sUAS sensor selection and utilization for remotely sensing crop health and production parameters

Abstract

Since the beginning of 21st century technological innovations are aiding to improve the agricultural industry. Precision Agriculture (PA) systems are one of those technologies which are creating the greatest impact on agricultural practices. PA is defined as the farming practices which optimizes the farm inputs with a goal to increase overall yield and minimizing harmful environmental impacts. To accomplish such targets, collection of high resolution spatial and temporal data on crop health and physiology becomes critical in helping to make decisions on optimizing farm inputs. Due to this, there has been increase in the use of small unmanned aerial systems (sUAS) in agricultural industry as a PA tool. sUAS are used for remote sensing applications such as crop phenotyping, crop water stress estimation, crop disease detection among others. With the development of robust sUAS sensors, collection of vegetative based data in field is evolving into standalone system. Sensors chosen for such applications include RGB sensors, thermal infrared sensors, modified color infrared sensors, multispectral sensors and Hyperspectral sensors. Usually a sUAS sensor is selected based on the spectral ability of the sensor and specific configuration are not considered. Improper selection of sensors leads to false data collection affecting the studies done using these sensors. In this thesis thermal infrared, modified color infrared and multispectral sensors were identified and compared based on the parameters specific to the sensor involved, to understand and improve future selection of sUAS sensors. Multispectral sensors are more commonly used as compared to other UAS sensors as multispectral sensors provide information both in visible and infrared spectrum. Multispectral sensors can be divided into two categories, namely narrowband and broadband. A study was done to compare and evaluate performance of the two types of multispectral sensors available in precision agriculture systems. The sensors were examined on different parameters to check their ability to provide remote sensing data with high-accuracy. Spectral response, ground resolution and statistical correlation with ground data were evaluated for both the sensors. Results shows the sensors performed differently in different parameters, but the spectral data provided by them was in close correlation with each other. A need for developing better ground data collection methods was observed. Thermal image quality is critical to accurately quantify water stress patterns in field crops. Image data quality from a thermal sensor is impacted by several factor. A second experiment was conducted with goal to compare accuracy of canopy temperature quantification and assess the quality of thermal orthomosaic when using thermal sensor of different focal length and image acquisition at varying flying altitudes of a sUAS. Three thermal infrared cameras were selected with focal lengths of 9mm, 13mm, and 19mm. All three cameras were flown at altitudes of 20m, 50m, and 80m. Results showed that 13 mm focal length and 50 m altitude produce finer resolution orthomosaic which provide a robust and accurate information on canopy temperature. Overall, the canopy temperatures were quantified accurately regardless of altitude and focal length by efficiently and accurately utilizing the ground reference system.