Aphids Hyperspectral Images

Aphids cause severe damage on agricultural crops resulting in significant economic losses and the use of pesticides at high frequency with low efficiency. Monitoring aphid infestations by regular field surveys is time-consuming and does not always provide an accurate spatiotemporal distribution of pests. Therefore, an automated non-destructive method to detect and evaluate aphid infestation would be beneficial for targeted treatments. With this aim, we have created a machine-learning model to identify and quantify aphids, localizing their spatial distribution over the leaves, using One-Class Support-Vector Machine and Laplacian of Gaussians blob detection. To train this model, we built this large database of aphids' hyperspectral images, captured in a controlled laboratory environment. This database contains more than 160 images from three aphid lines distinctive in color, shape and developmental stage, and laying on leaves or neutral backgrounds.