Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Real-time Cricket Sorting By Sex

Juan Manuel Cantarero Angulo, Matthew Smith

TL;DR

The study addresses the need for scalable, sustainable cricket farming by automating sex-based sorting of Acheta domesticus. It integrates a lightweight embedded vision system (YOLOv8 nano) on a Raspberry Pi 5 with a servo-driven sorting arm to physically separate crickets by sex in real time. The prototype achieves a mAP@0.5 of 0.977 for sex detection and an overall sorting accuracy of 86.8%, with higher accuracy under low-stress conditions (>90%). This work demonstrates the feasibility of affordable, autonomous insect sorting and lays groundwork for breeding optimization and nutritional differentiation in industrial cricket production.

Abstract

The global demand for sustainable protein sources is driving increasing interest in edible insects, with Acheta domesticus (house cricket) identified as one of the most suitable species for industrial production. Current farming practices typically rear crickets in mixed-sex populations without automated sex sorting, despite potential benefits such as selective breeding, optimized reproduction ratios, and nutritional differentiation. This work presents a low-cost, real-time system for automated sex-based sorting of Acheta domesticus, combining computer vision and physical actuation. The device integrates a Raspberry Pi 5 with the official Raspberry AI Camera and a custom YOLOv8 nano object detection model, together with a servo-actuated sorting arm. The model reached a mean Average Precision at IoU 0.5 (mAP@0.5) of 0.977 during testing, and real-world experiments with groups of crickets achieved an overall sorting accuracy of 86.8%. These results demonstrate the feasibility of deploying lightweight deep learning models on resource-constrained devices for insect farming applications, offering a practical solution to improve efficiency and sustainability in cricket production.

Real-time Cricket Sorting By Sex

TL;DR

The study addresses the need for scalable, sustainable cricket farming by automating sex-based sorting of Acheta domesticus. It integrates a lightweight embedded vision system (YOLOv8 nano) on a Raspberry Pi 5 with a servo-driven sorting arm to physically separate crickets by sex in real time. The prototype achieves a mAP@0.5 of 0.977 for sex detection and an overall sorting accuracy of 86.8%, with higher accuracy under low-stress conditions (>90%). This work demonstrates the feasibility of affordable, autonomous insect sorting and lays groundwork for breeding optimization and nutritional differentiation in industrial cricket production.

Abstract

The global demand for sustainable protein sources is driving increasing interest in edible insects, with Acheta domesticus (house cricket) identified as one of the most suitable species for industrial production. Current farming practices typically rear crickets in mixed-sex populations without automated sex sorting, despite potential benefits such as selective breeding, optimized reproduction ratios, and nutritional differentiation. This work presents a low-cost, real-time system for automated sex-based sorting of Acheta domesticus, combining computer vision and physical actuation. The device integrates a Raspberry Pi 5 with the official Raspberry AI Camera and a custom YOLOv8 nano object detection model, together with a servo-actuated sorting arm. The model reached a mean Average Precision at IoU 0.5 (mAP@0.5) of 0.977 during testing, and real-world experiments with groups of crickets achieved an overall sorting accuracy of 86.8%. These results demonstrate the feasibility of deploying lightweight deep learning models on resource-constrained devices for insect farming applications, offering a practical solution to improve efficiency and sustainability in cricket production.

Paper Structure

This paper contains 8 sections, 3 equations, 14 figures, 8 tables.

Table of Contents

  1. Introduction and Related Work
  2. Methodology
  3. Hardware and Physical Setup
  4. Dataset and Annotation
  5. Object Detection Model: YOLOv8 Nano
  6. Control Logic and Actuation
  7. Experiments and Results
  8. Conclusions and Future Work

Figures (14)

  • Figure 1: Overview of the cricket sorting prototype: top — drawing of the device; center — top view of the sorting device; bottom — Raspberry Pi AI camera and sorting arm.
  • Figure 2: Servomotor and 3D-printed sorting arm located at the end of the bridge.
  • Figure 3: Comparison between male and female Acheta domesticus; note the ovipositor.
  • Figure 4: Training image containing three Acheta domesticus.
  • Figure 5: Spatial distribution of crickets in the dataset.
  • ...and 9 more figures