Unlike previously available systems, OMRON's Pattern Vision technology can be incorporated into devices as small as a cellular phone, while offering unprecedented performance in character/mark recognition. To realize these amazing capabilities, OMRON had to overcome many technical hurdles in capturing characters and marks quickly and compactly.

Look at the patterns in the two boxes below. Can you tell the difference between the right and left images?

Most people would not notice a difference. How about the characters shown in the boxes below?

Now, if you look closely you can see that the upper parts of the characters are different. In fact, the previous two patterns are also different in the top half. As you can see, people can unconsciously judge whether a pattern has meaning or not and then figure out the information it carries. If a machine can detect a mark in much the same way that people do, it should be possible to detect and process characters and marks in a more compact form.

This way of thinking led OMRON to devise its proprietary "Quest Theory," which greatly reduces the amount of information needed for recognizing each character. As a result, significantly less memory is needed when compared to conventional technology that overlaps and compares a character with a pre-registered image. This breakthrough enabled OMRON's Pattern Vision technology to be implemented in an amazingly compact unit like a cellular phone, while still offering exceptional recognition speed.
Presently, character recognition mainly focuses on English. But the demand for recognizing more complex characters such as Chinese has been increasing. To meet such requirements, OMRON is devoting significant R&D resources to further refine its Pattern Vision technology.

Return to Vision Sensing Top