The invention discloses an aggregate particle identification method. The method comprises the following steps: shooting a high-definition image for aggregate at an appropriate position (such as a coldmaterial pile, a discharge hole and a conveyor belt); carrying out numerical processing on the image information by utilizing a matlab program; selecting an appropriate filtering technology for preprocessing the image according to the image acquisition effect in multiple filtering technologies; pre-segmenting and binarizing the image; performing morphological processing on the image by using opening and closing operations; performing preliminary segmentation on the image by using a watershed algorithm based on distance change; and carrying out fine calibration and secondary segmentation on the particles by using morphological characteristics such as pits. According to the method, the sampling positions of the aggregate particles are not limited, and only the image quality needs to meet the requirements. The result shows that the method can effectively segment aggregate particles to obtain particle information, and provides technical support for aggregate detection and grading controlin subsequent construction of the asphalt mixture.