Sensing and imaging of plant disease through the lens of science mapping

Increasing efforts have been made to applying digital farming technologies such as sensing and imaging (S&I) and associated data analysis pipelines to plant pathology and disease management. Although several recent reviews have synthesized and discussed S&I advances in phytopathology, there has been no quantitative investigation of research trends, the evolution of research topics, and publication landscapes in this domain using a structured, bibliometric approach. Here, we applied science mapping—implemented in VOSviewer software—to systematically analyze and visualize scholarly output from journal and proceedings publications ( n = 582) in the domain of S&I of plant disease. The publication landscape was dominated by first authors from China, the USA, and Germany, whereby Chinese institutions were among the most prolific publishers (in numbers of articles) and the most collaborative institutions (in terms of link strength in the institutional coauthorship network). Only one disciplinary plant pathology journal was represented among the top-10 journals publishing research in the plant disease S&I domain, with the other top journals focused on sensors and physical processes, precision agriculture, and interdisciplinary plant science. Analysis of term co-occurrence in the literature corpus revealed a term map with four major clusters which were interpreted as research themes corresponding to wheat diseases and associated spectral and correlation indices, aspects of plant physiology, analytical approaches, and data acquisition sources. Temporal overlay visualization applied to the term map showed that articles in the latter two thematic clusters tended to have later publication dates, suggesting that domain knowledge in pathogen biology and plant physiology had to be established first before collaborating with engineers and computer scientists in more recently emerging areas such as machine learning and big data analytics. Our findings illuminate research habits, publication trends, and collaboration patterns and provide a baseline for future research on scientific networks in this interdisciplinary domain.