The purpose of this project is to be able to detect bees in an image regardless of orientation (vertical, diagonal, horizontal and anything in-between), track their movement and ultimately be able to count how many bees have entered of left the hive.
Visit our blog here for more detail on our progress throughout the project's lifespan
Mainly used for image processing, storing matricies, performing matrix operations such as finding the covarience of a matrix, performing matrix multiplication, transposing a matric and doing a principal component analysis on the training and test data.
We created a custom cropping tool in order to faciliate the creation of sample images from a given video file. It allows us to easily crop a portion of an image with predetermined width and height values.