Ricoh's Industrial Cameras Have Wide Depth of Field, Recognize 6 Colors

Dec 8, 2012
Ikutaro Kojima, Tech-On!
A demonstration
A demonstration
[Click to enlarge image]
A comparison of depth of field (data courtesy of Ricoh)
A comparison of depth of field (data courtesy of Ricoh)
[Click to enlarge image]

Ricoh Co Ltd exhibited two industrial cameras that it has been developing at International Technical Exhibition on Image Technology and Equipment 2012, which took place from Dec 5 to 7, 2012, in Yokohama, Japan.

One is a camera whose depth of field is wider than other industrial cameras. The other is capable of recognizing a larger number of colors.

Camera with wide depth of field

First, Ricoh demonstrated a camera whose depth of field is three times wider than normal cameras. With a wide depth of field, it is possible, for example, to read QR codes attached to goods located near and far from the camera at the same time in a warehouse.

For another example, it becomes possible to take images of the joint areas of IC packages dispersed on a large printed circuit board (PCB) at the same time. Therefore, the camera reduces the time required for tests and checks.

Ricoh widened depth of field by using a special lens (optical system) and an image processing technology for collected images. This time, it employed a commonly-used sensor for the camera. In the demonstration, images were processed by software. But, if it is done with an FPGA, the image processing speed can be improved, the company said.

Cameras capable of recognizing 6 colors

Ricoh also demonstrated a camera that recognizes six colors. Normal cameras for production lines recognize three colors (red, green and blue). Because the camera detects small color differences, it realizes high-accuracy color checks. The company expects that the camera will be used for checking the surface colors of molded objects, coated objects and light-emitting displays.

The camera was realized by using a (1) "spectral filter," which collects several spectral data of an object, (2) "microlens array," which distributes the spectral data among sensor pixels, and (3) "image processing part," which creates an image based on each spectral data by using an image taken by a sensor.

Traditionally, to increase the number of colors to be recognized, it is necessary to increase the number of cameras. On the other hand, with the new method, the number of colors can be increased without increasing the number of cameras. As a result, it becomes possible to reduce the size of color checking mechanism.

However, there is a trade-off between the number of colors (chromaticity) and resolution. And Ricoh plans to put priority on one of the two depending on application.