Skip to main content
ARS Home » Midwest Area » Columbia, Missouri » Cropping Systems and Water Quality Research » Research » Publications at this Location » Publication #61218
Title: CORN POPULATION SENSOR FOR PRECISION FARMING

Author
item BIRRELL, STUART - UNIV OF MO
item Sudduth, Kenneth - Ken

Submitted to: American Society of Agri Engineers Special Meetings and Conferences Papers
Publication Type: Abstract Only
Publication Acceptance Date: 6/22/1995
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: Corn population has a considerable effect on yield, and a decrease in population is normally a result of non-nutrient factors such as pests. Therefore, the ability to map corn population would provide additional information for the decision-making process and improve the use of yield maps for fertilizer recommendations. A combine-mounted corn population sensor has been developed. The sensor consists of a spring-loaded rod attached to a rotary potentiometer, mounted in front of the gathering chains on the row dividers. The corn stalks cause the rod to rotate back, resulting in an increase in potential across the potentiometer, which will drop off sharply when the stalk releases the rod. The buffered potential is fed through a low pass filter into an analog derivative circuit and digital filter circuit to convert the sharp drop in potential into a pulse recorded by a digital counter. The sensor was tested in the field by harvesting transects in which the distance between the stalks had been manually measured. The correlation coefficients between manual and raw sensor measurements were greater than 0.98 for all tests. However, the regression slopes ranged between 1.33-2.0, 1.11-1.23, and 1.03-1.16 for travel speeds of 3.2, 5.6, and 8 km/h, respectively. The higher predicted population was due to false counts caused by heavy weed infestations. However, if the raw measurements were filtered to remove multiple counts within 7.5 cm of each other, regression slopes were between 1.06-1.45, 1.01-1.05, and 0.93-1.00, respectively. The raw and filtered predictions of total population were within 20% and 5%, respectively of the actual total population for the 5.6 and 8 km/h tests. The sensor consistently identified gaps in the row regardless of speed and weed infestation.