Write your own game script and recommend the picture recognition algorithm.

for ai robots that want to play two dozen games based on their hobbies, first solve the first picture recognition. Most of the pictures are motionless or do not change much. For example, most of the tanks in the red alert are always, except for the direction of walking and the direction of the barrel.

my initial idea is to put the pictures of tanks into the picture library, such as the Grizzly Tank, and put its eight or even 80 Grizzly tanks into one library in time. Look directly in the game to see if there are any pictures that are very similar to them. if you find it, it"s the Grizzly Tank.

there are other ideas can also be provided, there is a complete solution (just provide solutions, specific algorithms I can learn), the whole idea is feasible, simple, and can try most of the 2d games

The purpose of

1.K-Means clustering is to divide n points (which can be an observation of a sample or an instance) into k clusters, so that each point belongs to the cluster corresponding to the nearest mean (that is, the cluster center), which is used as the clustering standard.
2.K-Means algorithm solves the problem of image feature clustering analysis, and then uses OpenCV,SIFT, ORB, RANSAC and other algorithms to solve the problem of point set verification

.
Menu