Abstract

Nowadays, using the stare-of-art algorithm, it is possible to create a system that helps people in-
crease the performance of work, avoid spending too much time on mundane tasks so they can
concentrate on work that requires human skills. Deep learning methods can complete tasks that
previously took human a lot of time to do.

Fresh food in supermarkets such as fruit, vegetables are sold quickly that why employ-
ees should re-fill the empty boxes with fresh fruit, vegetables as soon as possible. In the current
workflows, employees need to check manually and it is very time-consuming.

The goal of the thesis is to demonstrate that computer vision and deep learning can solve the
task of monitoring and masking (segmenting) the location of boxes of fruit, vegetables that need
to refill. This thesis present state-of-art instance segmentation Mask-RCNN algorithm to train and
build a model on a very challenging, the solution opens new experimental approaches, saving time
and resources for employees and help supermarkets increase their profits.