Climate change and environmental pollution will have a great impact on food security worldwide. The temperature increases cause, and will continue causing, more frequent drought events and, as such, an

increasing concentrations of soluble chemicals such as salt in the field. More than 30 % of the word’s irrigated areas are estimated to be perturbed by high salinity concentration affecting the productivity of crops. Camelina, also known as false flax, is a flowering plant that is mainly cultivated as an oilseed crop that has many potential economic benefits; it can be used in food products, converted into biofuel, industrial applications (e.g. hydraulic fluids, greases and polymers) and in animal feed. However, natural disasters due to climate events have led to significant crop losses.

In this work, we developed a computer vision approach to analyze the effects of high, low and no salt concentrations in Camelina Sativa using the McGill Plant Phenomics Platform (MP3). We present an adapted protocol to be applied with phenomics facilities in a greenhouse environment and the most effective way to high-throughput phenotyping. We evaluated different views and cameras from the phenomics equipment to choose the most informative ones, as such, reduce the measurement time. We built statistical multi-dimensional phenotypic profiles to assess more than 30 traits and we described “phenomics proxys” to important breeding traits to be used in selecting parent lines among other applications such as genetic screening.