The world population is growing steadily, and now it has reached around 7.9 billion people. To feed such a growing population, a substantial increase of global food production by 70 percent by 2050 is required. As a result, agriculture is facing a challenge, especially in combination with labor shortages and consumers asking for more sustainable food. To take on the food and labor challenges in the agriculture sector, significant interventions in arable farming and greenhouse horticulture are needed. One such innovation is the increased application of fully automated data analysis in greenhouses. One of the applications of this fully automated data analysis is automatically recording and monitoring seedlings in greenhouses to identify the growth, size, and color information over time. For this application, AI algorithms are used which are trained on annotated datasets. But, these annotations require time and money, especially if the model needs to be robust for several growing stages and light circumstances. Therefore, it is interesting to use AI for developing synthetic training data including annotations.
The task of this project is to investigate how AI could be used to develop synthetic training data. An example of an AI algorithm that is able to do that is a generative adversarial network (GAN), of which an example is shown in the image below (acquired from https://www.mdpi.com/2072-4292/13/1/23/htm). The aim of this project is to develop an AI network which output is similar to the example shown below, but then applied for a tray of seedlings.
30% Theory, 10% data collection, 40% implementation, 20% writing report
Le Viet Duc, email@example.com