About

Matthew Digman is an Associate Professor in the Department of Biological Systems Engineering at the University of Wisconsin–Madison. He holds a B.S. in Mechanical Engineering from the Milwaukee School of Engineering, obtained in 2003, and both his M.S. and Ph.D. in Biological Systems Engineering from the University of Wisconsin–Madison, completed in 2006 and 2009 respectively. His research interests include the impact of autonomy on agricultural machine forms, application of sensors to predict chemical and physical properties of agricultural materials, and the fractional utilization of herbaceous biomass. He is actively involved in advancing knowledge in these areas through his academic and research activities.

Research topics

• Computer Science
• Mathematics
• Artificial Intelligence
• Geography
• Machine Learning
• Agronomy
• Ecology
• Statistics
• Computer vision
• Materials science
• Engineering
• Animal science
• Chemistry
• Food science
• Environmental science
• Remote sensing
• Agricultural engineering
• Biology

Selected publications

• Day-to-day variation in forage and mixed diets in commercial dairy farms in New York

  Applied Animal Science · 2021 · 14 citations

  • Animal science
  • Agronomy
  • Mathematics

  We evaluated variation in sampling and analysis of forages and quantified day-to-day variation in silages and TMR in typical New York dairy farms. Alfalfa-grass haylage and corn silage samples were collected daily from 7 dairies, for a total of 24 wk for haylage, 22 wk for corn silage, and 16 wk for TMR samples. Multiple samples also were collected at 4 dairies to evaluate both sampling and subsampling variation. Based on SD, sampling for DM varied from actual DM by up to ±2 percentage units. Haylage was more variable than corn silage, likely due in part to variability of grass percentage within fields. The most practical parameter to measure for daily rebalancing of rations is DM, and DM had considerable day-to-day variability for haylage, with less variability for corn silage and TMR. Assuming a 7 percentage-unit threshold for a weekly range in DM is great enough to benefit from daily rebalancing, this threshold was exceeded 14% of weeks for corn silage, 25% of weeks for TMR, and 42% of weeks for haylage. A better understanding of day-to-day variability will help determine the accuracy required for on-farm silage moisture determinations. Nutrient composition of fed rations differs from formulated rations due to day-to-day variation in DM concentration and nutrient composition of forages. Although providing excess feed likely will mitigate the effects of day-to-day silage variability, it not only increases feed costs but also is less environmentally acceptable.

  PublisherDOI

• Handheld NIRS for forage evaluation

  Computers and Electronics in Agriculture · 2021 · 31 citations

  • Computer Science
  • Environmental science
  • Mathematics
  PublisherDOI

• Augmenting Crop Detection for Precision Agriculture with Deep Visual Transfer Learning—A Case Study of Bale Detection

  Remote Sensing · 2020 · 66 citations

  • Computer Science
  • Artificial Intelligence
  • Computer Science

  In recent years, precision agriculture has been researched to increase crop production with less inputs, as a promising means to meet the growing demand of agriculture products. Computer vision-based crop detection with unmanned aerial vehicle (UAV)-acquired images is a critical tool for precision agriculture. However, object detection using deep learning algorithms rely on a significant amount of manually prelabeled training datasets as ground truths. Field object detection, such as bales, is especially difficult because of (1) long-period image acquisitions under different illumination conditions and seasons; (2) limited existing prelabeled data; and (3) few pretrained models and research as references. This work increases the bale detection accuracy based on limited data collection and labeling, by building an innovative algorithms pipeline. First, an object detection model is trained using 243 images captured with good illimitation conditions in fall from the crop lands. In addition, domain adaptation (DA), a kind of transfer learning, is applied for synthesizing the training data under diverse environmental conditions with automatic labels. Finally, the object detection model is optimized with the synthesized datasets. The case study shows the proposed method improves the bale detecting performance, including the recall, mean average precision (mAP), and F measure (F1 score), from averages of 0.59, 0.7, and 0.7 (the object detection) to averages of 0.93, 0.94, and 0.89 (the object detection + DA), respectively. This approach could be easily scaled to many other crop field objects and will significantly contribute to precision agriculture.

  DOI

Frequent coauthors

• Paul J. Weimer

  University of Wisconsin–Madison

  27 shared

• K. J. Shinners

  University of Wisconsin–Madison

  21 shared

• R. E. Muck

  U.S. Dairy Forage Research Center

  19 shared

• Ronald D. Hatfield

  U.S. Dairy Forage Research Center

  17 shared

• Bruce S. Dien

  United States Department of Agriculture

  17 shared

• Michael D. Casler

  University of Wisconsin–Madison

  12 shared

• Kevin J. Shinners

  11 shared

• Hans‐Joachim G Jung

  Agricultural Research Service

  8 shared

Education

• Ph.D., Biological Systems Engineering

  University of Wisconsin-Madison

  2005

• M.S., Biological Systems Engineering

  University of Wisconsin-Madison

  2001

• B.S., Agricultural and Biological Engineering

  University of Wisconsin-Madison

  1998

Similar researchers at University of Wisconsin-Madison

• John Ralphh-178
• Brad Astorh-119
• Weibo Caih-113
• Matt Turnerh-113
• Paul Ahlquisth-110