Insect_Detect_classification Computer Vision Project
Overview
This version is deprecated! Please use the updated Insect_Detect_classification_v2 dataset with more images and classes.
The Insect_Detect_classification dataset contains images of various insects and some other arthropods, sitting on or flying above an artificial flower platform. All images were automatically recorded with the Insect Detect DIY camera trap, a hardware combination of the Luxonis OAK-1, Raspberry Pi Zero 2 W and PiJuice Zero pHAT for automated insect monitoring.
This classification dataset contains the cropped bounding boxes, exported from the Insect_Detect_detection dataset.
v2 insect_detect_classification includes a new class: episyr_balt, which contains images of Episyrphus balteatus. 290 new images of E. balteatus were added and all images of E. balteatus that were previously included in the class hovfly (v1) were transferred to episyr_balt.
Classes
The following classes were annotated in this dataset:
- wasp (mostly Vespula sp.)
- hbee (Apis mellifera)
- fly (mostly Brachycera)
- hovfly (various Syrphidae, e.g. Eupeodes corollae, Scaeva pyrastri)
- other (all Arthropods with insufficient occurences, e.g. various Hymenoptera, true bugs, beetles)
- shadow (shadows of the recorded insects)
- episyr_balt (Episyrphus balteatus) (new in v2)
View the Health Check for more info on class balance.
Deployment
You can use this dataset as starting point to train your own insect classification models. Check the model training instructions for more information.
To deploy the image classification model (ONNX format) on your PC for fast CPU inference, follow the provided step-by-step instructions. Open source Python scripts to deploy the trained model can be found in the insect-detect-ml
GitHub repo.
License
This dataset is licensed under the Creative Commons Attribution 4.0 International License (CC BY 4.0)
Citation
You can cite this dataset as:
Sittinger, M. (2023). Image dataset for training of an insect classification model [Data set]. Zenodo. https://doi.org/10.5281/zenodo.7725970
You can cite this project as:
Sittinger, M., Uhler, J., Pink, M. & Herz, A. (2023). Insect Detect: An open-source DIY camera trap
for automated insect monitoring [Preprint]. bioRxiv. https://doi.org/10.1101/2023.12.05.570242
Trained Model API
This project has a trained model available that you can try in your browser and use to get predictions via our Hosted Inference API and other deployment methods.
Connect Your Model With Program Logic
Find utilities and guides to help you start using the Insect_Detect_classification project in your project.