Abstract
Better matching of the timing and amount of fertilizer inputs to plant requirements will improve nutrient use efficiency and crop yields and could reduce negative environmental impacts. Deep learning can be a powerful digital tool for on-site, real-time, non-invasive diagnosis of crop nutrient deficiencies. A drone-based RGB image dataset was generated together with ground truthing data in winter wheat (2020) and in winter rye (2021) during tillering and booting in the long-term fertilizer experiment (LTFE) Dikopshof. In this LTFE, the crops were fertilized with the same amounts for decades. The selected treatments included full fertilization including manure (NPKCa+m+s), mineral fertilization (NPKCa), mineral fertilization but no nitrogen (N) application (_PKCa), no phosphorus (P) application (N_KCa), no potassium (K) application (NP_Ca), or no liming (Ca) (NPK_), as well as an unfertilized treatment. The image dataset consisting of more than 3600 UAV-based RGB images was used to train and evaluate in total of eight CNN-based and transformer-based models as baselines within each crop-year and across the two crop-year combinations, aiming to detect the specific fertilizer treatments, including the specific nutrient deficiencies. The field observations showed a strong biomass decline in the case of N omission and no fertilization, though the effects were lower in the case of P, K, and lime omission. The mean detection accuracy within one year was 75% (winter wheat) and 81% (winter rye) across models and treatments. Hereby, the detection accuracy for winter wheat was highest for the NPKCa+m+s (100%) and the unfertilized (96%) treatments as well as the _PKCa treatment (92%), whereas for treatments N_KCa and NPKCa the accuracy was lowest (about 50%). The results were similar for winter rye. In the cross-year and cross-cereal species transfer (training on winter wheat, application on winter rye, and vice versa), the mean accuracy was about 18%. The results highlight the potential of deep learning as a digital tool for decision-making in smart farming but also the difficulties of transferring models across years and crops.
Abstract from the paper’s OpenAlex record; open copy licensed cc-by. Read at the source ↗
The text the model read, with the source record’s inline markup (italics, sub- and superscripts) rendered and its character entities decoded; quote offsets refer to the raw text.
Claims 10
Claim · model-written summary
Nitrogen omission and no fertilization strongly reduced crop biomass.
- Taxon
- winter wheat and winter rye
Secale cereale
Triticum aestivumt0 - Treatment
- N omission and no fertilizatione0
- Outcome
- biomasse0
- Direction
- declinee0
Not reportedComparator
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“The field observations showed a strong biomass decline in the case of N omission and no fertilization”
t0“winter wheat and winter rye”
Claim · model-written summary
Phosphorus, potassium and lime omission had smaller biomass effects.
- Taxon
- winter wheat and winter ryeSecale cerealeTriticum aestivumt0
- Treatment
- P, K, and lime omissione0
- Outcome
- effectse0
- Direction
- lowere0
Not reportedComparator
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“though the effects were lower in the case of P, K, and lime omission”
t0“winter wheat and winter rye”
Claim · model-written summary
Mean within-year detection accuracy was 75% for winter wheat.
- Taxon
- winter wheatTriticum aestivume0
- Outcome
- detection accuracye0
- Value
- 75%e0
- Setting
- within one yeare0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“The mean detection accuracy within one year was 75% (winter wheat) and 81% (winter rye) across models and treatments”
Claim · model-written summary
Mean within-year detection accuracy was 81% for winter rye.
- Taxon
- winter ryeSecale cerealee0
- Outcome
- detection accuracye0
- Value
- 81%e0
- Setting
- within one yeare0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“The mean detection accuracy within one year was 75% (winter wheat) and 81% (winter rye) across models and treatments”
Claim · model-written summary
Winter wheat detection accuracy for NPKCa+m+s was 100%.
- Taxon
- winter wheatTriticum aestivume0
- Treatment
- NPKCa+m+se0
- Outcome
- detection accuracye0
- Direction
- higheste0
- Value
- 100%e0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“the detection accuracy for winter wheat was highest for the NPKCa+m+s (100%)”
Claim · model-written summary
Winter wheat detection accuracy for the unfertilized treatment was 96%.
- Taxon
- winter wheatTriticum aestivume1
- Treatment
- unfertilizede0
- Outcome
- detection accuracye1
- Direction
- higheste1
- Value
- 96%e0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“the unfertilized (96%) treatments”
e1“the detection accuracy for winter wheat was highest”
Claim · model-written summary
Winter wheat detection accuracy for _PKCa was 92%.
- Taxon
- winter wheatTriticum aestivume1
- Treatment
- _PKCae0
- Outcome
- detection accuracye1
- Direction
- higheste1
- Value
- 92%e0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“the _PKCa treatment (92%)”
e1“the detection accuracy for winter wheat was highest”
Claim · model-written summary
Winter wheat detection accuracy for N_KCa and NPKCa was about 50%.
- Taxon
- winter wheatTriticum aestivume1
- Treatment
- N_KCa and NPKCae0
- Outcome
- accuracye0
- Direction
- loweste0
- Value
- about 50%e0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“for treatments N_KCa and NPKCa the accuracy was lowest (about 50%)”
e1“winter wheat and winter rye”
Claim · model-written summary
Winter rye showed similar treatment detection results.
- Taxon
- winter ryeSecale cerealee0
- Outcome
- resultse0
- Comparator
- similare0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“The results were similar for winter rye”
e1“the accuracy was lowest (about 50%)”
Claim · model-written summary
Cross-year and cross-cereal species transfer achieved about 18% mean accuracy.
- Taxon
- winter wheatTriticum aestivume0
- Outcome
- mean accuracye0
- Value
- about 18%e0
- Qualifier
- cross-year and cross-cereal species transfere0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“In the cross-year and cross-cereal species transfer (training on winter wheat, application on winter rye, and vice versa), the mean accuracy was about 18%”
The same questions in other papers
fertilization · growth · Secale cereale (rye)Positive and negative results
4 papers in the same direction as this paper · 15 in the opposite direction
- positive
- positive
- positive
16 papers more in this group
See all 20 papers →fertilization · growth · Triticum aestivum (wheat)Positive and negative results
91 papers in the same direction as this paper · 663 in the opposite direction
- positive
- positive
- positive
751 papers more in this group
See all 755 papers →Claims are grouped by organism, treatment and what was measured or targeted. A paper counts as positive when any of its claims in the group is positive.