Dataset for varroa mite detection on sticky boards

<p>This Zenodo entry contains several supplementary files to the paper <i>"Analysis of Varroa mite colony infestation level using new open software based on deep learning techniques"</i> by Jose Divasón, Ana Romero, Francisco Javier Martínez de Pisón, Miguel A. Silvestre, Pilar Santolaria and Jesús L. Yániz.</p><p>The code is also publicly available in Github: <a href="https://github.com/jodivaso/varroa_detector">https://github.com/jodivaso/varroa_detector</a></p><p>The following files are included in this entry:</p><ul><li><strong>dataset.zip: </strong>this file contains the dataset that consists of 64 images (dimensions 8064 x 6048) of sticky boards with varroa mites. The compressed file also includes the labels (the annotations).</li><li><strong>images_deblurGAN.zip: </strong>this file contains the same 64 images, but after applying deblurGAN techniques.</li><li><strong>df_dataset.csv: </strong>CSV containing the information which images belong to the training set and which belong to the validation set.</li><li><strong>models.zip:</strong> it contains the different deep learning models that have been developed.</li><li><strong>example.zip: </strong>compressed file that includes a model (based on Faster R-CNN with resnet18 as backbone and fpn) and one image.</li></ul>