Deployment: Classification¶

The recommended processing pipeline uses a detection model with only one generic class ("insect"). The low input resolution enables a high inference speed, which is necessary to reliably track moving/flying insects. Images of the detected and tracked insects are cropped from synchronized HQ frames in real time.

By using the provided script for automated monitoring, cropped detections of individual insects are saved as .jpg files and relevant metadata is saved to .csv for each recording interval. The insect images can be classified in a subsequent step on your local PC, by using a classification model exported to ONNX format for faster CPU inference. The classification results are added to the merged metadata .csv files for post-processing in the last step.

Installation¶

If you followed the steps in Local Setup, you already have Python installed on your computer.

Optional: Set up an isolated Python environment

You can create a virtual environment before installing the required packages for YOLOv5 to avoid version and dependency conflicts of the packages, especially if you are working on different Python projects.

Install virtualenv:
```
py -m pip install virtualenv
```
Navigate to the YOLOv5-cls folder:
```
cd YOLOv5-cls
```
Create a new Python environment (folder in your current directory):
```
py -m virtualenv env_yolov5
```
Activate the environment by running:
```
.\env_yolov5\Scripts\activate
```
If you cannot run the activate script because of your Execution Policy settings, open a new PowerShell as administrator and run the following:
```
Set-ExecutionPolicy RemoteSigned
```
You can find more info here and here.
Now all packages will be installed only in the virtual environment, which you have to activate everytime you want to run a script that depends on these packages.
You can deactivate the virtual environment by running:
```
deactivate
```

Create the new folder C:\Users\<username>\YOLOv5-cls.
Download the custom yolov5 fork and extract it to the YOLOv5-cls folder.
Open a new Terminal (PowerShell) and navigate to the YOLOv5-cls folder:
```
cd YOLOv5-cls
```
Update pip:
```
py -m pip install --upgrade pip
```
Install the required packages by running:
```
py -m pip install -r yolov5-master/requirements.txt
```
If you are running the image classification on a computer with CUDA-enabled GPU, please edit the requirements.txt file and change onnxruntime to onnxruntime-gpu before installing the packages.

Run image classification¶

We will use the modified classify/predict.py script from the custom yolov5 repo together with the provided classification model to classify all insect images in the insect-detect/data folder (camera trap output) and add the prediction results to the merged metadata .csv files.

Download the insect-detect-ml repo and extract it to the YOLOv5-cls folder.
Copy your insect-detect/data folder, saved from the Raspberry Pi's SD card to the YOLOv5-cls folder. Make sure that only cropped detections are present! If you additionally saved full HQ frames, delete them before running the classification script.
Navigate to the YOLOv5-cls folder and start the classification script by running:
```
py yolov5-master/classify/predict.py --name camtrap1 --source "insect-detect/data/**/" --weights "insect-detect-ml-main/models/efficientnet-b0_imgsz128.onnx" --img 128 --sort-top1 --sort-prob --concat-csv
```
Change the --name of your prediction run accordingly, e.g. by including information about the camera trap location and collection date of the images. Instead of copying the insect-detect/data folder to the YOLOv5-cls folder, you can also insert its full path after --source.
Optional arguments
- --sort-top1 sort images to folders with predicted top1 class as folder name and do not write results on to image as text (which is the default configuration)
- --sort-prob sort images first by probability and then by top1 class (requires --sort-top1)
- --concat-csv concatenate all metadata .csv files and append classification results to new columns
Image Not Found

While running the classification script, in some cases you might run into the following error:
```
AssertionError: Image Not Found <image path>
```
This error can be caused by corrupt .jpg images, that are rarely generated during image capture. Run the process_images.py script to find corrupt images in the data folder and move them to a new folder with:
```
py insect-detect-ml-main/process_images.py -source "insect-detect/data"
```
After removing the corrupt .jpg images, the classification script should now run without throwing an error.

All results are saved to yolov5-master/runs/predict-cls/{name}. The *metadata_classified.csv still contains multiple rows for each tracked insect (= track_ID). During post-processing in the last step, the respective class with the highest weighted probability is calculated for each tracked insect. This will create the final .csv file, in which each row corresponds to an individual tracked insect.

If you used --sort-top1 as optional argument, the insect images are sorted to folders with the predicted top1 class as folder name. This allows for a quick identification of edge cases and these images can be used to retrain your classification model. By using --sort-prob additionally, the images are sorted by top1 probability first and then by top1 class. This can give you more information about cases where the model is unsure about its prediction. Adding images with a low probability to your training dataset can increase classification accuracy over time.