Toshiba Innovation

According to Toshiba, this synthetic photograph can be comparable to a APS size SLR. Multi-Frame synthesis this thing on the phone is Oppo Find7 proposed, at that time is 13 million pixels through multiple frames to synthesize 50 million pixels, later this technology is widely used. With all the different versions, now that Toshiba claims its own technology can rival SLR, is that a boast?

Weakness of traditional multi-frame synthesis technology

Multi-frame synthesis technology is very early on the camera, ha su 200 million pixel camera is actually used in the multi-frame synthesis. Used on the phone, is oppo Find7 to do the precedent.

The principle is the use of mobile phone CPU powerful processing capacity and now the ability of rapid exposure CMOS, quickly take a few, and then software synthesis, using a few photos to review the information, or to achieve the purpose of noise reduction, or to achieve the purpose of expanding pixels.

Because multiple frames have real pixel support, this effect is much better than interpolation, and the difference is clearly visible.

Because of its high practicality, many manufacturers have followed suit to learn, oppo in their own follow-up products Oppo N3 also use this technology.

Multi-frame synthesis results are obvious, but there are also weaknesses, because it takes a short time to take multiple frames, which requires the position of each frame roughly the same. And hand-held shooting, it is inevitable that a slight jitter, each frame position is different, the composite photo effect will be reduced.

Multi-frame noise reduction photos are less noisy, but often do not seem to have a high sharpness of the original photo, the multiple frame of the pixel photo shoot failure blurred the probability is not small.

In addition, this processing requires a large amount of memory for temporary storage of photos, the synthesis speed is also difficult, OPPO N3 to take care of the speed has to reduce the number of pixels of the super resolution.

Later, Huawei and Cool sent the technology of upstream vendors to make some changes to the multi frame synthesis. Using a fixed-position two-camera at the same time, two photos taken simultaneously to synthesize, so there is no jitter problem. But the cost of the dual-camera hardware has increased by one-fold, and only two frames have been synthesized, with limited quality improvement.

Toshiba Innovation

Toshiba's technical principles are virtually indistinguishable from ordinary multiple-frame synthesis, but Toshiba has improved on how to synthesize them.

Toshiba's "Infinite High quality" is also a multi-frame synthesis, but instead of taking 10 8-piece synthesis, Toshiba is a piece of synthesis, the first and second synthesis, while taking the third, one or two synthetic diagram and third synthesis, while taking the fourth picture ...

Toshiba's Synthesis method has been two in the synthesis, this kind does not need to store n picture to the memory, saves the promotion, the promotion speed, the Toshiba currently can synthesize 60, is equal to 60 times uses the CMOS area, this greatly improves the picture quality.

In order to solve the problem of jitter and ambiguity in multi frame synthesis, Toshiba uses the high-precision motion detection technology to compensate the fuselage jitter, which is similar to the camera's image stabilization, but it ensures the exact position of each frame and avoids the jitter resulting in different frame position.

This high-precision motion detection can not completely solve the problem, so Toshiba is also in the detection of ways, not each point to point to the simple synthesis, but to detect the contour of the object synthesis, so that the jitter caused by the synthesis of fuzzy problems completely solved.

In this way, not only can greatly increase the details of the photos, reduce impurities, increase the sensitivity, but also can improve the quality of the picture a lot.

According to Toshiba's data, "infinitely high quality" technology can stack up to 60 photo shoot. And in the case of using 13 million megapixel cell phone lens to burst 16 photos and then overlay, the quality of the photos can basically be comparable to the ordinary APs Senor's SLR.

Of course, Toshiba's technology also supports hyper-pixel, allowing 13 million of the camera to take 26 million-pixel photos, but this is not the point, the improvement of quality is the focus.

Because of the thickness of the smart phone can not plug the larger CMOS, multi-frame synthesis is an important way to improve the quality, Toshiba now has a new development in the synthesis technology, worthy of concern. Maybe six months or a year later, the real phone that can be comparable to SLR will come to our side.

How far is Silicon Valley? Click here for more information and join us in Silicon Valley!