The invention provides an optical lens surface defect detection method and
system. The method comprises the following steps: acquiring a corresponding multi-polarization-state image group through an industrial camera; generating a self-adaptive texture
mask matched with the normal texture of the lens through a
deep learning network based on the image features, and synchronously performing
gray level layering
processing on the multi-polarization-state image group according to the self-adaptive texture
mask to generate a corresponding target polarization-state image; calculating polarization and gray features of pixels of the target polarization state image, performing synchronous fusion to construct a corresponding
feature set, and performing clustering
processing on the
feature set by adopting a density peak clustering
algorithm so as to obtain an initial defect candidate area through threshold value screening; and extracting multi-dimensional feature parameters of the initial defect candidate region to construct a corresponding feature
verification library, and removing the
stain edge and the pseudo-
defect region of the initial defect candidate region through the feature
verification library to detect a corresponding target
defect region. The detection efficiency can be effectively improved.