Because the internal quality of walnuts is not easy to detect and a certain proportion of shelled walnuts in supermarkets are of poor quality, the principle of judging the internal quality of walnut according to the density was discussed, and the method and equipment of walnut sorting were studied. The principle of sorting walnuts was as follows: firstly, the walnut image collected by the camera was processed to estimate the volume of walnut; the weight of walnut was obtained; the gas pipe was selected according to the density of walnut to output gas; as a result, different walnuts fell into different containers. Walnut can also be detected and sorted according to density equivalent parameters. Secondly, when the walnut fell to the platform, the angle between its split plane and the platform was always about 60°. With this feature, the front and side cameras can be set in the correct orientation to obtain the contours of walnut, calculate the density of walnut and improve the sorting accuracy. In conclusion, this technology can indirectly detect the internal quality of walnut and promote intelligent and advanced detection. The sorting scheme had no adverse effect on walnut, working environment and operators. With the decrease of camera price, the decline of image processing cost and the improvement of computer operation speed, the cost of walnut sorter would be greatly reduced. The food processing plant would be able to arrange multiple work stations and multiple sorters at the same time, so as to meet the requirements of production efficiency when large quantities of walnut were sorted. Compared with the existing methods, the proposed method can realize nondestructive detection and had high academic value. © 2021, Chinese Society of Agricultural Machinery. All right reserved.