Table of Contents
Advertisement
Outline
(3) Degree of match inspection
Purpose
Application
Example
(4) Point measurement
Purpose
Application
Example
Compare a good criterion image to a test image by inspecting matching levels
using the gray scale search function. (Determine whether the part is acceptable or
NG by checking similarities between the criterion image and the workpiece (test)
image.)
Detect positional deviation of labels, detect contamination of different parts, inspect
the mounting of electronic parts on PC boards, detect mis-prints, inspect for missing
electric parts such as terminals, and simple letter inspection.
Detecting label deviations on packages
Criterion image
for positioning
SERIAL NO.
8F053G26
MODEL
IV - C35M
Criterion image
for measuring object
Good label
[Measurement results]
- Degree of match compared with the reference image
- Detected coordinates (X/Y) of the measurement image.
- Light level in the measured image (average light level/absolute value of difference)
- Inspection procedure
1 Conduct a gray scale search of the criterion image position
2 Correct the position of the object being measured from the coordinates for
the criterion image obtained in item 1 above.
3 If the matching level of the test image is low, the IV-C35M can determine
that the label position is NG.
Evaluate whether workpieces are present or not at multiple specified points.
- Gray processes or converts to binary the image and evaluates whether specified
pixel area is white or black.
- Obtains average density of specified pixel area, and evaluates whether the
density is within the specified range.
Existence inspection of packing parts, inspect lighting conditions of mobile phones,
LEDs, and fluorescent tubes, and sorting products.
Inspection for LED lightings on mobile phones
CHR
CLR
Inspection procedures
Image capture
Binary conversion
Average light level
Search area
for positioning
Criterion image
for positioning
Criterion image
for measuring object
NG label
PWR
Number of points (max.)
: 128 points at average light levels
256 points in binary images
Point size: 2 m x 2 n pixels
Black/white evaluation of points
Light level evaluation of points
1-10
Search area
for positioning
(m, n = 1 to 16)
Advertisement
Table of Contents
