An object detection model that detects high dense pollen grains in microscopic images and provides the count of grains in each image.
YOLOv8 state of the art, object detection model had used to train the model for detecting the grains in the microscopic images. The YOLO variant YOLOv8s has given tremendous results. The developed model can be used for detecting the grains in future. You can also get the count of grains in an image that you need for your research purposes.
The developed model evaluated with the metrics mAP (Mean Average Precision) and also tried to derive the accuracy by calculating the ratio of number of grains in predicted image to Ground truth image. The accuracies of training and testing are 98.27%, 94.02%, whereas the mAP of train, valid and test sets are 0.985, 0.946, 0.944 respectively.