Maize Growth Monitoring Based on Embedded Vision System

Monitoring the growth of maize rapidly and accurately is meaningful in research and management. However, traditional plant growth monitoring methods based on machine vision required ideal environment for image measurement, lack of exploration for the application of complex environment. This paper proposed embedded devices in Maize growth environment, extracting features of plant growth in the embedded vision system by the artificial, then monitoring by the user-defined algorithm. Plant growth monitoring system based on embedded device integrates a standard process flow contained five part such as filter, color space transformation, image segmentation, morphological operations and feature quantization. Each step in the process includes a variety of algorithms, users can combine algorithms freestyle to get analysis values of plant on different environment. Continuously monitor the maize plants at jointing stage, the results showed that the determination coefficient of system analysis value and manual measurement value reaches 0.907, which indicated that the system can be used for monitoring maize growth under complex environment.