A Precise and Efficient Computational Code System for a Real Time 2D Analysis and Mapping of Crops through Unmanned Aerial Vehicle Surveys
The use of Unmanned Aerial Vehicles (UAV) in precision agriculture allows to constantly monitor crops by air at a relative low altitude. Thus, after a UAV mission is completed and given the high-resolution set of mounted cameras, the set of image data obtained could lead to find a potential problem. Therefore, there exists the need of having an efficient computational tool to process such information and provide the required answers in a relatively fast manner. Hence, a novel real time computational code system is presented: AGHAWK MAPS. This computational code system has been specifically created for data processing of crop imaginary. AGHAWK MAPS is mainly constituted of five components which can be accessed trough a graphical user interface (GUI): (1) Images are loaded from a given directory in the user’s computer. (2) Image coordinates are placed on map given their GPS locations and the user chooses those images to be analyzed. (3) Images are stitched according to their relative distance by following the UAV path and its angle of orientation. Also, a planar projection is always considered to speed up the calculations. Moreover, a unique technique for image blending has been implemented: the most important pixels approach. The main advantage of this technique is that it preserves each pixel information by considering those that belong to each image center. In other words, the most important pixels approach does not mix pixels among overlapped images, but it provides the essential information taken during the aerial survey. In addition, the stitched image is geo-rectified, and a final mosaic is obtained. (4) Vegetation indexes and estimators are obtained at the pixel level according to the camera filter type. Thus, data is processed and filtered through several algorithms which allow user interaction for specific constraints and custom values. (5) The user can refine the mosaic projection on a web map service and the final mosaic can be exported on different geographic data format for a posteriori use.
Learning Objective #1
The recent advances about UAV use in precision agriculture will be discussed. Furthermore, a computation tool specifically designed to process UAV imaginary: AGHAWK MAPS. The advantage of this tool for use in precision agriculture will change the way agricultural stitching is done.
Learning Objective #2
A specific UAV has been created with a custom camera system to collect image data for AGHAWK MAPS. Therefore, preliminary results of the use of AGHAWK MAPS will have a better understanding of the usefulness of UAV systems in precision agriculture.