1. Training Machine Learning algorithm
2. Environmental interference
3. Integrating technologies
4. Improper data availablity
5. Night time videos were difficult to capture the car count.
6. Computational power of our personal systems

We took some time to find and pick the most appropriate algorithm that would fit the needs and scope of this project. We were able to solve the the night time issue by contrasting the video and capture its outline. Since our systems were unable to process a lot of data, we reduced the resolution of processing to bare minimum. We were able to better train the system by feeding it a lot of data we had taken from Kaggle. By feeding a lot of data we able to remove environmental disturbances from the algorithm such as walking passerby, moving trees, etc.