The number of vehicles that increase every year has a major impact on the occurrence of congestion and accidents and causes a significant increase in the volume of vehicles, especially on the highway. With this increase, many officers find it difficult to be able to anticipate or supervise vehicles directly. The research that we made, entitled Automatic Vehicle Counter System Based on Blob Detection for Highway Surveillance Using OpenCV, is a solution to this problem because by utilizing image transformation it makes it easier for the system to be able to detect vehicles and identify the number of vehicles entering the lane. The results obtained show an accuracy value of 97.11% based on testing with 10 video samples, with a total of 1329 vehicles detected out of a total of 1362, meaning that the total error is only 3.02%.