Submitted to: International Conference on Precision Agriculture Abstracts & Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 7/17/2002
Publication Date: 12/1/2002
Citation: HUMMEL, J.W., DRUMMOND, S.T., SUDDUTH, K.A., KRUMPELMAN, M.J. SENSING SYSTEMS FOR SITE-SPECIFIC ASSESSMENT OF CORN PLANTS. PROCEEDINGS 6TH INTERNATIONAL CONFERENCE ON PRECISION AGRICULTURE. 2002. CD-ROM (UNPAGINATED). AMERICAN SOCIETY OF AGRONOMY. MADISON, WI. Interpretive Summary: Plant population has a significant effect on corn yield potential and with the exception of climatic conditions, can be the predominant factor limiting crop yield. In precision agriculture, producers need a map of plant population for a field, so that relationships between population and other variables such as soil type, can be used in optimizing application of inputs. The cost and time required for the intensive sampling needed, using hand sampling techniques, makes implementation impractical. On-the-go population sensors might be used to collect the desired data. Commercially-available photoelectric sensors were incorporated into a 4-row population sensor system and used to collect early-season corn plant diameter and spacing data. At harvest, the same sensors, along with prototype mechanical sensors, were mounted on a combine for field testing. Enhanced software filtering techniques were developed to improve plant diameter, spacing, and population estimation. The technology could be used by crop consultants and corn producers to optimize target fertilizer application rates across a field, resulting in a potential reduction in the amount of fertilizer that moves out of the field in surface and subsurface flow.
Technical Abstract: Early-season site-specific estimation of corn plant spacing and diameter could be useful in planting equipment evaluation, comparison of hybrids, and establishing input levels of nutrients during the current growing season, and harvest populations appear useful in interpreting yield maps. A four-row photoelectric corn population sensor system was evaluated for estimating early season population, plant spacing, and plant diameter. This system, along with four mechanical counters, was mounted on a combine for data collection at harvest. A Visual Basic-based post-processing program was developed to convert raw sensor data into useful information about the corn crop, and to facilitate comparison of sensor data with manually collected data. With appropriate software filtering of the data collected by the photoelectric sensor, accurate estimates of plant population, plant diameter, and plant spacing were obtained at lay-by and at harvest. At harvest, population estimates by the two sensors used were comparable to manually counted populations.