Chapter 6

Remote Sensing in Precision Agriculture

Structure from Motion Remote Sensing

Structure from motion (SfM) is a remote sensing technique that utilizes a series of 2-dimensional photographs of an object or feature to create a three-dimensional set of points corresponding to the surface of the feature (each with X, Y, Z coordinates) called a point cloud with associated RGB coloration. LiDAR point clouds do not contain such spectral information. Each SfM point is colored according to the feature it represents (e.g., a green tree canopy point would appear green). RGB information is stored as attribute data for each point, allowing for natural color point cloud visualization and analysis of ground features/land covers (e.g., selection/differentiation of vegetation points using the green information).

Applications of Structure from Motion Remote Sensing in Precision Agriculture

In agriculture, structure from motion (SfM) is primarily used to create detailed 3D models of crop fields and plant structures by analyzing overlapping images taken from various angles, enabling precise measurements of plant growth parameters, crop health assessments, and monitoring of soil erosion patterns, all with high spatial resolution and accuracy, often utilizing drones for data collection. The SfM approach with UAV imagery can be conducted at a relatively low cost using normal RGB cameras to suit the requirements of agricultural applications. SfM is used in crop monitoring to assess plant height, canopy cover, and Leaf Area Index, track crop development stages, identify stress indicators, and analyze plant architecture and disease spread patterns.

Limitations of Structure from Motion Remote Sensing in Precision Agriculture

While SfM offers cost-effective 3D modeling without specialized LiDAR sensors, it has several limitations. SfM relies on detecting common features in images to generate a 3D model. Errors accumulate when large-scale datasets are processed. SfM is a passive sensing method that relies on natural light, making it sensitive to lighting variations. SfM struggles to match features in low-texture environments (e.g., fields with uniform crops, bare soil, or water bodies).

Click on the following topics for more information on remote rensing in precision agriculture.