7.3. Features for Shape and Size Characterization
In a number of image analysis applications, an important piece of information is the shape and size of an object of interest within the image. For example, in medical applications the shape and size of nodules are crucial in classifying them as malignant, or benign. Nodules with an irregular boundary have a high probability of being malignant, and those with a more regular boundary are usually benign. Also, it has been observed that in certain cases nodules with a perimeter of more than 3 cm are usually malignant [Cavo 92].
Another example in which the shape of the object is of major importance is the automatic character recognition in an optical character recognition (OCR) system [Mori 92 ...
Get Pattern Recognition, 4th Edition now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
