The PlantVillage dataset, with over 54,000 images spanning 14 plant species and 26 disease types, has been widely used for leaf disease classification. However, it is limited in both scale and diversity. To address these limitations, we developed LeafNet, a large-scale dataset designed to support foundation models for leaf disease diagnosis. LeafNet comprises over 186,000 images from 22 crop species, covering 43 fungal diseases, 8 bacterial diseases, 2 mould (oomycete) diseases, 6 viral diseases, and 3 mite-induced diseases, categorized into 97 classes.

This dataset includes spectra of 250 corn samples with different vitality levels, with a data size of 250*256, categorized into five vitality grades. The imaging spectrometer employs a series spectrophotometer, model N17E, with a spectral range of 874-1734nm and a spectral resolution of 5nm. The CCD used is model ICL-B1410, featuring 1600×1200 pixels, and is equipped with an OLES22 lens with a focal length of 22mm.

Ripe and unripe pistachios lies in their appearance, taste, and texture, as well as their uses. Ripe Pistachio: The kernel inside is vibrant green with purple skin, larger, and fully developed. More aromatic and flavorful than unripe pistachios. Widely used in desserts (ice creams, baklava, pastries) and savory dishes. Unripe Pistachio: The nut kernel is smaller and pale green or yellowish. Less sweet and not as flavorful as ripe pistachios. Sometimes used in specialty cuisines, pickling, or as a garnish. This dataset contains 966 images for ripe and 966 images for unripe classes.

The IP102 dataset comprises 75,222 images of 102 pest species, out of which 12 classes are chosen for detection tasks. The custom pest dataset contains 10 categories of pests commonly found in crops like rice, maize, soybean, and canola. It includes images captured under varied real-world conditions, such as different lighting, occlusion, and complex backgrounds, making it highly representative of practical agricultural scenarios.