Deep Learning in Agriculture whats happening Nathaniel Narra - - PowerPoint PPT Presentation

▶
deep learning in agriculture what s happening
SMART_READER_LITE
LIVE PREVIEW

Deep Learning in Agriculture whats happening Nathaniel Narra - - PowerPoint PPT Presentation

May 08, 2023 462 likes •668 views

Deep Learning in Agriculture whats happening Nathaniel Narra Prof. Tarmo Lipping Sigala group, Signal Processing Lab, TUT/Pori System of input and output: simplified Stimulus Response System of input and output: simplified

slide-1
SLIDE 1

Deep Learning in Agriculture – what’s happening

Nathaniel Narra

  • Prof. Tarmo Lipping

Sigala group, Signal Processing Lab, TUT/Pori

slide-2
SLIDE 2

System of input and output: simplified

Stimulus Response

slide-3
SLIDE 3

System of input and output: simplified

Stimulus Response Water Solar radiation Soil prop. Temperature …. Yield

slide-4
SLIDE 4

Temperature Solar radiation Precipitation Humidity … Soil properties Soil type Mineral content (N,P,K,..) Organic content Moisture … Weather Irrigation Fertilizers Compost Herbicides … Intervention

slide-5
SLIDE 5

Temperature Solar radiation Precipitation Humidity … Soil properties Soil type Mineral content (N,P,K,..) Organic content Moisture … Weather Irrigation Fertilizers Compost Herbicides … Intervention Remote Sensing Artificial Intelligence

slide-6
SLIDE 6

Temperature Solar radiation Precipitation Humidity … Soil properties Soil type Mineral content (N,P,K,..) Organic content Moisture … Weather Irrigation Fertilizers Compost Herbicides … Intervention Remote Sensing

Machine Learning Deep Learning CNN

slide-7
SLIDE 7

("machine learning" OR "deep learning" OR "artificial intelligence" OR "neural network") AND ("agriculture")

slide-8
SLIDE 8

Machine Learning Deep Learning CNN

(Convolutional Neural Networks)

Soil type Mineral content (N,P ,K,..) Organic content Moisture … Irrigation Fertilizers Compost Herbicides … Temperature Solar radiation Precipitation Humidity … Remote sensing image data

+

Sensor Data Agriculture information processing Agriculture production system optimal control Smart agriculture machinery equipment Agricultural economic system management Artificial Intelligence Methods Agronomy

https://granular.ag/farm-management-software/

slide-9
SLIDE 9

Machine Learning Deep Learning CNN

(Convolutional Neural Networks)

Soil type Mineral content (N,P ,K,..) Organic content Moisture … Irrigation Fertilizers Compost Herbicides … Temperature Solar radiation Precipitation Humidity … Remote sensing image data

+

Sensor Data Plant Animal Land Mechanization Artificial Intelligence Methods Subject areas

slide-10
SLIDE 10

Machine Learning Deep Learning CNN

(Convolutional Neural Networks)

Soil type Mineral content (N,P ,K,..) Organic content Moisture … Irrigation Fertilizers Compost Herbicides … Temperature Solar radiation Precipitation Humidity … Remote sensing image data

  • Hyperspectral
  • Multi-spectral
  • SAR
  • Infrared/Thermal
  • LIDAR
  • NIR
  • Optical
  • X-ray

+

Sensor Data Plant Animal Land Mech. Artificial Intelligence Methods Subject areas

  • Crop classification
  • Phenology recogn.
  • Disease detection
  • Weed/pest detection
  • Fruit counting
  • Yield prediction
slide-11
SLIDE 11
  • Crop classification
  • Phenology recogn.
  • Disease detection
  • Weed/pest detection
  • Fruit counting
  • Yield prediction

Kussul et al. 2017; DOI: 10.1109/JSTARS.2016.2560141 Rebetez et al. 2016; ISBN: 978-287587027-8

slide-12
SLIDE 12
  • Crop classification
  • Phenology recogn.
  • Disease detection
  • Weed/pest detection
  • Fruit counting
  • Yield prediction

Yalcin, Hulya. “Plant phenology recognition using deep learning: Deep-Pheno.” 2017 6th International Conference on Agro-Geoinformatics (2017): 1-5. Cotton Pepper Corn

slide-13
SLIDE 13
  • Crop classification
  • Phenology recogn.
  • Disease detection
  • Weed/pest detection
  • Fruit counting
  • Yield prediction

Mohanty et al. 2016; DOI: 10.3389/fpls.2016.01419 accuracy of 99.35%

slide-14
SLIDE 14
  • Crop classification
  • Phenology recogn.
  • Disease detection
  • Weed/pest detection
  • Fruit counting
  • Yield prediction

McCool et al. 2017; DOI: 10.1109/LRA.2017.2667039 Dyrmann et al. 2017; DOI: 10.1017/S2040470017000206

slide-15
SLIDE 15
  • Crop classification
  • Phenology recogn.
  • Disease detection
  • Weed/pest detection
  • Fruit counting
  • Yield prediction

Bargoti & Underwood 2016; arXiv:1610.03677v2 Chen et al. 2017; DOI: 10.1109/LRA.2017.2651944

slide-16
SLIDE 16

What next?

“…one key shortcoming: no major company has really delivered on the promise of facilitating better in-season decision-making.” (Barclay Rogers, agfundernews, Sep 2018)

The next big wave in agtech will be better in-season decision-making, including:

 Directing resource allocation based upon on actual field performance  Informing in-season fertilizer applications  Detecting pest and disease pressure  Evaluating product performance  Guiding irrigation decisions  Forecasting field-level yields  Providing better management zones

https://agfundernews.com/whats-next-for-agtech.html/

slide-17
SLIDE 17

Future?

Hyperspectral imaging : greater source of data for analysis

Drone tech

Crop models: AI methods

Databases and decision making?

https://agfundernews.com/growing-impact-hyperspectral-imagery-agrifood-tech.html/

VTT creates the world's first hyperspectral iPhone camera

https://phys.org/news/2016-11-vtt-world-hyperspectral-iphone-camera.html

slide-18
SLIDE 18

Machine Learning Deep Learning CNN

(Convolutional Neural Networks)

Soil type Mineral content (N,P ,K,..) Organic content Moisture Impedance? … Irrigation Fertilizers Compost Herbicides … Temperature Solar radiation Precipitation Humidity … Remote sensing image data

  • Hyperspectral
  • Multi-spectral
  • SAR
  • Infrared/Thermal
  • LIDAR
  • NIR
  • Optical
  • X-ray

+

Sensor Data Plant Animal Land Mech. Artificial Intelligence Methods Subject areas

  • Crop classification
  • Phenology recogn.
  • Disease detection
  • Weed/pest detection
  • Fruit counting
  • Yield prediction

MIKÄ-DATA context