Research on insect pest image detection and recognition based on bio-inspired methods

Insect pest recognition is necessary for crop protection in many areas of the world. In this paper we propose an automatic classifier based on the fusion between saliency methods and convolutional neural networks. Saliency methods are famous image processing algorithms that highlight the most relevant pixels of an image. In this paper, we use three different saliency methods as image preprocessing to train 4 different convolutional neural networks for every saliency method. We obtain several trained networks. We evaluate the performance of every preprocessing/network couple and we also evaluate the performance of their ensemble. The best stand-alone network is ShuffleNet with no preprocessing. However, our ensemble reaches the state of the art accuracy in a publicly available insect pest dataset, approaching the performance of human experts. Besides, we share our MATLAB code at: https://github.com/LorisNanni/.