ES Journal of Agriculture and Current Research

Abstract

Monitoring a crop’s biophysical parameters such as the Leaf Area Index (LAI), crop coverage, growth stages, health, crop height etc. at a fine-scale is crucial for agriculture field management in regard to irrigation and fertilizer applications [1]. There has been a great demand for non-destructive and cost-efficient crop monitoring with high accuracy in precision agriculture. Field-based surveying and sampling methods have been used for crop monitoring. However, these traditional techniques are often time-consuming, laborious and not feasible in large areas [2]. The use of geospatial techniques such as remote sensing technology has been applied to estimate various biophysical parameters. However, inadequate spatial resolution, along with the low temporal resolution of satellite data weakens the application of geospatial data [3]. Recently, high-resolution satellite data with daily coverage from Planet (https://www.planet. com/) is a game-changer in the agriculture field with the constellation of more than 150 active miniature satellites in the system. The emergence of lightweight and cost-efficient Unmanned Aerial Vehicles (UAVs) system has expanded the field of precision agriculture. UAV facilitates the availability of high spatial and temporal resolution earth observation to reveal high spatial details of crops. The authors conducted satellite and UAV observation of glass land, corn, and wheat fields. An index that shows the growth conditions at a glance is proposed. The comparison between satellite and UAV is examined. The resolution of Planet images is 3meters and that of UAV data is 0.1 meters. Integrating Planet derived Normalized Vegetation Index (NDVI) data with that of UAV derived data can utilize and explore the potential of crop monitoring and is expected to support agricultural practices to monitor crop growth with fewer visits to the field and thus leads to cost reductions. This study can help the region which is facing a shortage of human resources in rural areas and will be helpful for sustainable agriculture practices in helping to provide stable supplies of agriculture products. Deep learning is currently paving new avenues in the field of digital image processing and the possibility of classifying satellite images these techniques is also examined.