WEED DETECTION SYSTEM USING CONVOLUTIONAL NEURAL NETWORK

The ongoing presence of illegal drug issues in the Philippines, particularly concerning cannabis (marijuana),poses a serious threat to community safety and significantly burdens law enforcement. Law enforcementofficials find it challenging to conduct real-time identification of these substances because traditional methods,such as laboratory forensic testing, are often time-consuming and costly. This delay impacts policeinvestigations, as a major share of drug-related crime is linked to cannabis use or distribution. To directlyaddress this operational challenge, this study introduces a Weed Detection System Using Deep LearningAlgorithm. This mobile application is designed to provide quick, accurate, on-the-spot identification ofsuspected weed substances through image recognition, aiming to help police forces and security personnel. Thesystem's main objective is to minimize the analysis time for weed classification, thereby maximizing theworking time of law enforcers. It specifically aims to modernize the traditional process, provide a real-timeresponse, and analyze common types of weeds sold in the Philippines by their strains and details. The systemoperates by having the user capture or upload an image of the substance, which is then analyzed by aConvolutional Neural Network (CNN) trained on a large dataset to deliver rapid, reliable identification results.This innovation is focused strictly on expediting law enforcement investigations and shows significant potentialfor enhancing operational accuracy and streamlining workflows by leveraging data-driven, CNN-baseddetection