Full Surface Defect Detection of Spherical Fruit Based on Hyperspectral Online Sorting Technology

Abstract The whole surface hyperspectral image acquisition of spherical fruit is particularly important for the detection and quality classification of fruit surface defects. Because the light intensity at the edge of the spherical fruit is lower than that in the middle, the defects on the surface of the fruit cannot be effectively identified. In this paper, a hyperspectral online sorting device for the whole surface of spherical fruits is proposed. The correction of light intensity of spherical fruits is realized by using spectral characteristic peaks and the Non-uniformity Correction based on Quadratic Curve Fitting (QCF). First of all, the image data of navel orange was collected by online detection sorting equipment and the spectral image of the characteristic wave peak of 1655.72 nm was extracted. Then, the light intensity at the edge of the spherical fruit is enhanced by Non-uniformity Correction. Finally, the corrected image is segmented by the threshold to obtain surface defects. Ultimately, the online sorting test is carried out, and the detection accuracy is 100%. For defective navel oranges, the number of defect pixels after light intensity enhancement is effectively improved compared with that before light intensity enhancement. This indicates that this method not only overcomes the defect of the uneven light intensity distribution of spherical fruit surface images, but also effectively improves the sensitivity of defect detection. At the same time, the dimensionality reduction of hyperspectral data is also carried out, which is conducive to improving the efficiency of online detection.