Setting a desired threshold value as a standard and converting a gray image into values of 0 and 1 is called "binarization processing" (usually black is 0 and white is 1). By using this process, it is possible to focus on only the object of inspection and conduct various analyses.
The method of analyzing an image that has undergone binarization processing is called "blob analysis". A blob refers to a lump. Blob analysis is image processing's most basic method for analyzing the shape features of an object, such as the presence, number, area, position, length, and direction of lumps.

Blob applications

When blob analysis is performed on this image using white as the object

There are two lumps for analysis
One starts from the position X = 2, Y = 2 and has a width of 3, a height of 4, and an area of 12
The other starts from the position X = 9, Y = 3 and has a width of 5, a height of 5, and an area of 25

Presence or absence of lumps

It can be determined whether objects for inspection exist in the image.

Count

The number of objects for inspection in the image can be measured.

Size

The area of objects for inspection in the image can be measured (unit = pixels).

Measuring

The length, circumference, and so on of objects for inspection in the image can be measured (unit = pixels).

Distinguishing orientation

The orientation (angle) of objects for inspection in the image can be measured.

Position detection

The center of gravity and so on of objects for inspection in the image can be identified.

Inspection

Objects for inspection in the image can be treated as defects, and inspections can be performed on them by using the shape features above, etc.

Besides the applications above, blob analysis is also widely used as a basic method in many other applications.