I think that this paper is exactly what you are looking for.
PKLot – A robust dataset for parking lot classification
the algorithm classify each parking place at once as occupied or not, and the neural network was trained considering three different situations: sunny, cloudy and rainy days. Please, see the PKLot data set for more detailed information about the parking scenes.
Nice dataset! Again something to add to my collection!
We used YOLO in order to achieve this. We were able to obtain very satisfying results, they are detailed in this post: https://medium.com/picterra/the-xview...
@holger, the downvote is not mine, but my biggest guess, adding a comment, with a link to a website (that is cursed to be unreacheble someday) does not improve the answers already provided. They could have posted the relevant information here. The blogpost does not discuss empty parking lot detection (which to me seems the end goal here). But yes, Yolo can be used for this.
I would go for some sort of object model detection, making it rotation invariant or applying it in a series of different rotation to cope for the fact that the car will not always be straight horizontal or vertical.
Suggestions:
All of those are either in OpenCV main or contrib repository. Keep in mind that all of these techniques need training data, and in this case lots of it (since the variance in what is a car exactly).
Also, if the camera is fixed and always on the same position, more simpler techniques could be used, but I did not do that not to restrain the problem to much.
Asked: 2015-10-15 16:48:16 -0600
Seen: 2,600 times
Last updated: Jul 19 '18
