I am thinking about writing a seminar paper about machine learning especially deep learning to detect if objects are present or not. But before I commit myself to this usecase, I would like to know if you guys think there are going to be any problems and
whether my approche would make sense or not.
The usecase would be components are placed on a component-holder.
It should detect if a component is placed on a component-holder.
If there is a posibility to place more than one component in a holder it should detect where it is placed.
Since I do not need to differentiate between the types of the component, I do not know how to approche this.
I could teach with different components and holders. And have them just classified as component or holder.
I do not know if this approche is any good or I should take another one. Maybe you guys have the experience and an idea how to approche this usecase.