Camera Consulting

The most important thing in an image processing inspection is to take an image of the object of inspection properly. Cameras, lenses, and lighting are indispensible items for the inspection, just as the inspection system is.
In particular, lighting has a large influence on the contrast of the image, and it is no exaggeration to say that the precision of image processing depends on the quality of the lighting.

At ViSCO, taking inspection precision and inspection conditions into consideration, we propose the best optical conditions for achieving the client's request.

Points for Consideration in Camera Selection

The following three points for consideration when selecting a camera can be given.

  1. The size of the object for inspection.
  2. What is to be inspected.
  3. The size of the flaw or foreign matter to be detected as a defect.

Further, it is necessary for the camera to fulfill the following conditions.

  • The object must be projected in a large manner
  • An image which has the object in focus must be projected
  • A bright, clear image must be projected

Two methods of taking images

Area sensor camera

Area sensor camera

Two-dimensional images can be obtained using this method, and it is the most commonly used image-taking method in image processing systems. The size of the image depends on the camera. When strobe lighting is used, images of moving objects can be taken too. Cameras that employ this method are often referred to as a “300 K pixel camera” or a “5 megapixel camera”.

Line sensor camera

Line sensor camera

This is a method of taking images that operates on the same principle as scanners and photocopiers, which is that two-dimensional images are obtained by continuously obtaining one-dimensional images (one row's worth). Compared to an area sensor camera this method is effective with large objects, and it is ideal in situations where images are continuously obtained. When acquiring images it is necessary for either the camera or the object to be moving in a set direction. Cameras that use this method are often referred to as a “1024 pixel line camera” or a “4096 pixel line camera”.

Color cameras, black and white cameras

Color images are effective for detecting different colored inspection objects, determining color irregularities, and so on, but because they use three times the amount of data of grey-scale images, a high-performance image processing system is required.

Differences in detection due to pixel count

The resolution of the camera is decided according to the size of the flaw or foreign matter to be detected as a defect. The larger the size of the image (lengthwise and widthwise), the greater the amount of information that can be obtained, so small defects can be detected.

4 pixels are required to detect defects

Ex.: When detecting a defect in an object with a width of 640 mm

300 K pixel camera: Common image size = 640 pixels widthwise x 480 pixels lengthwise
In the case of taking an image of the 640 mm width object in which the object fills up the image, 1 mm = 1 pixel.
That is, defects of 4 mm or more can be detected.
5 megapixel camera: 2,488 pixels widthwise x 2,050 pixels lengthwise
When a 5 megapixel camera is used, 0.257 mm = 1 pixel, so a defect of 1 mm can be detected.

5 megapixel camera

5 megapixel camera

300 K pixel camera

300 K pixel camera

Lens selection - Points to consider

It is important to consider the following seven things when selecting a lens.

  1. Distortion (linearity, pitch, length and width total pitch)
  2. Brightness
  3. Difference of peripheral brightness (Difference between brightness of center and surroundings)
  4. Peripheral resolving power
  5. Telecentricity
  6. Depth of field
  7. Chromatic aberration (in particular, compare using the WD difference of red and blue)

It is difficult to inspect these items one by one and manage the decision using numbers, but by taking “grid chart” images and comparing, you can see "the real capability" of the lenses, which is important practically and is hard to understand from just a specifications chart in a catalog or the like, and the merits and demerits of the lenses can be easily understood.

Lens selection method

Things you need

Camera, grid chart (if possible, with 1 mm pitch and high dimensional accuracy), depth of field chart, LED lighting (a total of three kinds - blue and red two-color backlighting and blue bar lighting).

Lens adjustment

Conduct the evaluation with the aperture fully open (only narrow it when evaluating telecentricity)
In the case where there is focus adjustment set it to the fastest setting (FAR, infinity)
Adjust the brightness of the image according to the camera exposure time

Object of image-taking, and image taken

Grid Chart
Focus the camera in blue lighting and take one image, and then leaving things as is, switch to red lighting and take another image (two in total).
When comparing the brightness of individual lenses, set the lighting brightness to a set brightness and take images using each of the lenses.
Depth of field chart
Using blue lighting, take one image with the aperture fully open and one image with it closed (two images in total).

1. Take images of a grid chart in blue/red lighting (use for evaluating chromatic aberration)

Image taken with focus adjusted
in blue light

Image taken with focus adjusted in blue light

Image taken with the lens in the position for focus in blue lighting and the lighting switched to red
*Image is out of focus

Image taken with the lens in the position for focus in blue lighting and the lighting switched to red*Image is out of focus

2. Depth of field chart (used for evaluating telecentricity and depth of field)

Image taken with aperture fully open and depth of field made small

Image taken with aperture fully open and depth of field made small

Image taken with aperture closed and depth of field made large
*Telecentricity evaluation

Image taken with aperture closed and depth of field made large*Telecentricity evaluation

3. Image taken of grid chart (used for evaluating difference of peripheral brightness and peripheral resolving power)

3. Image taken of grid chart (used for evaluating difference of peripheral brightness and peripheral resolving power)

4. Measuring the pitch and offset (height) of the grid chart (used for evaluating distortion)

4.  Measuring the pitch and offset (height) of the grid chart (used for evaluating distortion)