An Empirical and Theoretical Investigation of Random Reinforced Forests and Shallow Convolutional Neural Networks

For many years, the global population of honey bees has been decreasing due to inconclusive reasons resulting in the syndrome Colony Collapse Disorder (CCD). This syndrome has been plaguing bees and affecting commercial agriculture pollination since 1998. Many researchers have suggested that pesticides, in-hive chemicals, pathogens, etc., might be the causes of CCD. Researchers also believe that any changes in a beehive can disturb the bees, which may negatively affect their health. Honey bees are the most vital among all the animal pollinators contributing to approximately 30% of the world’s commercial pollination services. As they are of keystone importance to their respective ecosystems, monitoring their hives is crucial for understanding the effects of CCD and enabling beekeepers to maintain the health of their hives. As beekeepers cannot monitor their hives continuously, electronic beehive monitoring (EBM) can help them keep an eye on their hives. EBM extracts the videos, audios, temperature using cameras, microphones, sensors for observing the forager traffic (incoming and outgoing flow of the bees through the hive) to track food and nectar availability, following the sounds of the buzzing, and monitoring the abrupt temperature changes. EBM reduces the number of invasive inspections and transportation costs incurred for traveling to the beehive location. This research proposes a new technique using reinforcement learning, a method based on a reward/punishment strategy and aims at providing both accurate and energy efficient classification techniques to improve individual bee recognition in bee traffic videos