Development and application of crop monitoring system for detecting chlorophyll content of tomato seedlings

A crop monitoring system was developed to nondestructively monitor the crop growth status in the field. With a two channel multispectral camera with one lens, controlling platform, wireless remote control module and control software, the system was able to synchronously acquire visible image (red(R), green(G), blue(B): 400-700 nm) and near-infrared (NIR) image (760-1 000 nm). The tomato seedlings multi-spectral images collection experiment in the greenhouse was conducted by using the developed system from the seeding stage to fruiting stage. More than 240 couples of tomato seedlings pictures were acquired with the Soil and Plant Analyzer Development (SPAD) value measured at the same time. The obtained images were available to process, and some vegetation indexes, such as normalized difference vegetation index (NDVI), ratio vegetation index (RVI) and normalized difference green index (NDGI), were calculated. Considering the SPAD value and the correlation coefficient between SPAD and other parameters in different fertilization treatments, the multiple linear regressions (MLR) model for tomato seedlings chlorophyll content predication was built based on the average gray value in red, green, blue and NIR, vegetable indexes, NDVI, RVI and NDGI in the 33.3% (N1), 66.6% (N2), and 100% (N3) nutrient levels during seeding stage and blossom and fruit stage. The R2 of the model is 0.88. The results revealed that the developed crop monitoring system provided a feasible tool to detect the growth status of tomato. More filed experiments and multi-spectral image analysis will be investigated to evaluate the crop growth status in the near future.