Artificial Neural Network-Based Seedling Phenotypic Information Acquisition of Plant Factory

This work aims to construct an artificial neural network (ANN) ant colony algorithm (ACA)-based fine recognition system for plant factory seedling phenotypes. To address the problems of complexity and high delay in factories, first, multiple cameras at different positions are employed collect images seedlings 3D images. Then, mask region convolutional networks (MRCNN) is adopted analyze Finally, optimized ACA optimize process timing factory, thereby constructing a phenotype identification via ANN combined with ACA. Moreover, model performance analyzed. The results show that plants have four stages phenotypes, namely, germination stage, rosette heading stage. accuracy stage reaches 97.01%, required test time 5.64 s. Additionally, optimization sequence proposed maintained 90.26%, energy consumption stabilized 20.17 ms 17.71, respectively, when data volume 6000 Mb. However, problem image acquisition occlusion construction still needs further study. Therefore, constructed ANN-ACA-based phenotypes can more real-time lower way provide reference integrated progression unmanned intelligent systems complete sets equipment later