TY - JOUR
AU - Visu, P.
AB - The agricultural industry faces challenges as pests and infections, which may damage a crop’s health and lower its production. Organizations like IPM (Integrated Pest Management) help farmers by providing pest control measures such as pesticide use and crop rotation. However, early discovery of pests is critical to preventing them from causing a significant problem. If pests are not identified at an early stage of producing, they might become a significant issue. Various CNN architectures and contemporary Deep Learning (DL) approaches are applied to solve pest identification challenges. Commercial approaches have grown more successful with the availability of sufficient pest data and modern algorithms for machine learning (ML) and architectures. In the work that follows, we provide a distinctive framework that is trained on diverse pest picture types, enabling simplicial-complex and closest neighbor feature extraction along with pattern analysis, assisting in the most accurate identification of different pests. Additional tests were performed on data sets for medical imagery classification that further validated the robustness and power of the proposed Adam optimization variants.
TI - Advanced Pest Identification Framework Using Deep Learning and Feature Extraction Techniques
JF - Journal of Electrical Engineering & Technology
DO - 10.1007/s42835-024-02111-3
DA - 2025-03-01
UR - https://www.deepdyve.com/lp/springer-journals/advanced-pest-identification-framework-using-deep-learning-and-feature-TZbcw52UBc
SP - 1803
EP - 1814
VL - 20
IS - 3
DP - DeepDyve
ER -