Reading the Visual Language of Massive Outdoor Biological Systems

RA Position Opening (2-14-2000)

By virtue of its ability to reveal the aspects of this planet previously hidden from view, from a new angle, the remote sensing (RS) technique has revolutionized modern geoscience, environment studies, and resource management technologies. With new developments in imaging sensors, the target can now be sensed with higher levels of spatial, spectral, and temporal resolution. There is the potential to extend remote sensing technology to new areas, for example, that of large scale biological object sensing. Emerging technologies for global positioning systems (GPS) and electronic-based information systems (EIS) represent a unique opportunity for utilization of remote sensing information in near real-time applications. Remote sensing is expected to become the sensor of a close-loop control system rather than merely a tool for monitoring. Traditional remote sensing targets are generally steady objects such as lakes, airports, crop types and field areas, etc. With large-scale high resolution RS we can sense living objects such as individual plants, or animals. With the monitoring of the growth stages and the responses to the environmental conditions by biological objects, site-specific management (SSM) can eventually be realized.

Figure 1. Low-altitude high-resolution aerial imaging system		Figure 2 High resolution digital camera

We have integrated and evaluated a near-real-time, multi-function biological target remote sensing prototype system (Figure1). Using an airborne high-resolution digital camera (Figure 2) and a real-time image processing and pattern recognition system, we have investigated some fundamental problems related to new high-resolution remote sensing techniques. While satellite remote sensing systems can only fly over a site at a certain time and obtain a spatial resolution of several meters per pixel, our system is near real-time and can obtain images with much higher spatial resolution. This system enable us to relate image patterns to ground truth more closely, almost simultaneously. More details about the targets have been found through image processing.

Example images:

Agricultural Engineering Farm (7-28-97)

Campus (around Morrow plots)

Details of Corn plants in Morrow Plots

One example of classified images

Publications (New 7-10-98)

"The general objective of this project is to establish a foundation for low-altitude/large-scale, biological object, remote-sensing technology. This involves finding the best settings to more effectively sense and archive in-field biological target variability information, and creating the massive chronological referenced database that supports real-time image data processing."

For more information please contact Dr. Lei Tian in Agricultural Engineering Dept., University of Illinois at Urbana-Champaign.

Go back to the Department Home Page