A Study on Image-Based Detection of Varroa Destructor Using Superpixel and Color Features

Abstract

The parasitic damage of the varroa mite is a serious problem for honey bees, and the early detection of mites is indispensable for its control. W e study on a basic method for the image based detecti on of mites on the body surface of adult bees on the hive frame for beekeeping. Our method generates the s uper p ixel from the hue components of the image and detect s mites based on the shape and color features. We report the detection accuracy of mites in this paper.