In order to improve the detection and recognition ability of potato internal diseases and insect pests, image visual feature recognition method was used to detect potato diseases and insect pests, and a method of potato internal diseases and insect pests feature recognition based on machine vision image was proposed. A two-dimensional visual image acquisition model of potato internal diseases and insect pests was constructed, and the visual images of potato internal diseases and insect pests were detected by block fusion, and the characteristics of diseases and insect pests were detected according to the distribution of potato green leafin texture. The visual fractal features of potato internal diseases and insect pests were extracted, the surface texture registration and block adaptive detection methods were used to calibrate the feature points of diseases and insect pests, and the wavelet transform method was used to decompose the visual images of potato internal diseases and insect pests. According to the difference of color gradient change, the characteristics of potato diseases and insect pests under machine vision were recognized. The simulation results show that the accuracy of the method is close to 90%, which improves the ability of the prevention and identification of the internal diseases and insect pests of potato.