System and methods for the detection of irregularities in objects based on an image of the object

A method for the detection of a defect in a solder element based on an X-ray image of the solder element, the method comprising the steps of

• acquiring an X-ray image of the solder element obtained by directing X-rays through the solder element for collection by an X-ray detector;
• binarizing the X-ray image at a plurality of binarization thresholds to obtain a plurality of binarized images, each having a respective binarization threshold;
• for each of the binarized images, extracting information from the binarized image;
• combining the information extracted from each of the binarized images;
• and
• detecting the defect based on the combined information,
• wherein the X-ray image and each of the binarized images comprise a set of pixels and each pixel in the X-ray image has a corresponding pixel in each of the binarized images, further comprising the step of assigning each set of corresponding pixels in the X-ray image and the binarized images an accumulator;
• wherein the step of extracting information from the binarized image comprises identifying pixels in the binarized image that satisfy pre-determined criteria, and
• wherein the step of combining the information extracted from each of the binarized images comprises, for each of the identified pixels in each of the binarized images, incrementing the accumulator corresponding to the identified pixel.

A method for the detection of an irregularity in an object based on an image of the object, the method comprising the steps of:

• acquiring an image of the object obtained by an image acquisition device;
• binarizing the image at a plurality of binarization thresholds to obtain a plurality of binarized images, each having a respective binarization threshold;
• for each of the binarized images, extracting information from the binarized image by estimating the regular object resulting from the binarization, at the respective binarization threshold, of an image of a version of the object in which the irregularity is absent;
• combining the information extracted from each of the binarized images;
• and
• detecting the irregularity based on the combined information.