Do you think it is possible to learn the app, how to autonomous evaluate good or bad parking of the bikes? The thing is you need to take a picture with your phone and app need to decide according to this pictures. The parking will be in the streets. Deciding will be based on the huge database with the pictures of the bikes which parked good/bad.
It is definitely possible!
You just need thousands of labeled pictures, a team of ML engineers, a few months, and some tens of thousands dollars to invest.
Jokes aside, it is not a simple task. Also, the labeling of the images will be a pain if they are not already labeled. You will need a team for that.
It would be a nice idea to ensure that your image dataset also has parking signs as a part of it, along with few other parameters that you can focus on. That shall help app in concluding whether a spot is available for public vehicle parking or not. I believe there is already an app available (Can I Park Here) on AppStore for this purpose. You may also like to take advantage of Google Maps API (collects info about parking spots, which are very well categorized), if publicly available. So the answer is YES, this is possible