Detection of skin defects on loquat using the hyperspectral imaging combining both band radio and improved three-phase level set segmentation method

Abstract Background and objectives Skin defects are one of the primary problems that occur in post-harvest grading processing loquats. The loquats with skin will lead to loquat being easily destroyed during transportation storage, which cause risk other infected, affecting selling price loquat. Materials Methods In this paper, a method combining band radio image improved three-phase level set segmentation algorithm (ITPLSSM) is proposed achieve high accuracy, rapid, non-destructive detection Principal component analysis (PCA) was used find characteristic wavelength PC images distinguish between four types defects. Determine best ratio based on wavelength. Results (Q782/944) PC2 segmented image. Based Pseudo-color enhancement, morphological processing, local clustering criteria, has better contrast defective area normal Finally, ITPLSSM segment (Q782/944), accuracy 95.28 %. Conclusions effective distinguishing Meanwhile, it also effectively segments intensity inhomogeneities.