Deep Learning for Crop Science

Due to the COVID-19 pandemic, the conference will run remotely, using the Zoom platform.

IJCNN programme has been already published here. Our special session will be held in conjunction with the Deep Learning Methods for Wildlife Bioacoustics and Ecology the 20th of July from 8pm-10pm (British Summer Time - BST)

Plants provide the main source of food, not only to us, but also for animals. However, plant crops are endangered by several factors, such as drought, unpredictable climate changes, pathogens, and so forth. Combinations of these factors have an effect to the plants’ yield potential, which not only has an impact of the economy of farmers, but also to food security. Therefore, it is paramount to understand how plants grow in relation to the surrounding environmental conditions to maximise the yield.

Crop assessment is important to quantify and characterise plants, which is typically performed manually. Clearly, the manual assessment of crops is a time-consuming activity and also error-prone. Hence, we argue that image-based computer algorithms can provide an important tool to plant researchers, breeders, and agronomists to monitor crops. However, such algorithms need to be robust to obtain reliable visual information from plants. In particular, in the recent years, machine learning is making strides in the plant community, demonstrating that data-driven algorithms are robust and reliable to perform plant visual characterisation. Deep learning is a particualr branch of of machine learning that has emerged in the agriculture and plant sciences communities due to its versality to deal with large amounts of data and solve complex problems.

In this special session, we invite researches engaged in the plant community to submit research contributions focussing on image analysis of plants using deep learning. We will encourage research contributions including (but not limited to) the following topics:

• • Image-based plant analysis to improve state-of-the-art results and/or propose new challenges (classification, regression, image generation, tracking, etc.)

  • Crop assessment and monitoring with the use of RGB or multi-spectral imaging

  • New datasets for crop visual assessment, with suitable benchmark results

  • New deep learning architectures to address problems in plant phenotyping and crop assessment/monitoring

Template

Latex and Word templates can be downloaded from the following IEEE web page

For Authors using Latex, further instructions can be found here and bibliographic files can be downloaded from here.

Style Information

• • Only PDF papers will be accepted

  • Paper Size: US Letter

  • Paper Length: up to 8 pages are allowed, including tables, figures, and references. Additional pages (up to 2, for a total of 10 pages) are accepted with at an extra charge of 100USD/page

  • Paper Formatting: double column, single spaced, #10 point Times Roman font.

  • Margins: Left, Right, and Bottom: 0.75″ (19mm). The top margin must be 0.75″ (19 mm), except for the title page where it must be 1″ (25 mm).

  • No page numbers please. We will insert the page numbers for you.

Note: Violations of any of the above specifications may result in rejection of your paper.