부산대학교 도서관

Our environment depends on honeybees. In the face of global honey bee decline, colony strength assessment can help apiary management and provide valuable research data. Counting honey, brood, pollen, larvae, and bee cells manually and classifying them based on visual judgement and estimation is time-consuming, error-prone, and requires a qualified inspector. Digital image processing and AI developed automated and semi-automatic solutions to make this arduous job easier. Prior to classification of cells, it is necessary to detect all the cells correctly. The objective of this study is to provide an efficient method to identify all the uncapped cells in honey comb images. The paper is organized with the discussion of problem's background and evaluation first. Further, we give a thorough review of the existing work and explore a possible application. After that, we discussed the methodology adopted in literature. Overall, our investigation is thorough and extensive.