DETECTION AND ESTIMATION OF TREE CANOPY USING DEEP LEARNING AND SENSOR FUSION
Highly accurate spraying is one of the leading research zones in modern agricultural spraying applications, where the key factor knows the location and volume of the target. By utilizing cutting-edge information technologies, the unique idea of precision agriculture seeks to boost the production and efficacy of agriculture. Growers are better equipped to monitor every step of the production process and administer precise treatments chosen by machines with remarkable precision due to the most recent advancements in automation, artificial intelligence, and networking. Methods to reduce the number of human labour required in agriculture are still being developed by experts. Precision farming develops into a training system that gets smart every day as vital information resources get better. With regard to autonomous feature extraction, deep learning networks offer a considerable advantage over earlier algorithms because they don't require human participation. Deep learning (AI) and LiDAR technology combined with smart agricultural equipment can optimize spraying procedures. This study used the ficus, guava, and palm as case studies to create and assess canopy detection and estimating system. To scan trees for its height, distance, categorization, and canopy identification, the prototype included a LiDAR, machine vision, sensor fusion, and AI. Results for tree canopy estimate provided by a smart canopy detecting system showed a relatively low average error and predicted accuracy.