A dataset of fortunella margarita images

This article created a Fortunella margarita image data set. The images of fortunella margarita are divided into seven categories: (a) mature, (b) immature, (C) growing, (d) containing mature and immature in one image, (e) mature and growing, (f) immature and growing, and (g) image data at all three stages set. Then the dataset is sorted into two types of files: (1) Growth stage classification images (Growth stage classification) and (2) Manual labeling and annotation files (Labels).
(1) Growth stage classification images: The number original images is 1031. Figure 1 shows examples of the original images for seven fortunella margarita categories. After data augmentation, we have a total of 6611 images. The size of image is 3024 * 4032 pixels, and the image format is JPEG. The dataset is divided into training, verification and test sets for machine learning or deep learning. Table 1 shows the number of training, verification and test images of the seven categories before data amplification.
(2) Manual labeling and annotation files (Labels): This dataset contains 6611 annotation images. The file format is XML. Each file is manually labeled with the growth stage and location as shown in Figure 2. This file can be used by researchers to train deep learning models such as YoLo, R-CNN etc.