Establishment and application of a method for measurement of rice grain shape based on image analysis

In this study，a new method with accurate，efficient，high-throughput measurement of rice grain shape was established based on image analysis. Digital images of rice grain samples were fetched batching using scanner，segmented preprocessing with image cropping tool and analyzed by SmartGrain software to get grain shaperelated metrics. Using this method，we could measure a large number of different samples of grain length，width，area and roundness and other traits simply and rapidly. The same samples were measured by manual scale measurement method and SmartGrain software，and their results were used to do significance test. The results showed that no significant difference of measurement result in the two kinds of measurement methods，with correlation coefficient value nearly equaling to 1，meaning very strong correlation. The new measurement method for rice grain shape was applied to a recombinant inbred line population in high-throughput determination of multiple grain shape traits，to improve efficiency of measurement.