Application of Machine Vision in Identifying Size And Surface Defect of Huanghua Pear

Yibin Ying,Hansong Jing,Yang Tao

2000-01-01

Abstract:Huanghua pear is one of the important fruits of China. The size and surface defects are important indexes in classification of Huanghua pear. To detect the surface quality of Huanghua pear, a machine vision system was set up on the basis of investigating the spectral reflectance of the Huanghua pear. The effects of different backgrounds on the acquired images were studied, and the method to seek the optimum background in the light of the grayscale histogram was found to be very effective. The clearer images could be acquired when the background was white. The images of Huanghua pears were acquired and the image preprocessing methods were studied. The Sober operator and Hilditch methods were used to fulfill the edge detection and edge thinning respectively. The image preprocessing method adopted could not only retain the original image information necessary, but also smooth the image and greatly decrease the redundant information. The maximum diameter, which was perpendicular to the line joined the center of pear and the intersection point of the stem and pear, was used to represent the size of pear, and it was found that the correlation coefficient of real size versus detected size was 0.96. In the light of color difference in the joint area of the defected and non-defected surface, the grayscale value of R (Red) frame and G (Green) frame were used to find the suspectable defected surface, and the whole defected surface was found by region growing method. To decrease the relative errors, the pixel transform method was adopted to recover the geometrical feature of sphere-fruit surface from umbriferous image while the area of defected surface was calculated. Moreover, a method to revise the estimated area was advanced. These results lay a solid foundation for further developing Huanghua pear quality detection system with machine vision.