The work aims to count the corrugated cardboard accurately. A hand-held corrugated cardboard counting system based on machine vision was proposed and the hardware structure and application scenario of the system were introduced. The blob enhancement algorithm and the blob projection strategy based on slope correction were adopted and compared with the existing counting algorithms for analysis. Finally, counting experiments were carried out on corrugated cardboard laminates with different shapes, and the robustness of the proposed algorithm was further verified. The existing corrugated cardboards with different shapes were tested for many times, and the accuracy of the proposed counting system reached 99%. In the actual working scenario, a large number of tests show that the system can accurately count corrugated cardboard laminates.