Smart agriculture has benefited greatly from the widespread use of deep learning, which has proven critical to the industry. Reliability of data annotation and poor data quality, on the other hand, will severely limit the performance of intelligent applications because deep learning models are limited by these factors. We approaches, distance-entropy to distinguish the good and bad data from the perspective of information. DenseNet-121 was used as the backbone network and the IP06 dataset was used in trials. The findings highlight the frequency of duplicate data by demonstrating that almost 50% of the dataset has sufficient redundancy to produce test accuracy scores that are comparable. In addition, a thorough examination of representative samples resulted in the development of recommendations for enhancing dataset efficiency. These recommendations provide a useful road map for data-driven smart agriculture research, advancing knowledge and the use of data to advance agricultural innovation and sustainability.
Hover or tap a name to highlight its co-authors. Double-click to open the author's page.
National Institute of Technology Hamirpur
Different organisations
Same organisation
Affiliation unknown
Thicker line = more joint articles · bigger circle = more articles
Show as table
Co-author
Organisation
Joint articles
Affiliations come from OpenAlex; journal OAI-PMH metadata carries none. Authors without an OpenAlex ID are matched by name, which can merge or split people.