Automated early yield prediction in vineyards from on-the-go image acquisition

  1. Aquino, A. 1
  2. Millan, B. 1
  3. Diago, M.-P. 1
  4. Tardaguila, J. 1
  1. 1 Instituto de Ciencias de la Vid y del Vino

    Instituto de Ciencias de la Vid y del Vino

    Logroño, España

    GRID grid.481584.4

Computers and Electronics in Agriculture

ISSN: 0168-1699

Year of publication: 2018

Volume: 144

Pages: 26-36

Type: Article

Export: RIS
DOI: 10.1016/j.compag.2017.11.026 SCOPUS: 2-s2.0-85035801258 WoS: 000425072400004 GOOGLE SCHOLAR


Cited by

  • Scopus Cited by: 30 (12-06-2021)

Journal Citation Reports

  • Year 2018
  • Journal Impact Factor: 3.171
  • Best Quartile: Q1
  • Area: AGRICULTURE, MULTIDISCIPLINARY Quartile: Q1 Rank in area: 5/57 (Ranking edition: SCIE)
  • Area: COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Quartile: Q2 Rank in area: 31/106 (Ranking edition: SCIE)

SCImago Journal Rank

  • Year 2018
  • SJR Journal Impact: 0.95
  • Best Quartile: Q1
  • Area: Agronomy and Crop Science Quartile: Q1 Rank in area: 46/363
  • Area: Animal Science and Zoology Quartile: Q1 Rank in area: 36/429
  • Area: Computer Science Applications Quartile: Q1 Rank in area: 114/1875
  • Area: Forestry Quartile: Q1 Rank in area: 17/159
  • Area: Horticulture Quartile: Q1 Rank in area: 4/86


  • Year 2018
  • CiteScore of the Journal : 5.5
  • Area: Horticulture Percentile: 96
  • Area: Forestry Percentile: 93
  • Area: Agronomy and Crop Science Percentile: 90
  • Area: Computer Science Applications Percentile: 82


Early grapevine yield assessment provides information to viticulturists to help taking management decisions to achieve the desired grape quality and yield amount. In previous works, image analysis has been explored to this effect, but with systems performing either manually, on a single variety or close to harvest-time, when there are few rectifiable agronomic aspects. This study presents a solution based on image analysis for the non-invasive and in-field yield prediction in vines of several varieties, at phenological stages previous to veraison, around 100 days from harvest. To this end, an all-terrain vehicle (ATV) was modified with equipment to autonomously capture images of 30 vine segments of five different varieties at night-time. The images were analysed with a new image analysis algorithm based on mathematical morphology and pixel classification, which yielded overall average Recall and Precision values of 0.8764 and 0.9582, respectively. Finally, a model was calibrated to produce yield predictions from the number of detected berries in images with a Root-Mean-Square-Error per vine of 0.16 kg. This accuracy makes the proposed methodology ideal for early yield prediction as a very helpful tool for the grape and wine industry. © 2017 Elsevier B.V.