-regression approach. Additionally, the sex prediction module demonstrated an F1 score of 0.75 and a prediction accuracy of 74 %. These results underscore the feasibility of leveraging CV techniques for predicting the sex and body traits of BSF larvae, representing a significant advancement toward the automation of selective breeding in the context of insect farming.",

keywords = "Automated Phenotyping, Black soldier fly, Computer vision, Deep learning, Insect farming, Selective breeding",

author = "Sarah Nawoya and Quentin Geissmann and Henrik Karstoft and Kim Bjerge and Roseline Akol and Andrew Katumba and Cosmas Mwikirize and Grum Gebreyesus",

note = "Publisher Copyright: {\textcopyright} 2025",

year = "2025",

month = aug,

doi = "10.1016/j.atech.2025.100953",

language = "English",

volume = "11",

journal = "Smart Agricultural Technology",

issn = "2772-3755",

publisher = "Elsevier B.V.",

}

Prediction of black soldier fly larval sex and morphological traits using computer vision and deep learning. / Nawoya, Sarah; Geissmann, Quentin ; Karstoft, Henrik et al.
In: Smart Agricultural Technology, Vol. 11, 100953, 08.2025.

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaper › Journal article › Research › peer-review

TY - JOUR

T1 - Prediction of black soldier fly larval sex and morphological traits using computer vision and deep learning

AU - Nawoya, Sarah

AU - Geissmann, Quentin

AU - Karstoft, Henrik

AU - Bjerge, Kim

AU - Akol, Roseline

AU - Katumba, Andrew

AU - Mwikirize, Cosmas

AU - Gebreyesus, Grum

PY - 2025/8

Y1 - 2025/8

N2 - The growing interest in insect farming as a sustainable protein alternative has given rise to the commercial production of key species like the Black Soldier Fly (BSF), primarily for use in livestock, fish, and pet nutrition. Despite the heightened interest in BSF production, there is a need for increased efficiency, particularly in the context of large-scale measurement of various traits for selective breeding as well as management optimization. The unique insect production systems, coupled with the challenges posed by their small size, fragility, and metamorphic life cycle underscores the necessity for innovative approaches to streamline production. This study explores the potential of computer vision (CV) in predicting the larval sex and morphological traits of BSF, offering a non-invasive, rapid, and automated method for trait measurement. The study explores algorithms utilizing You-Only-Look-Once (YOLOv8) in detection and segmentation, ResNet for feature extraction and classification, and regression analysis mechanisms. We assess the ability of our models to predict larval weight from images through morphometric weight prediction and CNN-regression approaches. A notable contribution of this study is the pioneering effort to classify BSF larval sex using CV and deep learning (DL). In the analysis of larval weight prediction, a coefficient determination (R2) of up to 0.80 between measured and predicted weight was achieved using the morphometric weight prediction approach, along with an R2 of 0.71 through the CNN-regression approach. Additionally, the sex prediction module demonstrated an F1 score of 0.75 and a prediction accuracy of 74 %. These results underscore the feasibility of leveraging CV techniques for predicting the sex and body traits of BSF larvae, representing a significant advancement toward the automation of selective breeding in the context of insect farming.

AB - The growing interest in insect farming as a sustainable protein alternative has given rise to the commercial production of key species like the Black Soldier Fly (BSF), primarily for use in livestock, fish, and pet nutrition. Despite the heightened interest in BSF production, there is a need for increased efficiency, particularly in the context of large-scale measurement of various traits for selective breeding as well as management optimization. The unique insect production systems, coupled with the challenges posed by their small size, fragility, and metamorphic life cycle underscores the necessity for innovative approaches to streamline production. This study explores the potential of computer vision (CV) in predicting the larval sex and morphological traits of BSF, offering a non-invasive, rapid, and automated method for trait measurement. The study explores algorithms utilizing You-Only-Look-Once (YOLOv8) in detection and segmentation, ResNet for feature extraction and classification, and regression analysis mechanisms. We assess the ability of our models to predict larval weight from images through morphometric weight prediction and CNN-regression approaches. A notable contribution of this study is the pioneering effort to classify BSF larval sex using CV and deep learning (DL). In the analysis of larval weight prediction, a coefficient determination (R2) of up to 0.80 between measured and predicted weight was achieved using the morphometric weight prediction approach, along with an R2 of 0.71 through the CNN-regression approach. Additionally, the sex prediction module demonstrated an F1 score of 0.75 and a prediction accuracy of 74 %. These results underscore the feasibility of leveraging CV techniques for predicting the sex and body traits of BSF larvae, representing a significant advancement toward the automation of selective breeding in the context of insect farming.

KW - Automated Phenotyping

KW - Black soldier fly

KW - Computer vision

KW - Deep learning

KW - Insect farming

KW - Selective breeding

UR - http://www.scopus.com/inward/record.url?scp=105002845100&partnerID=8YFLogxK

U2 - 10.1016/j.atech.2025.100953

DO - 10.1016/j.atech.2025.100953

M3 - Journal article

AN - SCOPUS:105002845100

SN - 2772-3755

VL - 11

JO - Smart Agricultural Technology

JF - Smart Agricultural Technology

M1 - 100953

ER -

Prediction of black soldier fly larval sex and morphological traits using computer vision and deep learning

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this