Research on Maturity Detection of Crisp Plum Based on Translucent Fast Vision Technology
Crisp plum is a characteristic fruit in southwest China,and the industry has developed rapidly.After ripening,the metabolism of crispy plums is enhanced,the endolytic matter is degraded,and the resistance will be significantly reduced,and making them easy to be harmed by latent infection bacteria so that lead to rotten fruit.In view of the problems of fruit decaying,short quality guarantee period and difficult storage of crisp plum after picking,this study utilized the translucent fast vision detection technology to detect the color and translucency of crisp plum,thereby completing the rapid maturity detection of crisp plum.The first step is building the computer vision translucency detection platform,then selecting the RGB color space as the method of crisp plum color extraction,and the minimum external rectangle method is selected to calculate the sum of pixels in the connected domain of the target area.The crisp plum maturity analysis is carried out.At the same time,the maturity detection program of crisp plum based on Matlab GUI is designed.The result shows that the threshold can be defined by the ratio of red pigment and green pigment,which can be used to distinguish the maturity grade of crisp plum.The maturity of crisp plum is positively correlated with the white area(i.e.translucency)of the translucent image which has been processed.The crisp plum samples were tested by the maturity detection program,and the average accuracy rate was as high as 93.57％.The research results provide a theoretical basis and technical support for the scientific and automatic detection of crisp plum,effectively reduce the loss rate of crisp plum during the ripening period,maximize the value of different grades of crisp plum.This study has both theoretical and practical value to promote the grade selection research and industrial development of crisp plum.
NETL
NSTL
万方数据
