Fruit Grading Systerm Based On DSP Machine Vision

China is the largest producer of fruit in the world. However, the backwardness of the grading equipment and technology results in low level of post-natal treatment to agricultural products, which results the appearance of poor quality and lack of market competitiveness of domestic fruit. In recent years, grading system based on DSP machine vision has already started commercial applications abroad, but domestic research in this area is still at an early stage. In view of this, the fruit shape grading based on DSP machine vision is studied in this paper.Firstly, the image segmentation on the actual fruit grading line is studied. It is put forward different threshold methods, including the state method, the minimum error method, iterative method, the maximum variance method. Segmentation experiments are carried out and the results are compared and analyzed.Secondly, the dynamic grading of fruit size and color is studied. This paper detects fruit axis using the fruit symmetry to solve the problem that the fruit axis can't be detected due to the identification difficulty of stem and calyx caused by fruit movement. For color grading, the stability of the HSI is studied, and the color composition of the fruit surface is analyzed using H values.Then, the defects inspection method of fruit is researched ,the Gaussian multi-scale method had used to construct multivariate image data of the fruit surface, and then used the principal component analysis to extract the image information which can represent the main independent elements, calculated the residuals matrix Q statistic of the these main independent elements, at last the Q statistic had used to detect defects in the fruit surface images.At last, the DSP-based fruit grading machine vision system design and implementation details have focus on, analysis based on machine vision fruit grading system uses the benefits of DSP image processing system. introduced the details of the grading system's overall structure and main principles, key describes the DSP image processing system of fruit composition and method, and then the grading system in the PC section also provided a detailed description.This study is of great significance to improve the commercialization level of domestic fruit.