In this paper, a method for testing the loudspeaker of artificial layered packing based on machine vision technology was put forward. First of all, the collected image underwent preprocessing such as median filtering, then, the canny operator was used for edge detection to extract the edge contour of the image, and the pixels of the distance between the current layer edge and the selected fixed area boundary were solved, which was used as the basis of determining layer amount. By determining the amount of layers, the average gray levels of the 20 regions were solved respectively, so as to determine whether there was a lack of speakers. The experimental results showed that this method could accurately distinguish the layer and number of the loudspeakers, with good recognition effect, fast detection speed, and good real-time performance.