Method for detecting optical defects in transparencies
A method of detecting optical defects in a transparency may comprise the steps of providing a digital image of the transparency having a plurality of image pixels and detecting at least one candidate defect. The candidate defect may be detected by determining a grayscale intensity of each one of the image pixels and calculating an intensity gradient across adjacent pairs of the image pixels. Each image pixel may be assigned a gradient value comprising a maximum of the absolute value of the intensity gradients associated with the image pixel. A gradient image may be constructed comprising the gradient values assigned to corresponding ones of the image pixels. Image pixels may be identified as candidate pixels if such image pixels have a gradient value exceeding a gradient threshold. The candidate pixels may comprise the optical defect.
1. A method of detecting optical defects in a transparency, comprising the steps of:
providing a digital image of the transparency, the digital image including a plurality of image pixels each having a grayscale intensity;
detecting at least one candidate defect by performing the following:
determining a grayscale intensity of each one of the image pixels;
calculating an intensity gradient across adjacent pairs of the image pixels;
assigning to each image pixel a gradient value comprising a maximum of the absolute value of the intensity gradients associated with the image pixel;
constructing a gradient image comprising the gradient values assigned to corresponding ones of the image pixels;
identifying as candidate pixels the image pixels having a gradient value exceeding a gradient threshold, the candidate pixels comprising the candidate defect;
comparing a quantity of the candidate pixels to a pixel quantity threshold to identify the candidate defect as an optical defect; and
transforming, using a fixed coordinate transformation, a two-dimensional location of the optical defect on the digital image into a three-dimensional location on the transparency.
2. The method of claim 1 further comprising the step of:
detecting a perimeter of a viewing portion of the transparency;
wherein the step of detecting the candidate defect comprises detecting the candidate defect within the viewing portion.
3. The method of claim 2 wherein the step of detecting the perimeter comprises:
selecting a predefined perimeter of the viewing portion.
4. The method of claim 2 wherein the step of detecting the perimeter comprises:
determining a grayscale intensity of each one of the image pixels; and
comparing the change in grayscale intensity across a series of the pixels to a threshold intensity change rate.
5. The method of claim 1 further comprising the steps of:
providing a digital image of the transparency comprising a plurality of image pixels;
converting the digital image from color format to grayscale format; and
providing an image of the transparency comprising image pixels each having a corresponding grayscale intensity.
6. The method of claim 1 wherein the step of providing the digital image of the transparency comprises:
providing a digital image of the transparency comprising a plurality of image pixels in color format; and
converting the digital image from color format to grayscale format.
7. The method of claim 1 further comprising the step of characterizing at least one of size and shape of the optical defect by performing a corresponding one of the following steps:
characterizing the size of the optical defect relative to the transparency using a scaling transformation; and
characterizing a shape of the optical defect using a convolution integral.